More Recent Comments

Wednesday, June 17, 2015

Creationists discover that human and chimp genomes are only about 70% similar!!!

Watch Jeffrey Tomkins of The Institute for Creation Science teach us about the very latest (not!) research in genomics. I haven't bothered counting the errors and misrepresentations in this video. I'm pretty sure it would take me a lot longer than 17 minutes to show where he is wrong. That's the problem.

Click on the figure below the video to show the human/chimp sequence comparison for chromosome 2 (top) and chromosome 7 (bottom). Note that the red dots fluctuate between 98-99% across the entire lengths except for a few gaps and duplications. Keep your eye on the figure as you listen to Jeffrey Tomkins tell us how the real difference is 66-85%.

Also, note how this level of similarity extends all across chromosome 2 where the comparison is to two separate chimp chromosomes. Examine the figure while Tomkins explains that the chromosome fusion idea is wrong.




193 comments :

Joe Felsenstein said...

A good way to point out the silliness of creationists' assertions. They love to argue that the difference is not 1.23% but is Much Bigger Than That. This they do by concentrating on the genome rearrangements and not on the base substitution differences in alignable parts of the genome.

One can make up various measures of difference between the genomes. Some will be considerably bigger than 1.23%, But the point that creationists* ignore is that all of these measures will show that the distance to chimps is much smaller than the distance to macacques.

Your figures do a great job of making that point.

*Footnote: I use the term "creationist" because that's who is making these arguments against close genealogical relationship between humans and chimps. Their argument has nothing to do with "Intelligent Design" arguments -- it is a straight creationist trope, and calling these creationists "ID theorists" is just muddying the waters.

c0nc0rdance said...

I can forgive the likes of Ken Ham and Kent Hovind... they're ignorant. Anyone with this level of training know that he's lying and does it anyway.

Is he talking about ungapped alignments of whole chromosomes? I wonder what similarity he would find if he tried the same alignment in a few hundred humans from various populations?

John Harshman said...

Come on, Larry. Don't make me actually listen to it. What's his measure of distance?

Unknown said...

Joe makes the most valid point. Regardless of the method used for comparison, chimps still come out as our closest relative. The creationists want to make the difference as great as possible. If they choose a method that makes the difference wider, they make the difference between chimps and other animals even wider.

Mikkel Rumraket Rasmussen said...

Dick pics.

Shazzbot said...

You are making a completely irrelevant point. The main point of the argument
ISN'T to argue that the chimp isn't the MOST similar amidst all other organisms to humans.
It's to argue that it's not similar ENOUGH.

Shazzbot said...

Sorry, I meant to say "It's to argue that it's not AS SIMILAR AS EVOLUTIONISTS CLAIM".
Though, yeah, it just may be used to argue for the former as well.

Joe Felsenstein said...

From the video, I think Tompkins is tossing back in the parts of the genome that were tossed out. These either are chunks of repeat sequences, or unalignable stuff. I don't know what he computes for these, but it must give similarity down near 0, so it raises the dissimilarity a lot.

For an even sillier creationist computation, see the one by UD author "niwrad" here. Niwrad takes 30-base random chunks from one species, finds the closest fit in the other, which is basically at the corresponding sequence, and then only counts it as similar if all 30 bases match perfectly. Otherwise "niwrad" counts a match of zero.

Here, at Panda's Thumb, I reanalyze that and show that the 65% similarity that "niwrad" gets is nearly what you expect when taking two genomes 1.23% apart and using this method of counting matches.

John Harshman said...

If you add the unaligned bits (and count each indel as a number of differences equal to its length) you only get to 95% similarity. That's what Roy Britten did. I don't think you can possibly get to 70% that way.

Niwrad's weird measure would get you into the ballpark, but was that what Tomkins did?

Robert Byers said...

They do a good research job in debunking how genes is a fuzzy way to compare us and primates.
yet i disagree with my fellow YEC here.
Even if its 70% Still thats close enough to make the point of how alike we are. Whats the gain to avoid 90% ?
The fact is that by looks and genes we are closer to primates then any other KIND of creature is close to another. lions to pandas etc.
YEC is wrong to desire great genetic differences with apes.
We are living in apes bodies.

The better arguement is that we uniquely live in another creatures body BECAUSE we uniquely are made in gods image. So we can not have our own body that represents this identity. This because biology is a general blueprint.
It only shows a material world and all creatures identity is only in their part of the blurprint. Thier bodies are thier identity.
Our souls are our identity and so we can only rent a body type in the blueprint.
the primate body type is the best for fun and profit.
Therefore this YEC desires 99% likeness with apes.
it won't be that for details but i want it.

Its not a logival path for YEC to desire lower % inlikeness. its high enough to have done the dirty deed.

Diogenes said...

No. Tompkins got the wrong answer not because (as he claims) evolutionists ignore sequences they can't align while Tompkins compared everything. No, Tompkins got the wrong answer because he encountered (or exploited) a bug in BLAST whereby it does not return a comparison if you feed it too many requests at one time. When BLAST did not return a result, Tompkins the moron assumed that the sequences being compared were 100% different. This blunder was IIRC pointed out by our friend Ace of Spades at a Reddit thread. Tompkins was informed of his blunders. He kept defending his bullshit, so it stopped being incompetence and became lying.

Diogenes said...

His measure of similarity is the same as ours. The moron just has a bug in his program. He was informed of the bug. He kept repeating the bullshit. At some point it stopped being incompetence and became lying.

TheOtherJim said...

So you are saying that the main "point" is that Tomkins et al. don't actually understand what the data is telling us?

We already suspected that, but thanks for clarifying the issue.

Shazzbot said...

"So you are saying that the main "point" is that Tomkins et al. don't actually understand what the data is telling us? "
No, CLUELESS, what I'm saying is the point of Thomkins' argument is to prove that the chimp and human are not nearly as similar as Evos claim..
EVEN IF the data told you that the chimp was the most similar organism to the human, that would still not debunk his argument.
"We already suspected that, but thanks for clarifying the issue. "
No, thank YOU for clarifying just how CLUELESS you are.

Shazzbot said...

As for whether Thomkins is correct or not (regarding the level of similarity), I'm really not sure. That wasn't what I was aiming for , anyways.

Jass said...

Isn't it a bigger problem for evolutionists if chimp's genome is 98%+ identical and yet the body plans, the brain, learning capacity ect. are not?

Confused said...

Your body and the bodies of some people with various syndromes are very different, yet your genomes are 99% similar. Is that a problem?

Faizal Ali said...

KevNick, your brain and learning capacity are severely deficient compared to those of most human beings, and yet your genome is more than 99% similar to ours.

Ed said...

"As for whether Thomkins is correct or not (regarding the level of similarity), I'm really not sure.
That wasn't what I was aiming for , anyways."

It's pretty clear what you're aiming for: to evangelize. You're not sure if Tomkins is right, you really don't have a clue what he's saying, but he seems to be an AUTHORITY and he seems to disagree with evolution. Thus for you it's simple, goddidit. That's all that really matters for you doesn't it?

John Harshman said...

Where did this happen?

Joe Felsenstein said...

It was discussed here last August. "Aceofspades" gave an account of the refutation of Tompkins by AceOfClades. Diogenes has more details and links in that discussion, including link to a detailed comment at Uncommon Descent that Aceofspades25 made. (I think I detect a common ancestor of all these Aces).

The 2014 Sandwalk subthread can be found here.

The links provides by Aceofspades and by Diogenes there are worth reading.

Anonymous said...

No, CLUELESS, what I'm saying is the point of Thomkins' argument is to prove that the chimp and human are not nearly as similar as Evos claim.

And Joe's comment was about how that doesn't change the fact that chimps are the most similar species to ours (and our closest relative). So my question would be: if all this guy is saying is that chimps and humans are not as similar as "evos" say, why would this be so important for creationists? If not to pretend that we share no ancestry, then what for?

EVEN IF the data told you that the chimp was the most similar organism to the human, that would still not debunk his argument.

Again, what's the purpose of this "argument." If not to pretend that chimps are not our relatives, if not to pretend that the similarity doesn't match evolutionary scenarios, then what?

No, thank YOU for clarifying just how CLUELESS you are.

I would suspect that ignoring the purpose of the "not as similar" claim would qualify as clueless.

Anonymous said...

Diogenes,

How was this problem figured out? Did this idiot present the work in writing, methods and all, somewhere?

Diogenes said...

I'll write a summary below, give me a few minutes to collect the links.

Diogenes said...

I'll write a summary below, give me a few minutes to collect the links.

Diogenes said...

P.S. Ace of Clades (= Aron Ra) was not involved at all, except that I speculated (incorrectly) that Ace of Spades might be Ace of Clades. In fact Ace of Spades did all the work.

nmanning said...

Even without the bug, Tomkins was being disingenuous by only searching for UNGAPPED returns in the first place. For those like photo that do not understand this, it means that Tomkins told BLASTn to return only sequences that were 100% identical in the first place. To wit - if his human search string was ATTGTGTACCGG and the chimp's corresponding sequence is ATCGTGTACCGG, he woul;d have gotten a 0% similarity on the return. He rigged the analysis to give him what he thinks the "truth" is, the bug only made it all that much worse.

nmanning said...

Indeed. I emailed Tomkins (via CMI) when his original paper came out, asking if he was going to do a follow up using pairs of humans as well as taxa that YECs think are descended from the same Kind, and and got no response. I have asked an Italian group that came to similar conclusions as Tomkins did, and at least they replied, claiming that they were not going to compare kinds, but did send me a graph in which they had thrown in 2 humans, and golly gosh - using that technique, the two humans differed by more than 5% - the level that YECs were claiming proved humans and chimps cannot be related back when Britten came out with his "let's include indels" paper in 95.

nmanning said...

Look at the crowd for his other talk on vimeo....

https://vimeo.com/109955495

Wow...

Shazzbot said...

"It's pretty clear what you're aiming for: to evangelize. You're not sure if Tomkins is right, you really don't have a clue what he's saying, but he seems to be an AUTHORITY and he seems to disagree with evolution. Thus for you it's simple, goddidit. That's all that really matters for you doesn't it? "
1.Actually, I DO HAVE A CLUE on what Thomkins MEANS. I just don't know whether he's correct or not.
I'm not defending that! I'm defending bogus objections to the point of the argument itself.
2."Goddidit". A typical atheist fallacy. Anything can be expressed in a simple and mocking manner. Evolution can be worded in a number of mocking ways as well!
3.No, it's not due to 'simplicity'(or personal desire or ignorance) that I embrace design .Design is easily detectable, sure. It's very intuitive to us. It's actually a coherent hypothesis. And I consider that it constitutes the only possible explanation for all the forms of life we see today. In particular, I agree with Message Theory(And I reject both chemical and biological evolution, consequently)
4.Arguments from AUTHORITY?Or are you implying that I act with some form of confirmation bias? I'm not an evolutionist to do that!
-----------------------------------------------------------------------------------------------
"And Joe's comment was about how that doesn't change the fact that chimps are the most similar species to ours (and our closest relative). So my question would be: if all this guy is saying is that chimps and humans are not as similar as "evos" say, why would this be so important for creationists? If not to pretend that we share no ancestry, then what for?"
1st. Clueless, similarity DOES NOT imply ancestry.
2nd. And I'm about to repeat for the second time that Thomkins was
attempting to debunk the Evo-proposed LEVEL of similarity.
Even if apelike creatures are way more similar to us than any other
creatures, that still does nothing to harm his argument!
"Again, what's the purpose of this "argument." If not to pretend that chimps are not our relatives, if not to pretend that the similarity doesn't match evolutionary scenarios, then what?"
1st. Again, similarity != relation. It can also be interpreted as common design. Common design expects both similarities AND differences.
2nd. "If not to pretend that the similarity doesn't match evolutionary scenarios, then what? ".
A)Anything can be ordered in degree of similarity.
I can order cars that way and infer that it "complies" to an evolutionary scenario.
B)The purpose is just to show that we are not as similar as Evolutionists claim. That's it!
If there is a set of 5 objects {A,B,C,D,E} and ,say, B is similar to A by 70%. C,D & E are similar by 50% or lower.
Say an Evolutionist were to say that A & B are 99% similar. Then a Creationist came to say that they are only 70% similar. Then saying that , amidst other set objects, B is still the most similar to A would NOT debunk the Creationist's argument.
Since the Creationist merely attacked the claim on the DEGREE of similarity.
C) How much similarity can "match" evolutionary scenarios, can you clarify? I think any amount can! You can say anything about the lack of "in-betweens"!
As for any other possible PURPOSE of Thomkins' argument- maybe introduce an even more destructive form of Haldane's dilemma. Or maybe just expose how the Evolutionists distort evidence to fit their views. Or maybe both.

Shazzbot said...

Correction: "C,D & E are similar to A by 50% or lower"

Diogenes said...

N. Manning, you're basically correct from what I understand-- this means that Tomkins' error was similar to Niwrad's dumb metric, but Tomkins used BLAST as an intermediate step, so the bug in BLAST magnified Tomkins' low score. I'll paste in some links and quotes below.

Anonymous said...

Judge Rez, a major problem is that Thompkins results are spurious. They come from misinterpretation of a known problem with the BLAST program used to compare gene sequences. Read Diogenes first response on the next string of comments.

Diogenes said...

N. Manning says: "I have asked an Italian group that came to similar conclusions as Tomkins did, and at least they replied, claiming that they were not going to compare kinds, but did send me a graph in which they had thrown in 2 humans, and golly gosh - using that technique, the two humans differed by more than 5%"

Good heavens, you need to write that up in a post. The creationists have proven all humans cannot be descended from Adam and Eve!

Shazzbot said...

Another thing I'd like to add. Though, I'd like to be a bit careful here.
Let's focus on the other claim - that the Chimp is the most similar
organism to us even if the similarity is only 70%.
Let's look at how other organisms are genetically similar to us. By what degree, that is. Now, again, I don't claim that these numbers are final, but:
"Cat: 90%
Cow: 80%
Mouse: 75%
Fruit Fly: 60%
Banana: 50%"
"
Source: http://genecuisine.blogspot.com/2011/03/human-dna-similarities-to-chimps-and.html
(Of course, the author is an evolutionist. I'm not gonna misrepresent him here)
Thus, in this case, it probably can turn that other argument (even though it's irrelevant in regards to Thomkins' work) upside down.
Additionally, if we were to group all organisms by level of DNA similarity, I doubt that we would ever see any reasonable correlations to label them "evolutionary"
(It seems pretty ridiculous to put a cat in the lineage leading to a human. Especially so close.Same goes for the other animals. And unquestionably for the banana)
Again, I'm not claiming that these numbers are final.

Shazzbot said...

@bwilson295 Yeah, I've read it. He (Diogenes) may be correct, I'm not saying that Thomkins is absolutely right.
I'm just attacking bad attempts at attacking his argument.

Diogenes said...

P.S. I take back my statement "His mesaure of similarity is the same as ours." In fact his measure seems to be more similar to Niwrad's. But Tomkins used BLAST as an intermediary, he sliced up the human genome into what he calls "sequence optimized slices", submits each slice to BLAST. If Blast does not return anything, Tomkins counts that as 0% similarity (note that a **RANDOM** genetic sequence would return 25% identity because any random nucleotide C,T,G or A would match the human bp 25% of the time.) There was a bug in the version of BLAST that Tomkins used, it didn't return any result if you overload it with queries, so the bug magnified the error in Tomkins' Niwrad-style metric.

Tom Mueller said...

Let’s refocus

Scanning through genomic databases of multiple species, permitted the identification of some of highly mutated areas that clearly contribute to human-specific traits

These so-called "Human accelerated regions" are obviously subject to intense selective pressure and to my reading diminish Neutral Theory’s importance in explaining Human Chimpanzee divergence while simultaneously validating the evo-devo paradigm

http://rstb.royalsocietypublishing.org/content/368/1632/20130025.short

Dave Carlson said...

N. Manning, I haven't looked at what Tomkins specifically did, but there is a difference between looking for ungapped sequences, and looking for 100% identical sequences. In the example you provided, there is no gap in the alignment, but there is a mismatch at the third position. So if you're looking for ungapped hits but don't require 100% sequence identity, this would still qualify.

Diogenes said...

Attention all, I'll try to sort out what metric Tomkins used and why he got his numbers so wrong. I will break this into two section: first, Tomkins' erroneous similarites for one gene, GULO (debunked by Ace of Spades here at UD and please examine all his damning figures embedded here, especially towards the end), and second, Tomkins' erroneous overall similarities for the whole genome due to exploiting a BLAST bug (debunked by Roohif.)

Continued below.

Faizal Ali said...

(It seems pretty ridiculous to put a cat in the lineage leading to a human. Especially so close.Same goes for the other animals. And unquestionably for the banana)

Your argument seems to rest quite heavily on the premise that the phenotypic differences between, say, a human, a cat and a banana are greater than can be accounted for by the variations in their genomes. How to you figure that? Rigorous answer, please, and not just an appeal to your intuition.

Does it not seem correct that a cat is more similar to a human than is a fruit fly, and even more so compared to a banana tree? That seems to be obviously correct. What problem do you have with those figures?

Faizal Ali said...

Additionally, if we were to group all organisms by level of DNA similarity, I doubt that we would ever see any reasonable correlations to label them "evolutionary"

Does the term "nested hierarchy" have any meaning to you?

Diogenes said...

Tomkins' Metric of Similarity

Tomkins says that his analysis works by "sequence slicing the large contiguous sequence into optimized slice sizes" and then submitting the "slice" to BLAST to see if it returns a hit. This sounds suspiciously like Niwrad's silly metric, which was debunked by Joe Felsenstein at PT. But Tomkins stick in a BLAST search as an intermediary step, which introduces new errors.

If BLAST returns nothing, Tomkins counts that as 0% similarity (note that a **RANDOM** sequence will return 25% identity because a randomly chosen nucleotide C,G, T or A will match the human bp 25% of the time.) BLAST apparently didn't return a lot of matches because it has a bug so when Tomkins overloads it with queries, it does not reply, and Tomkins marks that down as 0% similarity.

When BLAST does return a match, I'm not sure how Tomkins computes the similarity.

But an additional error, pointed out by Ace of Spades, is how Tomkins handles "N"'s, that is, regions that were not sequenced, but marked as "N" in the database. This is significant because in some places the chimp genome is incomplete where it's complete in the gorilla, so Tomkins uses that to say, "Ha ha, evolutionists say we're more similar to chimps than gorillas, but my score is higher for goriallas than for chimps!"

Ace presents irrefutable proof here and here that Tomkins counts each N as a 0% similarity. Note again that a **RANDOM** sequence compared to human will return 25% identity minimum), but Tomkins counts N's as 0%.

It's important that Tomkins angrily denies counting N's as a 0% match, saying his matches were "performed after stripping all N’s from the data set", which directly contradicts the proof to the contrary presented by Ace to Tomkins. This is not a difference of opinion, it's contrafactual.

Diogenes said...

Here is Tomkins describing his method, attacking Ace: "The BLASTN analyses done in this paper were performed after stripping all N’s from the data set and sequence slicing the large contiguous sequence into optimized slice sizes – all done on a local server using optimized algorithm parameters. My data not only takes into account gaps, but sequences present in human and absent in chimp, and vice versa. Doing an amateur armchair analysis on the BLAST web server with default parameters never designed for a one-on-one large scale genomic regional comparison as noted in the comment above by aceofspades25 is bogus. Of course, if the paper was actually read in it’s entirety in regards to the above comments this would have been obvious.

Also, as noted in several evolutionary papers, which I cited in my paper, the large scale comparison and major differences in structural variability surrounding the GULO regions between humans and great apes in the intronic areas has been noted before. Interesting that the misleading post by aceofspades25 did not make note of that. My paper was in fact accurate in all respects and true to previous findings published by evolutionist themselves. My work just hashed out and exposed what was already known, but never previously elaborated upon because it shows just another aspect of what a complete fraud the human evolution paradigm truly is."
[Jeffrey Tomkins attacking AceofSpades at UD]

Faizal Ali said...

Anyway, Judge Rez, your fellow creotard Robert Byers takes issue with your claim a few posts down, and thinks a >90% similarity between chimps and humans should pose no problem, because God designed us using "ape bodies". Why don't you take the issue up with him, settle it thru the pertinent scientific experimentation, and then let the rest of us know when you creotards have figured out the degree of genetic similarity that should exist between organisms that have been designed by God to look as if they are related to each other. There's a good boy.

Shazzbot said...

"Your argument seems to rest quite heavily on the premise that the phenotypic differences between, say, a human, a cat and a banana are greater than can be accounted for by the variations in their genomes. How to you figure that? Rigorous answer, please, and not just an appeal to your intuition."
Yeah, you're right. Just an appeal to intuition would be absolutely ridiculous.
I won't go into the details, so here's my short, but nevertheless effective answer - just look at everything from an ENGINEERING perspective :all these interdependent parts, the parts themselves also being very tough to alter without breaking, plus the fact that these organisms an existential need for a lot of functions at once.
I believe it's impossible to incrementally alter such an organism without making it "crash" - just like a program, metaphorically speaking.
There are other issues of course.
I can't go into the details, I'm not an expert in anatomy and physiology, but this I certainly know.
Not only that, but when we also put sexuality into consideration (amidst all the complementary features of life forms )- how does complementarity evolve in parallel so perfectly? Chance? Seems a bit too much to me, and I think my intuition is reliable on this one.
Necessity of evolving in the necessary complementary directions? I don't think so .
Again, I hope I'm not making an argument based off incredulity here.
"Does it not seem correct that a cat is more similar to a human than is a fruit fly, and even more so compared to a banana tree? That seems to be obviously correct"
I really don't know, my intuition isn't really that deep when considering those kinds of thoughts. Asking which one is more similar to the other seems tough to answer.

Shazzbot said...

"Anyway, Judge Rez, your fellow creotard Robert Byers takes issue with your claim a few posts down, and thinks a >90% similarity between chimps and humans should pose no problem, because God designed us using "ape bodies". Why don't you take the issue up with him, settle it thru the pertinent scientific experimentation, and then let the rest of us know when you creotards have figured out the degree of genetic similarity that should exist between organisms that have been designed by God to look as if they are related to each other. There's a good boy."
1.Nice ignorant rant! Of course, insults as I expect from any Evolutionist.
2.No, clueless, they DON'T look like they're related to each other. The unity
of life gives the powerful impression of a common designer. And I believe that, upon technical inspection, the impression gets confirmed.
Even Evolutionists (who support both chemical and biological Evo) frequently use teleological language. (I think I don't even need to bring examples)
3. Again, similarity does not mean ancestry.

Shazzbot said...

"Additionally, if we were to group all organisms by level of DNA similarity, I doubt that we would ever see any reasonable correlations to label them "evolutionary"

Does the term "nested hierarchy" have any meaning to you?"
Sorry, my goof. Yeah, I've noticed that .
Even if we group them in nested hierarchies, that is still not remotely believable at all.

Diogenes said...

Here is Ace of Spades describing Tomkins' method: "Here is the reference he [Tomkins] gives for his method of comparing optimized sequence slices. It it seems to me that all he is doing is chopping up the sequence from humans into smaller chunks and then running a standard BLAST search on each of these chunks. This is effectively the same as what I have done except I didn’t chop the human sequence up into chunks first. If he had done this as he claims then that search would have clearly shown him that each of these chunks is between 97 and 98% identical. Dishonestly (in my opinion) he makes no mention of this inconvenient fact within his paper." [Ace of Spades debunks Tomkin's GULO comparison at UD]

Diogenes said...

But what happens when BLAST does return a result? What if the two "slices" that Tomkins compares via BLAST are a little different, how does he score that? I don't know, but here is a comment from Scottwsu at the Reddit thread Ace started on Tomkins' GULO numbers:

"I have read several of Tomkins' anti-chimp/human essays and I am curious about a couple of things, and am wondering (being new to Reddit) if they have been brought up already. First, I am curious as to Tomkins' use of "-ungapped" as one of his parameters for returns in the BLASTn analysis from his article found here:

https://answersingenesis.org/answers/research-journal/v6/comprehensive-analysis-of-chimpanzee-and-human-chromosomes/

I don't use BLASTn, so I ran his parameters by one of our bioinformatics technicians and was told that this returned only sequences that were a 100% match (i.e., no gaps) to the search string. It seems to me that this all but guaranteed that the % similarity would be much lower than doing an actual site-by-site analysis."
[Comment by Scottwsu at the Reddit thread Ace started on Tomkins' GULO numbers]

Diogenes said...

Before I go into Tomkins' error in his 70% overall similarity between human and chimp genomes, and the bug in BLAST which Tomkins exploited to get it so low (pointed out by Roohif here and here), I first want to focus just on Tomkins' analysis of one gene, GULO. For this one gene, Tomkins made the remarkable claims that human-chimp is 84% similar while human-gorilla are 87% similar. This is remarkable not just because the real numbers are 97.5% and 96.6%, but also because Tomkins picked GULO to get a human-gorilla similarity that's higher than human-chimp, whereas scientists say humans are more similar to chimps than to gorillas.

Ace of Spades gave a detailed description of his methods and results in checking Tomkins' GULO numbers at UD here. Tomkins reponds with pure denialism, not really addressing Ace's data but just using ad hominems against Ace ("armchair analysis", "bogus", etc.)

Ace's take-home message is:

Humans are more similar to chimps than gorillas in this 28,800bp region [the GULO gene; this is the opposite of Tomkins' claim]

Humans and Chimps are 97.5% identical in this region (not 84% [as Tomkins claims])

Humans and gorillas are 96.6% identical in this region (not 87% [as Tomkins claims])


Why is Tomkins so wrong? Ace presents irrefutable proof that, amongst other problems, Tomkins rates "N"s (unsequenced base pairs) as not matching any nucleotide, not C, T, G or A and counting them as 0% similarity.

The smoking gun is a particular part of the GULO gene which Tomkins calls the region of "extreme discontinuity" where human and chimp are counted by Tomkins as 0% similar. Amazingly, Tomkins apparently never just LOOKED at the sequence he called "extreme discontinuity" to see how they could be so different! Of course, the chimp sequence in that region is all N's. (Or maybe Tomkins did look, and saw it was all N's, and is deliberately being deceptive.)

I will link to Ace's damning figures below. You have to see Ace's figures,, they're amazing.

So was Tomkins lying or just incompetent? Ace went to great lengths to point out these obvious errors to he and Tomkins responded to Ace with total denial, insults and hysterical personal attacks.

Diogenes said...

Dave, there is a comment from Scottwsu on this topic that I copied below.

Diogenes said...

Now I'm going to link in the damning figures that show, not only that Tomkins made a huge blunder in human-chimp GULO comparison, but Tomkins has tried to conceal his blunder by recently removing the relevant figures from the internet, without admitting he was wrong.

Tomkins' figure is-- or was-- here. It's a dead link now; Tomkins must have removed it recently, because the Wayback Machine has a fairly recent archived version, which you can look at: Tomkins' figure archived as recently as May 31, 2015, just two weeks ago. But it's gone now, as of June 18, 2015. Note this must be Tomkin's own webpage as the URL is "designed-dna.org"; the full URL was http://www.designed-dna.org/resources/human-chimp_gulo_muscle_alignment.png [dead link shortly after May 31, 2015].

The important part of this figure is the caption for B,

"B. Area of extreme discontinuity zoomed in for emphasis" [Caption of Tomkin's now-deleted GULO figure, archived at Wayback]

Of course the region he calls "extreme discontinuity" is almost all N's, unsequenced base pairs, which Tomkins counts as 0% similar to humans.

Amazingly, Tomkins either never took a look at this amazing 0% region he calls "extreme discontinuity", or he did look, and he saw it was all N's, and he is lying outright.

(A note to John Harshman: will you continue to maintain that creationists are mostly just ignorant and not deliberately dishonest after looking at that?)

Here are Ace's figures showing the "extreme discontinuity" is almost all N's. Please, all you guys, please click on Ace's figures.

Ace: "Here is an image of the aligned sequences, clearly showing that your region of "extreme discontinuity" corresponds to a large gap (unsequenced region) in the chimpanzee genome.
The MUSCLE algorithm made a few mistakes in the alignment and so it is far from ideal, but it still shows sequences that are 97.98% identical.

This diagram shows the regions where the MUSCLE algorithm could and couldn't align the sequences.

Here is your region where you claim there is "extreme discontinuity". [Diogenes notes: pay attention to all those N's]. As you can see, this region is perfectly continuous at the start and then we run into a large section where the chimp genome hasn't been sequenced. I am rather curious as to how your diagram is showing small regions of alignment within this area [note: link to archived version] when the chimp genome hasn't even been sequenced here.

Here are the MUSCLE aligned sequences for you to download and check.

Once again, if you count the differences, there are 566 SNPs and 45 gaps out of a total of 30,312 complete positions. Adding this up, we find that the sequences are 97.98% identical.
" [Ace of Spades debunking Tomkins' GULO comparison at Reddit]

judmarc said...

Let's focus on the other claim - that the Chimp is the most similar
organism to us even if the similarity is only 70%.
Let's look at how other organisms are genetically similar to us. By what degree, that is. Now, again, I don't claim that these numbers are final, but:
"Cat: 90%
Cow: 80%
Mouse: 75%
Fruit Fly: 60%
Banana: 50%"


Since the blog post you quote gave two estimates of genetic similarity between humans and chimps, one 98.8% and the other 96%, thus agreeing with the proposition that we are closer genetically to chimpanzees than any other animal, please tell us why we should not dismiss you as another dishonest fool for (1) oh so conveniently leaving that information out, and (2) having any sort of thought that what you'd done wouldn't be immediately obvious.

Diogenes said...

Before I continue with the numbers, I want to pause to examine the infuriating and dishonest creationist responses to Ace's detailed takedown above. This is of course from UD.

Tomkins attacking Ace: The BLASTN analyses done in this paper were performed after stripping all N’s from the data set and sequence slicing the large contiguous sequence into optimized slice sizes – all done on a local server using optimized algorithm parameters. My data not only takes into account gaps, but sequences present in human and absent in chimp, and vice versa. Doing an amateur armchair analysis on the BLAST web server with default parameters never designed for a one-on-one large scale genomic regional comparison as noted in the comment above by aceofspades25 is bogus. Of course, if the paper was actually read in it’s entirety in regards to the above comments this would have been obvious.

Also, as noted in several evolutionary papers, which I cited in my paper, the large scale comparison and major differences in structural variability surrounding the GULO regions between humans and great apes in the intronic areas has been noted before. Interesting that the misleading post by aceofspades25 did not make note of that. My paper was in fact accurate in all respects and true to previous findings published by evolutionist themselves. My work just hashed out and exposed what was already known, but never previously elaborated upon because it shows just another aspect of what a complete fraud the human evolution paradigm truly is."
[Tomkins attacking Ace at UD]

Here he's not just attacking Ace, he's saying 1. "evolutionists" (scientists) already published his numbers, and 2. "evolutionists" (scientists) didn't already published his numbers because evolutionists ignore his fraudulent, concocted evidence because his fraudulent figures dispove evolution.

Can we say "scientific fraud" now!? Sure, it may have started out as incompetence. But when can we say "scientific $%^&ing fraud"!?

Now let's see the grovelling, uncritical response from Uncommon Descent. After Tomkins' content-free ad hominem and conspiracy theory, BornAgain77 grovels before the knees of Tomkins, creo-fellates the scientific fraudster, then closes the thread to eliminate further comments.

BornAgain77: Dr. Jeffrey Tomkins, thank you so much for taking the time to clear that up. ,,, Darwinists have been trying to diss your work ever since it came out. I have to admit that I did not know which data set to believe, theirs or yours, until I watched your interview with [fake "professor" creationist] Ian Juby. I then knew that you are a man who knows exactly what he is doing in this field and who takes his research seriously. Thanks again." [BornAgain77 begs for scientific fraudster Tomkins to gay-rape him at Uncommon Descent]

These are the proponents of what they call "critical thinking on evolution."

Diogenes said...

Above, I was mostly considering Tomkins' fraudulent numbers on the human-chimp comparison of just the GULO gene, as debunked by Ace of Spades. Now I want to switch to Tomkin's whole genome comparison, and Roohif's comments about it.

I've already discussed Tomkins' general methodology and metric above.

At the excellent blog, Eye on the ICR, commenter Roohif addresses his findings on Tomkin's whole genome comparison. I'll copy his comments.

Roohif: "I’ve figure[d] out how he [Tomkins] gets 70% on his Comprehensive Analysis paper, and I assume the same problem affects the GULO one.

In summary, he is using a version of BLAST+ with a bug in it that reduces the number of hits that are returned. Say for example, you submit 10,000 queries, one a time, you will generally get 10,000 results. What happened in Tomkins’ paper was that he submitted the 10,000 queries in one go, and only 7,000 or so results were returned. The bug exists in v2.2.27, and was fixed in 2.2.29.

I’ve written a paper on this and submitted it to Answers in Genesis’ journal. Obviously I’m coming up against a lot of resistance, but I’m grinding him down slowly, on every point that he has made."
[Roohif's comment of Jan. 16, 2015, at Eye on the ICR]

Here is his comment from a few months earlier:

I’m quite familiar with BLAST. The word length parameter really only finds 100% _INITIAL_ matches. So, if you use a word length of 7, you are likely to get a lot of “possible matches”. The BLAST algorithm then tries to extend the alignment in either direction. A word length of 10 is perfectly fine – I think the default is 11.

Tomkins’ unknowingly encoutered a bug in the BLAST software that does not return a proportion of hits when a large number of queries are submitted at once. So, for example, if Tomkins submits 300,000 queries, he might only get hits for 200,000 of those. The bug, is that if you re-submit the remaining 100,000 queries, you will get more hits. Repeat and resubmit until you don’t have any queries left.

I discovered this bug while I was writing my own response to Tomkins. I have been communicating with him over the last few months, and he is now aware of the problem, he just hasn’t spoken the magical words yet: “I was wrong”
[Roohif's comment of Sept. 17, 2014, at Eye on the ICR.]

I don't have more info on Roohif's work, but one of us should email him for more details. He gives us his email address:

If you want a sneak peak at the draft [Roohif's debunking of Tomkins' whole genome analysis], just send me a message. glenn at Delta, Uniform, Bravo, Zulu dot com dot au.

So I'll email him for more info.

Diogenes said...

Roohif must be Australian so there's a 12 (?) hour time difference.

Anonymous said...

Rez,

I did not say that similarity means common ancestry. I asked several times why lower similarity would mean anything to creationists unless they wanted to argue that we share no ancestry with chimps.

Matching an evolutionary scenario is not the whole story about why we accept evolution, but I will not add to this because you seem to have talent to miss the point. So I insist: Joe's point was that if we apply some measure that gives us less similarity, chimps still will be closer, then other organisms, etc. So, the similarities all along will be lower, that does not make the common ancestry, nor the evolutionary scenario disappear.

Again, seems clueless for you to ignore that there must be a reason these creationists want the chimp to be less similar. What do you think it is if not deny common ancestry. If not to deny common ancestry, then why did they make so much out of it.

Petrushka said...

I tend to agree that he knows exactly what he is doing.

Shazzbot said...

"Since the blog post you quote gave two estimates of genetic similarity between humans and chimps, one 98.8% and the other 96%, thus agreeing with the proposition that we are closer genetically to chimpanzees than any other animal, please tell us why we should not dismiss you as another dishonest fool for (1) oh so conveniently leaving that information out, and (2) having any sort of thought that what you'd done wouldn't be immediately obvious."

Jesus, big deal. I DID notice that. So what? That's still irrelevant here. I was referring to the claim that "even a 70% difference would mean that chimps are the organisms most similar to us".(attempting to refute it, that is)
So no, I am not dishonest or deceptive. I'm not your run-off-the-mill Evolutionist to be like that. (Not to say that any camp is 100% 'clean', but still)
----------------------------------------------------------------------------------------------------
"I did not say that similarity means common ancestry. I asked several times why lower similarity would mean anything to creationists unless they wanted to argue that we share no ancestry with chimps."
To which I responded - a)A bigger Haldane's dilemma b)Just exposing Evolutionist dishonesty c)Both a and b

"Matching an evolutionary scenario is not the whole story about why we accept evolution, but I will not add to this because you seem to have talent to miss the point"
Me? Or the evolutionists who completely mistook my first two comments?
I think I haven't demonstrated what you claim here,
"Again, seems clueless for you to ignore that there must be a reason these creationists want the chimp to be less similar."
I didn't ignore. And I listed three (well,two) possible reasons.

"What do you think it is if not deny common ancestry. If not to deny common ancestry, then why did they make so much out of it."
Not just to "deny" common ancestry. See b).
"So I insist: Joe's point was that if we apply some measure that gives us less similarity, chimps still will be closer, then other organisms, etc. "
Hmm, I really don't know whether that will be 100% true. Whether of the chimp or any other apelike organism in regards to a human being. IF the similarity estimates I gave are accurate, then judging by the phenotype sure is unreliable. And, for the reasons I gave above, a major problem for evolutionary scenarios.
But , let's say it's true! In that case, decreasing the amount of similarity still implies a).
"So, the similarities all along will be lower, that does not make the common ancestry, nor the evolutionary scenario disappear."
Same answer as above.

Shazzbot said...

judmarc I am not obliged to mention absolutely every detail about the sources I give. Unless omitting creates a grave misunderstanding. That, I did not do.

Diogenes said...

I emailed Tompkins and challenged him to show up here.

Hello Dr. Tomkins,

Your claim of 70% genetic similarity between humans and chimps is being demolished at the blog of real geneticist Prof. Larry Moran:

http://sandwalk.blogspot.com/2015/06/creationists-discover-that-human-and.html

Read the comments. You're coming off looking very bad indeed. That blog is frequented by several real scientists including Joe Felsenstein, John Harshman and Nick Matzke. So that's an audience you can't snow.

I challenge you to show up and defend your bogus 70% figure in front of an audience of real scientists. Right now you're looking epically bad.
Sincerely,
Diogenes

judmarc said...

judmarc I am not obliged to mention absolutely every detail about the sources I give. Unless omitting creates a grave misunderstanding. That, I did not do.

Well, you've certainly answered the inquiry as to whether you're another dishonest fool with a resounding "Yes!" to both.

You just happened to omit the most relevant figures from the blog source you quoted (the ones about similarity between humans and chimps, which is Tomkins' subject), which just happen to contradict Tomkins, meanwhile quoting Tomkins' figure for human-chimp similarity and comparing it to the blog's figures for kitty-cats and such. Yeah, that was sheer coincidence, and I'm the effing Queen of England.

You'll of course understand if I don't respond to much that you say from here on in except where it's just too damned entertaining to pass up.

Diogenes said...

If anybody has a Reddit account, could you please message Aceofspades25 and ask him to join in here?

Anonymous said...

Rez,

IF the similarity estimates I gave are accurate, then judging by the phenotype sure is unreliable. And, for the reasons I gave above, a major problem for evolutionary scenarios.

But here you're ignoring Joe's comment. If we were to inflate the difference between humans and chimps in some kind of methodology, we should use the same methodology when comparing us to other organisms. Then apes would be at the same position they were with the 98% similarity. You missed the point, and somebody else already explained this to you. It would be dishonest to use different methods for different organisms just to make the creationist case.

To which I responded - a)A bigger Haldane's dilemma b)Just exposing Evolutionist dishonesty c)Both a and b

Hum. My apologies, I did not see that because all the other times I was asking this question you would repeat that similarity and common ancestry are not the same (missing the point). So I assumed that was all you were going to give me. They sure use it to imply that "evolutionists" are dishonest, so you have a point.

I would answer those two claims (Haldane's and dishonesty, besides other things you said), but no need. We're talking about missing the point. I leave it there.

Anonymous said...

Rez,

"judmarc I am not obliged to mention absolutely every detail about the sources I give. Unless omitting creates a grave misunderstanding. That, I did not do."

Yes you did. By missing the estimates for humans and chimps, then using the creationist calculation with those, you're creating a grave misunderstanding. You're implying that the creationist estimate and those from that web site are compatible. Calculated in the same way. But the omitted numbers for chimps and humans tell a different story.

Jass said...

Genomes of 2 humans can be only 93% similar. If chimp's genome is 98-99% similar to human what genes are responsible for undeniable differences in body plans and both physical and intellectual abilities?

The question is obviously hypothetical because there is no way that genes alone can be responsible for ALL THE DIFFERNCES. Darwinists would like to be it, but I will wait for the actual evidence to support such a claims.

Jass said...

This is a topical post by Moran when he has no answers and no scientific proof. He is pissed again at ID but he has no answers so he lets "others" to vent. This is actually the most common trick done by blog hosts. I can bet he is dying to blow some steam, but he hardly does as himself. I'm just wondering who does the most steam blowing that Larry as the host should have done?

Glenn said...

Hi guys,

Got an email from diogenes this morning asking me to join the conversation, so here I am :D

Yes, Tomkins gets his 70% figure because of a bug in the BLAST software, and yes he uses "ungapped" incorrectly. I wrote a paper on this and submitted it to Answers Research Journal, basically to see if creationists will publish "evolutionists" papers.

The paper is here (2 PDFs):

https://www.dropbox.com/sh/dm2lgg0l93sjayv/AAATnWSJdER53EYEYZvcgiwma?dl=0

I submitted it back in September 2014, and went back and forth with Tomkins (through the editor) for a couple of months, and Tomkins now knows that he has a problem. Unfortunately, he has gone silent since December 13th 2014. The editor assures me that Tomkins is still working on a response, and that I'll have it any day now. But that's what he has been saying since mid-January.

Anyway, I think my paper is pretty clear, let me know your thoughts.

Glenn said...

Hey!

Yeah I'm here, and I'm awake. Gotta drop the kids at school though, and then run some errands :)

My paper is here:

https://www.dropbox.com/sh/dm2lgg0l93sjayv/AAATnWSJdER53EYEYZvcgiwma?dl=0

Tomkins-BLAST.pdf and Figures.pdf

Let me know what you guys think.

Faizal Ali said...

Genomes of 2 humans can be only 93% similar.

Ha ha ha! Including identical twins?

No, really, where did you come up with this figure? We could use a laugh.

Diogenes said...

Glenn,
Thanks for showing up. I'd read your paper, but 2 different apps of mine say it's not a valid PDF.

Unknown said...

You are too generous there, Diogenes. If you create a random sequence and look for the best BLAST match, you will generally get more that 25% sequence identity (though by how much depends on the length of the sequence you use). After all BLAST tries to align the sequence, and while the expected percentage of matches if 25%, BLAST will return the best result.

Glenn said...

Really? Dropbox has a built-in PDF reader as far as I can tell, and they work fine for me. Are you clicking on them, and then hitting "download" at the top right?

In any case, I'll email you a copy now ..

TheOtherJim said...

The question is obviously hypothetical because there is no way that genes alone can be responsible for ALL THE DIFFERNCES. Darwinists would like to be it, but I will wait for the actual evidence to support such a claims.

Actually, it is only you claiming that is what is believed. In reality, no one thinks it is 100% genetics. For instance, at the "Darwininst" hotspot like the Understanding Evolution web page it clearly states "In reality, of course, an organism's phenotype — the physical trait it ends up having (e.g., being four feet tall) — is usually the result of both genetic factors, environmental factors, and the interactions between them.

Stop misrepresenting. You god hates "bearing false witness".

Glenn said...

Just reading the comment about "ungapped" higher up in the thread. No, it doesn't require a 100% match, but using "ungapped" does not allow for indels to be included in the calculation.

BLAST works by finding a small match (and the length of that match is determined by the word_size parameter), and then it tries to extend that match in either direction. BLAST keeps a "score", that score increases each time a match is found at the next position, and takes a penalty if it is a mutation. It can also allow for indels: when it encounters a putative indel, it opens a gap (which is a penalty) and then it keeps extending that gap (penalty) until things start aligning again. If the gap gets too long, then your score gets too low, and it gives up.

You can modify all these parameters in BLAST to adjust the scoring as you see fit.

Glenn said...

Sorry, my takeaway point - which I completely left off - was that Tomkins uses ungapped in his comparisons, but in much of his other work is he is critical of studies that (he _believes_) exclude chunks of data. Rather hypocritical I think.

Glenn said...

Hi Joe and John,

I've written a paper on this bug, and how it affected his results:

https://www.dropbox.com/sh/dm2lgg0l93sjayv/AAATnWSJdER53EYEYZvcgiwma?dl=0

There are two PDFs there, let me know if you can open them (apparently diogenes had some trouble).

Cheers!

Glenn said...

I've also written a blog post on this "Niwrad-style" metric, because Tomkins cites a "paper" (and I use that term loosely) as supporting his case:

https://roohif.wordpress.com/2014/06/04/progetto-cosmo-genome-comparison/

christine janis said...

I think that what we're dealing with here is a Potomkin genetic analysis.

Faizal Ali said...

Link works fine form me. Thanks for sharing that, Glenn.

Another embarrassing moment in the annals of ID creationism.

Mikkel Rumraket Rasmussen said...

I'd like to ask him: If your fundamentalist creationist beliefs are really are true, why do you have to lie to support them?

Mikkel Rumraket Rasmussen said...

"1st. Clueless, similarity DOES NOT imply ancestry. "

No, not in and of itself. But nesting hierarchical arrangements of similary in branching patterns matching exactly what one would expect if the observed mechanism of inheritance is allowed to continue over geological time, does.

QED.

Glenn said...

Correct! Especially when those branching patterns are found in DNA/proteins that have been found - empirically - to be functionally equivalent. Pretty difficult to invoke common designer ...

Mikkel Rumraket Rasmussen said...

As all creationists Judge Failz neglects to consider the fact that the mechanism of inheritance and the resulting nesting hierarchical patterns it produces is observed empirical facts. Organisms can be observed to evolve and produce such branching patterns in the lab.

The similarity implies descent-argument isn't erected in a vacuum. We see organisms reproduce, we see that their offspring are different from their parents, genetically as well as physiologically. When we allow this process to iterate over many generations, we can directly see the nesting branching hierarchies that result from it when we sequence the genomes of the species in the experiments. We thus have a bona fide, emprically based expectation of such patterns. The fact that we then find them, when sequencing and comparing genes from any organism detected so far, agreeing to high degree with the same patterns reconstructed from comparative anatomy, the fossil record and so on, is what confirms evolution. Nobody thinks "omg these two things are similar, therefore they must be related". It isn't and never was that simple.

It's because we know organisms reproduce by replicating their genomes, because their genomes mutate, because their physiology and morphology is largely determined by their genomes etc. etc. that we draw the connection.

Faizal Ali said...

Yes you did. By missing the estimates for humans and chimps, then using the creationist calculation with those, you're creating a grave misunderstanding.

In fairness to Judge Rez, I don't think that is what he did there. It's hard for me to be certain, because (as is so often the case with these creotards) his writing is as inscrutable as his thinking is obtuse. But what he seems to be arguing is along these lines: "Even if we were to accept that the genomes of chimps and humans are 96-98% similar, look at all the other claims that 'evolutionists' make. They think cats are 90% similar to humans, and bananas 50%. That is obviously absurd." Am I correct there, Judge Rez? If so, you seem you have skipped the part where you explain why it is unreasonable to say that a cat's genome is 90% similar to a human's, and a banana plant's 50%. You just seem to arguing based on your expectation that their genomes should be more dissimilar. An expectation based on nothing more than your ignorance and stupidity.

Faizal Ali said...

@Judge Rez:

Again, I hope I'm not making an argument based off incredulity here.

Well you are, but that's not all. You're also basing your argument on your own ignorance regarding evolutionary theory. If I'm not mistaken, you are repeating Ray Comfort's question:

I ask questions about where the female came from for each species. Every male dog, cat, horse, elephant, giraffe, fish and bird had to have coincidentally evolved with a female alongside it (over billions of years) with fully evolved compatible reproductive parts and a desire to mate, otherwise the species couldn’t keep going. Evolution has no explanation for the female for every species in creation,

http://scienceblogs.com/pharyngula/2009/03/05/its-a-conspiracy-1/

To which the only appropriate response is to gape in astonishment at the depths of stupidity the human mind is capable of reaching

Anonymous said...

lutesuite,

In fairness to Judge Rez, I don't think that is what he did there.

But I Rez was. Rez was saying that humans and chimps at 70% would be a problem for the evolutionary scebnario if put together against those. He insisted too much on that.

He started that part with:
"Let's focus on the other claim - that the Chimp is the most similar
organism to us even if the similarity is only 70%."


The gives us those numbers from elsewhere to show that then chimps would not be the most similar to us.

Followed by:
"Thus, in this case, it probably can turn that other argument (even though it's irrelevant in regards to Thomkins' work) upside down. "

Rez is clearly suggesting that it's appropriate to put the 70% creationist measure against similarities calculated in a different way.

If that is not enough, Rez tells us:
"Jesus, big deal. I DID notice that. So what? That's still irrelevant here. I was referring to the claim that "even a 70% difference would mean that chimps are the organisms most similar to us".(attempting to refute it, that is)"

Attempting to refute it by presenting those other similarities, some higher than the 70%.

Then again, referring to the evolutionary scenario:
"IF the similarity estimates I gave are accurate, then judging by the phenotype sure is unreliable. And, for the reasons I gave above, a major problem for evolutionary scenarios."

So, clearly Rez was putting together the creationist chimp-human similarity with those available at that site to show that chimps would not be our closest relatives. That the evolutionary scenario would be broken by the lack of concordance between the phenotypes and the similarities.

Anonymous said...

Wow Glenn. That's patience.

Diogenes said...

My problems opening your PDF may be because I tried it on my droid. I'll try it on my laptop later.

Diogenes said...

No, it's Potemk... I see what you did there.

Diogenes said...

So Glenn, if we compare

HEYHOWAREYA

with

HYHOWAREYAZ

with the "ungapped" option, it would return a score of 1/11, because of the in del at position 2, which throws off the alignment?

It would seem such a measure of similarity would depend on window size and indel rate. I'd guess that if the similarity for single nuclotides were q (= 98.5% for human-chimp), then the "uncapped" option would on average return a score like

s = q

For comparisons with no indels, and

s = (1/2)*(q + .25)

For comparisons with indels. The .25 is for random sequence, and I'm assuming the indel is in the middle of the sequence.

The rate of having an indel would depend on window size L. If the probability of an indel at any given position = p, then the probability of no indel at any given position is 1-p, and no indels at all in a window of size L is (1-p)^L.

Therefore the average score would be

s = q*(1-p)^L + (1/2)(q + .25)*(1 - (1-p)^L)

The point is, with the option"ungapped", Tomkins' similarity s is not just a measure of q, but increases with window size L, and with pointwise indel probability p.

Glenn, could you point us to BLAST documentation for the "ungapped" option?

Diogenes said...

P.S. Note that for very large window sizes, Tomkins' score would approach not q (the actual human-chimp similarity, 98.5%) but would instead approach the average of 98.5% and 25%, = 61.75%.

What counts as a "large window" will depend on indel probability.

Diogenes said...

Glenn, that is a great blog post and quite hilarious. Glenn's post is about these Italian creationists, Progetto Cosmo, who make the same blunder Niwrad made, but they used a larger window, so they got an even lower similarity than Niwrad.

What's hilarious is that we all thought we'd never see the Niwrad blunder again! When Niwrad made his mistake, a couple years ago, it was so obvious and indisputable that many of us suspected Niwrad was a "Poe" pretending to be fundie so as to make creationists look hilariously stupid. But here the Niwrad error raises its ugly head again.

Glenn said...

_"with the "ungapped" option, it would return a score of 1/11, because of the in del at position 2, which throws off the alignment?"_

Yes - assuming there were a bunch of bases to the left of that string that were matching - then ungapped would stop at where the 'E' is missing. Although if you submitted those as-is to BLAST it would find the match in the middle: 'YHOWAREYA'. The example I give in my paper is clearer because the indel is right in the middle, not at the start :)

_"It would seem such a measure of similarity would depend on window size and indel rate."_

By 'window size' do you mean the 'slice size' that Tomkins talks about? Or the 'word_size' parameter to BLAST?

If you mean 'slice size', then yes, this is quite important if you are using ungapped, because you will hit an indel every few hundred bases. You'll see the results in Figure.pdf that show the change in alignment length with and without ungapped. Word_size isn't really important for something like humans and chimpanzees, because with such high similarity you'll get plenty of word matches. You should only lower the word_size if you're comparing animals that are fairly divergent.

I don't really have time right now to follow your maths, but I think I see what you're getting at. BLAST is fine at counting mutations, because the positive matches would continue after the mutation (and that increases the score). On the other hand if you have an 1bp indel, the sequences are then offset - you will get random 25% matches - and your score will go to zero. If you allow for indels (i.e. by NOT using ungapped), then BLAST has the chance to re-sync.

_"The point is, with the option 'ungapped', Tomkins' similarity s is not just a measure of q, but increases with window size L, and with pointwise indel probability p."_

His problem with using ungapped is that it will necessarily produce shorter alignments, but in his calculations he takes the full query length into account. You can't do both - you either use ungapped and alignment length, or no ungapped and max(query length, alignment length).

_"Glenn, could you point us to BLAST documentation for the 'ungapped' option?"_

http://www.ncbi.nlm.nih.gov/books/NBK279690 is the whole manual, which is searchable. Look for 'gapopen' and 'gapextend'. Using ungapped is effectively setting both these penalties to infinity.

Diogenes said...

But, Glenn, you did not need to do a full.Monte Carlo simulation to show that Progetto Cosmo's low sequence similarity, 63%, was entirely due to the big window size they used, 30 nucleotides. (At his blog Glenn tried simulated comparisons with different pointwise mutation probabilities until the average 30-bp window exactly matched for 63% of all windows, since Progetto Cosmo claim 63% overall similarity.) Instead of simulations, all you have to do is compute the probability of a 30-bp window containing zero differences if the probability of no difference at any given base pair is 98.5%. The probability of no differences for 30 bps is then

(0.985)^30 = 0.635

And that explains how Progetto Cosmo got their 63% human-chimp similarity!

Same dumb mistake as Niwrad.

Glenn said...

True, but I wasn't so sure if that quick probability is applicable if the mutations are not uniformly distributed. Given the results that I got, that might be a valid shortcut, I just felt the Monte Carlo would be a bit safer. I am by no means a statistician :D

Piotr Gąsiorowski said...

If I may quote myself,

It’s a good thing Tomkins doesn’t write plagiarism detection software. If someone copied a whole book (say, 200 pages of text) and then omitted a letter or inserted a comma once per page, Tomkins’s method would demonstrate that the original and the copy ale less than 1% similar.

Tom Mueller said...

Anticipating the IDiots next line of defense, their equivalent to calling up their Imperial Guard:

Scanning through genomic databases of multiple species, permitted the identification of some of highly mutated areas that clearly contribute to human-specific traits

These so-called "Human accelerated regions" are obviously subject to intense selective pressure and to my reading diminish Neutral Theory’s importance in explaining Human Chimpanzee divergence while simultaneously validating the evo-devo paradigm

http://rstb.royalsocietypublishing.org/content/368/1632/20130025.short

Bringing us to our next question:

How do the intellectually dishonest sophists at Uncommon Descent wave their hands to rescue their untenable paradigm?

http://www.uncommondescent.com/intelligent-design/more-accelerated-sequence-evolution/

http://www.gotquestions.org/human-chimp-DNA.html

Quick translation: the major premise of Neutral Theory and the minor premise of HARs elicit the inevitable conclusion that Evolution must be false.

I think I shall rework this as an essay question for my high school students. The solution to this non-dilemma would confirm whether or not my students indeed understood the message on this link:

http://evolution.berkeley.edu/evosite/evo101/IIIE5bNeutraltheory.shtml

Confused said...

"Genomes of 2 humans can be only 93% similar. "

You will have to explain this.

Shazzbot said...

Part I.
judmarc :
"
Well, you've certainly answered the inquiry as to whether you're another dishonest fool with a resounding "Yes!" to both. "
No, moron. I said IF the similarity were 70%, and the figures from the Evolutionist source for OTHER organisms were correct, THEN that would make the chimp farther to us than compared to those other organisms.
I know the ape-man figures contradict Thomkins' , that is STILL irrelevant. My argument involves the proposition that Chimp-Ape similarity is 70% .
(Even though I say that must not absolutely be the case)
And, yeah, it's entertaining not to miss up on your stupidity.
photosynthesis:
You had the same objection as judmarc. I answered that (except that,for you, without the offenses).
"Rez is clearly suggesting that it's appropriate to put the 70% creationist measure against similarities calculated in a different way."
No I'm not. I'm not saying that comparing similarities obtained via Creationists' methods and those via Evolutionists' methods (even though, for different organisms in relation to human beings) can be measured against reliably.
I'm not even saying that the estimates themselves(regardless of source) are 100% reliable.
Now let's handle lute-"TheEvotard"-suite:
I've read Myers' ranting post about how bad Comfort's book is and about how incredibly stupid he apparently is. And here's how Myers actually attempts to provide an explanation:
"Populations evolve, not individuals, and male and female elephants evolved from populations of pre-elephants that contained males and females. Species do not arise from single new mutant males that then have to find a corresponding mutant female — they arise by the diffusion of variation through a whole population, male and female."
Here, I really don't quite understand what Myers says.
1st."Populations evolve, not individuals". I've searched for what that meant.
Here:https://answers.yahoo.com/question/index?qid=20081109160705AAsZJeP
Basically, the answers from these guys fall in line with what I already know:
Individuals don't evolve in their lifetimes. Their successors will be born different from them. That's it.
2nd. "and male and female elephants evolved from populations of pre-elephants that contained males and females" . OK, asserting that sexually compatible organisms came from other, different, but in-themselves sexually compatible ones, but not really explaining how there came to be sexual compatibility (I'm talking about how the organisms miraculously happened to develop in parallel ,sexually, in a compatible fashion. Where some changes in one happened to be followed by necessary changes in the other. High improbability.).
3rd. "Species do not arise from single new mutant males that then have to find a corresponding mutant female" - all right, I knew that and wasn't implying that.
(his explanation)"they arise by the diffusion of variation through a whole population, male and female"
Science-y words, no explanations. He just asserts that they do.
Even if some organism wouldn't become different in his lifetime, that it'd be his successor, that STILL doesn't make the problem go away.
What if the organism's successor is born with a different sexual organ ?
Not only that. But what about the ORIGIN of sexuality in general? It'd have to be either
a) compatible right from the start from both partners(and they'd have to be close enough, of course and born in certain times to ) OR
b) the one who developed the sexual organ, or at least a building block for it, would still have to be able to reproduce asexually.(And then there's a complicated story, I need not go on)

Shazzbot said...

Part II.
Rasmussen. Another clueless Evotard:
""1st. Clueless, similarity DOES NOT imply ancestry. "
No, not in and of itself.//Agreed
But nesting hierarchical arrangements of similary in branching patterns...//Which we
can do for cars, planes,computers,motorcycles("evolving designs")etc...
...matching exactly what one would expect if the observed mechanism of inheritance is allowed to continue over geological time, does.//"the observed mechanism of inheritance".
So this is the mechanism that apparently creates the differences(branching points) from the initial organism( the "top point" of the generated nested tree).
This implies that we should also see two things : a) a coherent correlation between the differences. Which we DON'T. and b) it should be possible to make those transitions.
If those estimates I gave are correct, (whether we use a 70% estimate or a 98% estimate for the chimp,doesn't matter), that doesn't seem so likely. Again, you have engineering problems.
Back to lute-TheMoron-suite:
" It's hard for me to be certain, because (as is so often the case with these creotards)//It's easy for ME to be certain, as it is with you EvoTARDS
his writing is as inscrutable as his thinking is obtuse.//I use unambiguous terminology, I explain my points in detail. Your UNDERSTANDING is poor.
But what he seems to be arguing is along these lines: "Even if we were to accept that the genomes of chimps and humans are 96-98% similar//Well, I didn't say that, but , yeah, nested hierarchies would still look pretty ridiculous. (If the estimates are reliable, of course. Again, I utilize caution)
look at all the other claims that 'evolutionists' make.//Of course, there is no such thing as an "Evolutionist". Evolution is a fact, of course! Silly me!
They think cats are 90% similar to humans, and bananas 50%. That is obviously absurd." Am I correct there, Judge Rez?//Correct. No complaint here.
If so, you seem you have skipped the part where you explain why it is unreasonable to say that a cat's genome is 90% similar to a human's, and a banana plant's 50%. You just seem to arguing based on your expectation that their genomes should be more dissimilar. //In the case of an Evolutionary scenario, certainly.
An expectation based on nothing more than your ignorance and stupidity."//No, moron,
an expectation of correlation(in case of evolutionary scenarios) .An expectation that there
are no conceivable engineering problems.(and there are). Ignorance and stupidity (and "just so" stories) are YOUR work. Not mine.

Mikkel Rumraket Rasmussen said...

"But nesting hierarchical arrangements of similary in branching patterns...//Which we
can do for cars, planes,computers,motorcycles("evolving designs")etc..."


Organisms replicate themselves. Planes, computers and motorcycles don't, so that's that "argument" out the window.

Back to the drawing board kiddo.

Mikkel Rumraket Rasmussen said...

Here's the ultimate just-so story: Goddidit. In an instant.

With magic.

All problems solved. He did it "just so". *puff* and there it was. Life: explained.

IDcreationists should try to keep their hypocricy in check before they elect to open their mouthes about "just so" stories.

Shazzbot said...

Rasmussen. AGAIN.
As all creationists Judge Failz
neglects to consider the fact that the mechanism of inheritance and the resulting nesting hierarchical patterns it produces is observed empirical facts. //I'm not saying that VARIATION doesn't occur. I'm just saying that the mechanisms for these undirected changes taking place are not potent enough to make some of the proposed transitions.
Organisms can be observed to evolve and produce such branching patterns in the lab. //Yes, evolve from one gen to the next within certain LIMITS (Lenski's experiments show that .60,00 generations, for crying out loud).

The similarity implies descent-argument isn't erected in a vacuum. //Assertion, not proven.
We see organisms reproduce, we see that their offspring are different from their parents, genetically as well as physiologically.//Within certain LIMITS.
When we allow this process to iterate over many generations, we can directly see the nesting branching hierarchies that result from it when we sequence the genomes of the species in the experiments.//Yes! Except that the observed nested hierarchies are certainly limiting and don't give you the right to say that this mechanism was capable of producing the other alleged hierarchies!
We thus have a bona fide, emprically based expectation of such patterns.//In your looney dreams
The fact that we then find them, when sequencing and comparing genes from any organism detected so far//A lie. I see no coherent pattern. Also, how about CONVERGENCE? I.e. similarities being there where they SHOULDN'T be?Graph it and compare the graph to the common descent graph. Don't tell me how it doesn't turn common descent's entire head around. It contradicts. A common designer notion doesn't contradict itself. Especially in the light of Message Theory.
agreeing to high degree with the same patterns reconstructed from comparative anatomy,//Another lie. For SOME organisms grouped within nested hierarchies, YES.
But what if what I've shown is correct? What would comparative anatomy say if we grouped the cat and the human in the same nested hierarchy?
the fossil record and so on, is what confirms evolution.//In your looney , deluded dreams
Nobody thinks "omg these two things are similar, therefore they must be related". It isn't and never was that simple.//That's right. They have to envelop the organisms with stories, fictions.


It's because we know organisms reproduce by replicating their genomes, because their genomes mutate, because their physiology and morphology is largely determined by their genomes etc. etc. that we draw the connection. // A fictional and incoherent one.

Shazzbot said...

Rasmussen. You are beginning to annoy me.
"Here's the ultimate just-so story: Goddidit.//I noted this fallacy once.
In an instant. With magic."//Supernatural or Non-material != Magic. And not necessarily in an "instant".
Just because it sounds simpler, doesn't mean it's a "just-so" story. The notion of a common designer is a very coherent one.
"
Organisms replicate themselves. Planes, computers and motorcycles don't, so that's that "argument" out the window.

Back to the drawing board kiddo. "
NO FRICKIN WAY! Wow! I didn't know that! Except that they replicate, and their replica changes within CERTAIN LIMITS.

Shazzbot said...

A correction: By "they" I mean organisms, of course.
And by saying "their replica(successor) changes" I don't mean
that he/she changes within his/her lifetime. The change happens
before he/she is even born.

Larry Moran said...

These so-called "Human accelerated regions" are obviously subject to intense selective pressure and to my reading diminish Neutral Theory’s importance in explaining Human Chimpanzee divergence ...

Really? How much of the total divergence do you think is due to adaptation (natural selection) and not random genetic drift of neutral alleles? Keep in mind that 90% of the genome is junk and the total divergence is close to the population genetics value expected from drift of neutral alleles (molecular clock).

Anonymous said...

Rez,

"I said IF the similarity were 70%, and the figures from the Evolutionist source for OTHER organisms were correct, THEN that would make the chimp farther to us than compared to those other organisms."

Which is exactly where you missed Joe's point, and where you pretend that those two things, the creationist count and the "evolutionist" counts are compatible. You can't refute the evolutionary scenario by "fixing" only the human-chimp counts. This was Joe's point, and you continue missing it.

"I'm not saying that comparing similarities obtained via Creationists' methods and those via Evolutionists' methods (even though, for different organisms in relation to human beings) can be measured against reliably.
I'm not even saying that the estimates themselves(regardless of source) are 100% reliable."


Your "IFs," denials, and clarifications are inconsequential to the implication made by putting those numbers together. The implication continues to be that those numbers are compatible. Joe's point was, all along, that by using the same methods for calculating the similarities chimps would be at the same relative position as they are.

Mikkel Rumraket Rasmussen said...

"Rasmussen. You are beginning to annoy me."

How sad. Take a cold shower, I'm going to respond to your demonstration of your encephalitic subnormality soon.

Faizal Ali said...

@ photosynthesis.

I'm sorry I questioned you. I would have thought it was beyond even a creationist to be so stupid as to make the argument "A creationist says the chimp genome is 70% similar to the human genome. And competent scientists know that the cat genome is 90% similar. Therefore evolution is false."

I underestimated Judge Rez's stupidity. I should have learned by now: Never underestimate the stupidity of a creationist.

judmarc said...

Rasmussen. You are beginning to annoy me.

Lemme guess: You'd kick his ass, if you could only read the road signs on the way to his house.

Anonymous said...

lutesuite,

You're in good company. I often start giving them the benefit of the doubt, only to be disappointed again.

judmarc said...

Within certain LIMITS.

* * *

Also, how about CONVERGENCE? I.e. similarities being there where they SHOULDN'T be?


Oh noes, fellow evolutionists, UPPER CASE! All is lost, run away!

(Man, it just gets better. Please tell me you'll be here all week, Rez. Oh, and don't forget to tip your waiter.)

Anonymous said...

Rez,

"I've read Myers' ranting post about how bad Comfort's book is and about how incredibly stupid he apparently is."

Wow, if this is the "how did the female know to evolve at the same time as the male" or "how does the newly evolved male find a compatible newly evolved female," then sure, either Ray is incredibly stupid, or he's a genius of rhetoric and lemon-car salesmanship (I think it's the second). He knows that his followers are stupid enough to fall for that unimaginably stupid line.

And here's how Myers actually attempts to provide an explanation:
"Populations evolve, not individuals, and male and female elephants evolved from populations of pre-elephants that contained males and females. Species do not arise from single new mutant males that then have to find a corresponding mutant female — they arise by the diffusion of variation through a whole population, male and female."


Sure. Not easy to follow, but the idea is right there. Since mice, for example, evolved from populations of already sexually compatible individuals it is nonsense to ask how "the first male mouse" found his "first female mouse." Not just nonsense. Abject stupidity. I truly don't understand why this point escapes creationists. We hear that and we immediately notice the stupidity.

The question as to how sex evolved is a different thing. Notice that's not what Ray is asking/saying. Ray is implying that it's individuals that evolve (the "newly evolved male mouse" "finds" the "newly evolved female mouse"), that their sexual compatibility has to evolve anew each time (how does the newly evolved male mouse find the appropriately newly evolved female mouse), and that each gender evolves independently (you don't notice the stupidity? seriously?). That's what makes Ray's assertions/rhetorical questions into unsurmountable imbecility. He's not asking how did sex evolve. He's asking how these newly evolved male individuals find their corresponding newly evolved female partners, implying so much obvious nonsense that it's painful to see creationists falling for it.

Once you guys start getting it, you don't just notice how stupid that was, you move the meaning of that crap (or move the goal posts) into how did sex evolve (often "evolutionists" are kind and/or stupid enough to translate the nonsense into the question about how sex evolved themselves, but I'm never that kind when it comes to creationist bullshit).

Now, how sex evolved, of course, interesting question. Lots of ways to start answering it. But the astounding stupidity of that original question/assertion doesn't go away.

Ed said...

Nah judmarc, all was lost after this master piece from the judge, although
I will admit, it's the uppercase ENGINEERING which triggered our demise.

"I won't go into the details, so here's my short, but nevertheless effective answer - just look at everything from an ENGINEERING perspective :all these interdependent parts, the parts themselves also being very tough to alter without breaking, plus the fact that these organisms an existential need for a lot of functions at once. "

Gish gallop anyone?

Tom Mueller said...

Hi Larry

Re:
How much of the total divergence do you think is due to adaptation (natural selection) and not random genetic drift of neutral alleles?

First of all, I really see little divergence at all between Chimps & Humans.

http://m.naturalnews.com/news/036980_genius_chimpanzee_intelligence.html

Keep in mind that 90% of the genome is junk and the total divergence is close to the population genetics value expected from drift of neutral alleles (molecular clock).

I think you missed my point


HARs were found & identified precisely because they represented clusters of mutations that could not be explained by random background mutation - especially so since they are invariably found in NON junk regions such as crucial enhancers.

If HARs indeed do represent THE crucial speciation events between Human & Chimp lineages, the importance of evodevo has been vindicated and the relative importance of Neutral Theory has been diminished

Petrushka said...

"First of all, I really see little divergence at all between Chimps & Humans.

http://m.naturalnews.com/news/036980_genius_chimpanzee_intelligence.html"

That needs a Trigger Warning.

Jass said...

Morons,

The genes the you inherited from your mother and your father on average and at the maximum are about 93% similar. Sequence your parents genomes and your own and you will see...

This means that if the chimp and human similarity at 98-99% is correct, you are more closely related to a chimp than to yourself you imbeciles.

I really enjoy amusing, so stories told by confused Darwinists... They make me laugh to tears lol

judmarc said...

on average and at the maximum

'Cause these two things are identical, yeah. You must've been nearly as good at math as English.

Dazz said...

[i]on average and at the maximum[/i]

ROTFLMAO!

Under creationist logic the minimum is probably higher than both, which is strong evidence that miracles happen all the time!

Ed said...

According to lawyer kev's logic, forensic DNA typing cannot work. I wonder how many people have been killed or sent to prison on long term sentences based on DNA evidence? How can it work if there's already a 7% difference between direct descendants?

Hey kev were does 7% come from? Because current scientific literature states there's about 100 mutations on average difference between parents and children DNA. Or is 7% needed for hyper evolution which creationists believe in if the earth is only 6K years old?

Chris B said...

Too late, I'm already offended.

Chris B said...

No, KevNick,

On average, people are about 99.5-99.9% similar at the nucleotide level.

doi:10.1038/ng1435
doi:10.1038/ng1438

Keep loling, tho, it's all you've got.

Tom Mueller said...

Hi Larry,

My last reply was done by cell phone explaining why it did not track this conversation. I apologize

In the meantime, I did a quick google-whack and stumbled acros this neat PLoS paper.

check it out:
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0032877

substitutions in HARs tend to have come to fixation faster than substitutions elsewhere in the genome and that substitutions in HARs tend to cluster in time, consistent with an episodic rather than a clock-like process underlying HAR evolution.

Anonymous said...

Piotr,
Quite appropriate. Would be nice if somebody had the time to do the experiment with some text. I bet that different editions of the origin of species would come out as different books.

Piotr Gąsiorowski said...

Yep, and I'm not joking. In Tomkins's own words,

Given that the genome-wide analyses required a large amount of cumulative and comparative data, only the top alignment for each database hit (if it existed) was returned. Gapping was disallowed for a variety of reasons.

In my book example, the top alignment would be about two pages out of 200 (assuming a uniform distribution of "indels").

Aceofspades said...

Thomkins' methods for assesing genome similarity are a complete joke.

He made the mistake of releasing a paper in the creationist journal ARJ back in 2014 where he was trying to discredit the evidence for common descent from the GULO pseudogene which we share with other apes. Here is the paper. It is littered with errors but the two things that immediately jumped out at me were his claims that "The 28,800 base human GULO region is only 84% and 87% identical compared to chimpanzee and gorilla, respectively. The 13,000 bases preceding the human GULO gene, which corresponds to the putative area of loss for at least two major exons, is only 68% and 73% identical to chimpanzee and gorilla, respectively".

It is very easy to obtain these seqiuences and fact check these claims. I did that here. I have also provided the sequences so that you can also fact check Jeffry's claims yourself. It turns out that Jeffrey was either lying or grossly incompetent. It is a simple matter to align these sequences and count up the differences. In the 28,800bp region, we are actually 98% identical to both Chimps and Gorillas - the same is true for the 13,000bp region as well.

Jeffrey's mutation count would have had to have been an astonishing 7x the actual number!

Jeffrey has read my challenge and has so far responded dismissively, refusing to look into the glaring errors that he presents in his paper. When one has been made aware of glaring errors in their work, when they refuse to correct those errors and when they continue to parrot their claims regardless - at that point I can only be left to conclude that they are knowingly lying.

Aceofspades said...

Regarding his claims about the chromosome 2 fusion - I've addressed those here as well.

Diogenes said...

I would like to shift the discussion from Tomkins' bogus 70% similarity to the claim that the human chromosome 2 fusion isn't real. The most common creationist counter-argument is that a single gene spans the fusion point, therefore the fusion is functional, therefore it can't be a fusion.

Maybe we should start by reading AceofSpades' link above.

A while back Kenneth Miller disputed that a gene crossed the point of fusion. The creationists claimed that one database (ENSEMBL?) did list a gene as extending over the point of fusion, and they alleged that Miller had conceded the point in a private email to them.

What are the prospects of us dragging Ken Miller into this discussion?

Diogenes said...

AceofSpades, thanks for showing up. The gang's all here now. I'll read your link on chromosome 2 fusion.

Diogenes said...

Glenn, I can open your PDF in my laptop. So I'll read your paper ASAP.

Aceofspades said...

Actually, I would appreciate anybody's views on the transcript that spans the fusion site.

Here are the transcripts for this gene - note that only transcripts labelled b and e span the fusion site which lies within the intron which has been coloured green.

Aceofspades said...

Here is the so called "gene" that spans the fusion site on chromosome 2. Jeffrey claims it is functional - we don't actually know whether or not it is functional but the low levels of transcription seems to indivate that it more than likely is not functional. I have pointed this out to Ian Juby in the past - I guess he didn't think it was worth asking Jeffrey about this - the questions and answers here seem to be well rehearsed.

You will notice that only two of the 5 different transcripts for this "gene" span the fusion site (transcripts b and e). The fusion site exists within the green coloured intron in the link above (length 2165bp)

Perhaps somebody could help me better understand this data, but it appears that only a single transcript has ever been detected for version b and it was found in bone marrow.

For version e, 2 transcripts have been found in the prostate, 1 in the bladder, 1 in breast tissue (matching multiple genes) and 1 in schizophrenic brain S-11 frontal lobe. These numbers strike me as incredibly low for a gene that is supposed to be functional. I'm no expert, but I'm going to guess that these are spurious transcripts of no significance.

So much for Ian Juby playing up the concept that this is a highly critical gene.

Glenn said...

Hi Ace,

I thought the transcript that crosses the fusion site (are there 2??) was found in the prostate: NR_024004.1

Also, in the AceView expression table, how do we know which transcript we are detecting? Is it possible that it is measure transcripts that do not cross the fusion site?

Glenn said...

Yeah Ken Miller made a mistake on that - there is a sequenced transcript that crosses the fusion site It appeared at NCBI but not in Ensembl.

Ace: you say there are two transcripts that cross the fusion site - is one of them purely theoretical? Has it been sequenced? Or is it the product of a genebuild?

Aceofspades said...

Hi Glenn

Here is a screen grab from the UCSC Genome Browser which I captured a while ago. It shows one transcript which crosses the fusion site.

According to AceView, there are two. See the compact gene diagram on this page. The fusion site lies within the first intron of detected transcripts b and e. The only thing that seems to differentiate b and e is the slightly different length of the second intron.

If you click on b it shows that it has been detected once in bone marrow - Accession: DN999053. Yes.. It looks like it has been sequenced

There have only been two detections of the other transcript e which actually include the exon on the far side of the fusion site and they were both found in prostate tissue. The other detections don't actually span the fusion site. Accession: AK123946 (I can't give you the seconf accession number right now because the site is down)

Aceofspades said...

When you click on a transcript, it only shows the sequences which have been found to match it in the expression table. Note that most of these detected sequences don't actually cross the fusion site. If they did, they would have to contain data from exon 1 (positions 1 -> 122).

Here is the sequence for exon 1:

tttgcgagggcggagttgcgttctctttagcacacagccggagagcatcgcgagggcggagctgcgttctcctctgcacagacttcggggctattgcgaaggcggagcagagttcttctcag

Aceofspades said...

Interestingly... if you run a blat search for that first exon on the far side of the fusion site...

It appears to be a tiny piece of satellite DNA from the centromere on Chimp chromosome 2A. Coincidence?

Glenn said...

Ace, is there a way to breakdown the expression profiles by transcript? So for example, can we show the relative frequency of transcripts that cross the fusion site as opposed to the canonical transcript?

Glenn said...

Ok, answered my own question. If you hover the mouse over the exon, it will show you the RNA-seq reads.

Glenn said...

*intron.

Piotr Gąsiorowski said...

Yes, pure coincidence, like that intron which spans the alleged fusion site and just happens to contain pretelomeric, telomeric, inverse telomeric and inverse pretelomeric sequences, in that order.

Aceofspades said...

Glenn, I'm not a geneticist.

Could you explain to me what these numbers mean and what their implications are regarding the functionality or non-functionality of these sequences?

Glenn said...

Haha! Neither am I :D

The number of RNA-seq reads would be related to the expression data: so for the green intron, it was sequenced 687 times, while the "main" intron was sequenced 2,601 times, and a similar intron 599 times (total 3,200 times).

So, very roughly, transcripts that cross the fusion site make up only about 20% of the expressed transcripts.

I think :)

Glenn said...

So, to put the "functionality" into perspective:

First of all, it is a pseudogene - it doesn't actually code for an RNA helicase. Secondly it is one of a family of 18 (putative) similar pseudogenes, and thirdly these transcripts that cross the fusion site make up only about 20% of the transcripts of this one pseudogene.

So, there is some evidence that non-coding RNAs do have some regulatory function, but in this case, if we took away transcripts that cross the fusion site, we'd lose 20% of an 18th of the total regulatory work that the pseudogene might do.

Glenn said...

I could probably even go a step further and say that DDX11 (which is an RNA helicase) is part of a larger family of DDX genes. And each of these genes obviously produce a protein, but they also have multiple non-coding transcripts of their own.

Transcripts of DDX11L2 that cross the fusion site become rather insignificant when you see how far down the chain of importance they are.

Mikkel Rumraket Rasmussen said...

" not saying that VARIATION doesn't occur. I'm just saying that the mechanisms for these undirected changes taking place are not potent enough to make some of the proposed transitions."

Blind mindless claim and as such can be dismissed. The very fact of multiple nesting hierarchies in comparative anatomy, biochemistry and phylogenetics testifies to the falsity of that claim. A few centuries of selective breeding shows organisms usually have remarkable plasticity for change. In contrast creationsts have never been able to show any hard limit, they just mindlessly assert there must be one without any evidence.

"//Yes, evolve from one gen to the next within certain LIMITS (Lenski's experiments show that .60,00 generations, for crying out loud)."
What limits? You make the limit up.

"//Assertion, not proven."
I prove it in the very post you respond to you illiterate nutbag. Get off the Jesus weed for a moment and start paying attention to reality instead of your fevered halluscinations with magical creator fairies.

"/Within certain LIMITS."
The only limitation is time. Human beings only live so long, there's a limit to how much we can show in a single lifetime. That constraint isn't operating on nature as a whole.

Before your insignificant creationist brain explodes into a "MILLIONS OF YEARS IS AN EXCUSE" drooling rage, watch this you little kid: Micro vs Macro growth.

That's it, that's all your arguments annihilated in that single video.

Mikkel Rumraket Rasmussen said...

"//Yes! Except that the observed nested hierarchies are certainly limiting and don't give you the right to say that this mechanism was capable of producing the other alleged hierarchies!"
They are not limiting at all, they can be used to show the interrelatedness of all of life. Further supported by the pretty much universal genetic code, the similarities in the molecular machinery and so on confirming the same hierarchies.

"//A lie. I see no coherent pattern."
Not a lie, a demonstrable fact. What you personally see or not is immaterial with respect to the fact.

"Also, how about CONVERGENCE? I.e. similarities being there where they SHOULDN'T be?"
There is no place where there "shouldn't" be convergence. Who the fuck decides what should or shouldn't converge? Nobody. Natural selection is a pretty strong mechanism for generating convergent structures if the selective pressures are the same and the genetic and morphological substrates are sufficiently similar. We all live in the same world governed by the same laws of physics, we would expect a certain amount of convergence on that basis alone.

"Graph it and compare the graph to the common descent graph. Don't tell me how it doesn't turn common descent's entire head around."
It doesn't turn common descent's entire head around. For the reasons stated.

"A common designer notion doesn't contradict itself."
Everything in the universe can be rationalized in ad-hoc fashion when you include an omnipotent designer with no contraints. Literally no observation could falsify such a designer, you could always just rationalize that "the designer wanted to make it like this for reason X".

"//Another lie. For SOME organisms grouped within nested hierarchies, YES."
Not a lie, another demonstrable fact.

"But what if what I've shown is correct? What would comparative anatomy say if we grouped the cat and the human in the same nested hierarchy?"
We already do, they both belong to mammalia.

Mikkel Rumraket Rasmussen said...

"//In your looney , deluded dreams"
No that's another fact. It really doesn't take much effort to argue like you do, just blanketly deny everything. Well done, showed us all here you did! Bravo!

"//That's right. They have to envelop the organisms with stories, fictions."
If you say so my child.

"// A fictional and incoherent one."
These answers of yours are only worth responding to as a form of entertainment. They're not arguments, references or demonstrations, they're just your childish commentary. You might as well have just wrote "nyah-nyah and your mum spells like poopoo". What are you, a 12yo?

Confused said...

"The genes the you inherited from your mother and your father on average and at the maximum are about 93% similar. Sequence your parents genomes and your own and you will see."

What on Earth are you talking about? Me and my parents have identical genes.

Faizal Ali said...

@photosynthesis

You're in good company. I often start giving them the benefit of the doubt, only to be disappointed again.

What's really funny is, just after he confirms that my interpretation of his argument was incorrect, he then says it was correct. It seems he doesn't even know what he's saying.

Glenn said...

"Literally no observation could falsify such a designer, you could always just rationalize that 'the designer wanted to make it like this for reason X'."

I'd love to hear any "reason X" for why this designer put functionally equivalent sequences into different species, but when those sequences are plotted into a phylogenetic tree, they magically match trees constructed from independent data! What are the odds! :D

Tom Mueller said...

Uhmmm for the record

There are incredibly stupid IDiots and frustratingly obtuse IDiots

We are wasting much unnecessary effort on the former and ignoring the subtler rebuttals of the later

If one focuses attention on HARs then in fact it is true that

1 crucial regions of genome between chimps and humans differ by more than 30%

2 and such differences cannot be explained by the inexhorable ticking of a molecular clock as suggested by Neutral Theory leaving open the door for ID

Just saying...

Anonymous said...

"Isn't it a bigger problem for evolutionists if chimp's genome is 98%+ identical and yet the body plans, the brain, learning capacity ect. are not?"

WHICH genes differ is important. Chimp and human seem to differ greatly in the timing of development, which leads to many of the observed differences in their bodies. These differences are due to differences in the genes that control the processes of development.

By the way, the < 2% genetic difference between chimps and humans includes differences in genes that control development, differences in genes having other effects, and differences in structural DNA and non-functional DNA.

Unknown said...

I think a more important point concerns the likelihood that the transcript that crosses the fusion site arose after the fact. This is what the creationists would dispute. If there was recognizable DDX-like sequence on either side of the repeats, this would give the appearance that the gene was there prior to the telomeric sequences. I did a BLAST search of this region some time ago and did not identify any other DDX-related sequences on the far side of the telomeric repeats, supporting the conclusion that the fusion predates the transcript. Moreover, the DDX-like genes all have a similar size, structure and sequences across their exons, as shown Costa et al., 2009, the paper that describes the DDX11L gene. This reveals that the entirety of the recognizable DDX-like sequence resides on one side of the fusion site. Finally, the Costa paper concludes that the family of DDX-like pseudogenes was propagated to many sub-telomeric locations, lending further support to the conclusion that this region was previously a telomere.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2705379/

Diogenes said...

David Levin joins us. If you don't know, David is an important geneticist and experienced anti-creationist well known for his knock down, drag out fight with creationist Jonathan Sarfati in an Amazon thread that ran to 10,000 comments.

Our friend Christine Janis, a comparative anatomist, was there too.

the Costa paper concludes that the family of DDX-like pseudogenes was propagated to many sub-telomeric locations, lending further support to the conclusion that this region was previously a telomere.


Can you explain the importance of "propagated to many sub-telomeric locations"?

christine janis said...

Diogenes

We're still alive and kicking (mainly me, but also David occasionally) on the Darwin's Doubt discussions, where we're joined by invertebrate zoologist/marine biologist Aaron Baldwin.

Diogenes said...

I've emailed Kenneth Miller and asked him to join in here.

Aceofspades said...

Thanks for the information David. You've addressed the one side of this transcript (the side containing the DDX-like sequence) but it is also worth addressing the other side of this transcript as well (exon1 - on the far side of the fusion site)

As the annotation shows in this diagram, the exon lies within a piece of satellite DNA labelled "Repeat TAR1, family telo".

If you run a BLAT search for this satellite sequence, you will find that it is a pretelomeric sequence with paralogous sequences found on the ends of chromosomes 1, 2, 4, 8, 10, 13, 19 (both ends), 21 and 22.

If this is in fact a functional transcript that has always required this exon, it would seem to be a remarkable coincidence that this exon is in fact part of a larger sequence that apart from this one exception is found exclusively on the ends of chromosomes.

Diogenes said...

Great catch guys.

Unknown said...

Diogenes,

Thanks for the kind words.

"Can you explain the importance of "propagated to many sub-telomeric locations"?"

To understand this, it is important to recognize that telomeric and sub-telomeric regions are highly homologous and are sometimes used to repair the ends of chromosomes through homologous recombination. When this happens, the end of one chromosome is transferred to the end of another chromosome along with genes that reside within the sub-telomeric region. This has happened with human chromosomes, resulting in the proliferation of members of this family of expressed pseudogenes near the ends of 16-19 human chromosomes (there is some individual variation). The various copies of the DDX11L family all reside in the same location next to members of the WASH gene family and are in the same orientation relative to the centromere, supporting the idea that they arose through transfer of non-homologous chromosome ends. The internal copy of DDX11L at the chromosome 2 fusion site suggests that this region was once an active telomere.

Unknown said...

Ace: "it is also worth addressing the other side of this transcript as well (exon1 - on the far side of the fusion site"

Thanks for this. The most reasonable inference from all of this is that the transcript that reads through the fusion site either represents a new transcription start site, or an old one that was associated with a gene that was truncated by the fusion event, thus producing a chimeric transcript.

Unknown said...

In any event, what we see are two different telomere-specific signatures on either side of the fusion site. This region veritably screams "telomere".

Diogenes said...

Glenn, can you explain how we know DDX11L2 is a pseudogene?

Diogenes said...

Glenn, I read your paper. It's great. But my one complain is, you don't verbally dispute Tomkins' claim that his similarity is much smaller than the conventional similarity because "secular scientists" are excluding all the inconvenient DNA sequence, while Tomkins is including all that sequence. This claim-- the "evolutionists" are ignoring most of the DNA data, their 98.4% figure is only calculated from small bits they choose to look at, while the creationists look at all the data-- is endlessly repeated by the creationists.

However, your work shows clearly that Tomkins' explanation is bogus-- we know why Tomkins got a lower similarity than anyone else, and it was not because he was including data that evolutionists were excluding. You do a good job challenging Tomkins' number, but you don't challenge Tomkins' explanation for why he got the number he did. For that I think you need to quote Tomkins at length, and stick it to him.

Glenn said...

Glenn, can you explain how we know DDX11L2 is a pseudogene?

To the best of anyone's knowledge, DDX11L2 doesn't code for a protein. To function as an RNA helicase you need an ATP processing domain so that the enzyme can do some work, and a helicase domain that does the unzipping. According to Uniprot entry for the DDX11 gene, these two functions are coded for in the first 400 or so amino acids (http://www.uniprot.org/uniprot/Q96FC9#showFeatures), while the "best predicted protein" for DDX11L2 on AceView is only 127 amino acids long - and is missing the "DEAD" or "DEAH" motifs (which is where the protein family gets its name - "Dead Box Helicase").

Again, I'm not a geneticist - this is only what I've learned from other geneticists!

Glenn said...

You do a good job challenging Tomkins' number, but you don't challenge Tomkins' explanation for why he got the number he did. For that I think you need to quote Tomkins at length, and stick it to him.

Yup, completely understand the criticism, and the first draft I made of the paper actually included all the rebuttals of the secular papers that he cites - and boy howdy are there some pants on fire in there. The problem was that the paper became rather bulky, and the key result of the paper kinda got lost in a huge amount of discussion.

So I basically made an executive decision to narrow the scope down to just the comparison. The papers that he "re-evaluates", and the two creationist ones can be dealt with in a second paper (assuming this first one ever sees the light of day). Although, if the editor gives me the chance, I will probably add in a little paragraph about that - just saying that I didn't exclude anything, so Tomkins would need to "re-evaluate his re-evaluations".

nmanning said...

It is so cute when creationist want to bring up Haldane's dilemma, as they so often do not have a clue as to what the 'dilemma' even was supposed to be - or the fact that it was not a universally accepted issue:

"Perhaps the only disagreement I ever had with Crow concerned the substitutional load, because I never thought that the calculations concerning this load, which he and others carried out, were appropriate. From the very start, my own calculations suggested to me that Haldane’s arguments were misguided and indeed erroneous, and that there is no practical upper limit to the rate at which substitutions can occur under Darwinian natural selection."
- Warren Ewens

nmanning said...

I would love to - alas, I have been unable to find the graph they sent me, but I will keep looking!

Anonymous said...

Tom,

1 crucial regions of genome between chimps and humans differ by more than 30%


Regions of the human and chimps genomes might differ by more than 30%. That doesn't mean that they are "crucial."

2 and such differences cannot be explained by the inexhorable ticking of a molecular clock as suggested by Neutral Theory leaving open the door for ID

Oh, but you're jumping here. Random mutation doesn't mean equidistant and homogeneous mutation. It just means random mutation. Some time ago, people thought they had found "mutational hot spots," and a scientist with a bit better understanding of what random means showed that these "hot spots" did not occur at a higher frequency than expected from a random distribution of mutations. So there was no such thing as a mutational hot-spot. There might be, but finding them is a tad more difficult that just pointing to a region that has more mutations that other regions, since the distribution of mutations should actually vary.

Then, the molecular clock doesn't have to explain everything. When conceptualizing of the clock most scientists think of point mutations. Rearrangements, DNA insertions, retroviruses, etc, are not point mutations, yet they are just as natural. Etc. Etc.

So, no. Concentrated mutations don't just open the door for an intelligent designer. Now, more deeply, if we knew nothing about what random actually means, about population bottlenecks, about insertion elements, about etc and etc, that would still not leave the door open for an intelligent designer because, well, intelligent designers have a track record for being imaginary answers to open questions. Like volcanoes and thunder. Not knowing why there's a concentrated amount of mutations somewhere in the genome doesn't mean that some "intelligent designer" is the answer. It would just mean that we don't know how that happened.

Have a great week!

Aceofspades said...

The thing about HAR regions is that they are rare and incredibly small. They have on the order of about 100 base pairs each.

Even if these regions were under intense selective pressure, they represent such a tiny fraction of our genome that I don't see how they challenge neutral theory.

Tom Mueller said...

Hi Photosynthesis

Re:
Regions of the human and chimps genomes might differ by more than 30%. That doesn't mean that they are "crucial."

Agreed – such regions do not have to be crucial; any such contention would admittedly be contingent and not necessary.

That said, it appears that Human Accelerated Regions do indeed represent both mutational “hotspots” not due to regions of genomic instability but rather positive selection in functional regions of the genome. In fact, that is exactly how these regions were discovered (i.e hypermutability) AND it turns out these HARs are crucial in explaining human-chimp divergence.

Re:
So there was no such thing as a mutational hot-spot.

To my understanding there in fact are mutational hotspots for two reasons:
1 - Highly unstable regions of the genome
2 - Genomic regions subject to intense positive selection

For example, I direct your attention to

http://www.sciencemag.org/content/314/5800/786.abstract

Re:
the molecular clock doesn't have to explain everything.

Agreed! That was my whole point, the inexorable ticking of a molecular clock cannot explain HARs.

Re:
Concentrated mutations don't just open the door for an intelligent designer…Not knowing why there's a concentrated amount of mutations somewhere in the genome doesn't mean that some "intelligent designer" is the answer. It would just mean that we don't know how that happened.

Hmmm – but I direct your attention to Larry’s challenge:

Laurence Moran:”The onus is now on Intelligent Design Creationists to prove that the creation of an irreducibly complex system requires a step that cannot possibly be due to natural selection (or random genetic drift?)…

I don't know of a single example of such an irreducibly complex system.


And I counter that some of Dembski/Behe’s acolites are indeed citing HARs as proof positive that ID must be correct because of a false syllogism as outlined above. The major premise of Neutral Theory and the minor premise of HARs elicit the inevitable conclusion that Larry’s version of Evolution must be false.

Of course, the rebuttal is trivial: As Aceofspades explains, Neutral Theory does not deny any and all occurrences of intense positive selection; adaptation can be invoked to explain mutational hotspots such as HARs.

The champions of ID counter that such mutational clusters would have required fixation within a generation for divergence to proceed thereby invalidating evolution even at the level of intense positive selection. Too much/too fast dontchya know.

Here is a worksheet I provide my highschool students on the very subject:
http://www.indiana.edu/~ensiweb/lessons/Hum-Chimp%20DNA.pdf

I built it before I became a regular at sandwalk. Rereading it permits me to recognize some errors of omission and commission that require “tweaking”. This activity's next incarnation will focus on this article instead:

http://phenomena.nationalgeographic.com/2008/09/04/did-a-gene-enhancer-humanise-our-thumbs/

John Harshman said...

I think you are confusing mutation with fixation. Nobody claims that HARs are mutational hotspots, as far as I know. Strong positive selection doesn't increase the mutation rate; it increases the fixation rate.

Anonymous said...

Strong positive selection doesn't increase the mutation rate; it increases the fixation rate.

And so does population bottlenecks.

Tom, you also seem to mistake "random genetic drift" with one of its potential consequences, namely the potential for a molecular clock.

Anonymous said...

Tom,

Re:
So there was no such thing as a mutational hot-spot.

To my understanding there in fact are mutational hotspots for two reasons:
1 - Highly unstable regions of the genome
2 - Genomic regions subject to intense positive selection


I did not actually mean to say that mutations hot spots don't exist. I said that in that instance there was no such thing. I added that to find actual mutational hot-spots you have to do more than assume that concentrated mutations meant hot-spot.

Tom Mueller said...

Hi John - Hi Photosynthesis

Re: the confusion between mutation vs fixation rates.

Please do not attack the messenger. I am merely reporting the sophistry of the IDiots. Frankly I reckon too much band width has been wasted on their sillier strawman arguments and not enough effort on some of their more subtle albeit still spurious suggestions.

That said, I do welcome any efforts on your part to correct any lingering naïveté on my part. I thank you both.

It could be argued that biased gene conversion increasing the GC content of the genome may be responsible for any HACSN1 –driven human chimp divergence.

Check out this intriguing paper:

Hotspots of Biased Nucleotide Substitutions in Human Genes

http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1000026

I think IDiots miss the crucial point.

I checked the original 81 nucleotide sequence of HACSN1; 12 of the 13 of the total base pair variations are AT-to-GC “biased” substitutions. But hey, that is still a cluster 4 X faster than would be anticipated by "neutral substitutions" as compared to the rest of the genome... bringing us back to your distinction between mutation vs. fixation rates.

In other words- the IDiots have been summarily contradicted and further rebuttal on their part is futile.

These findings indicate that a recombination-associated process, such as biased gene conversion (BGC), is driving fixation of GC alleles in the human genome. This process can lead to accelerated evolution in coding sequences and excess amino acid replacement substitutions, thereby generating significant results for tests of positive selection.

Anonymous said...

Hi Tom,

I understand. Notice that I did not insult you (I hope I didn't). I just noticed that there were some problems with mixed understandings and misunderstandings, so I pointed them out. I often use my own concepts loosely. But when dealing with creationists we have to be careful and not let us go by mistaken notions, but rather try and understand what those notions are about. Concepts help understand, but they should not be mistaken with the things those concepts are trying to describe. With this distinction it becomes easier to explain when we misspeak and focus on the phenomena, better than playing word games. Remember that creationists are all about rhetoric, and a good deal of sophistry consists on using equivocations. Sometimes subtle ones, sometimes obvious ones.

For some reason, equivocation fallacies seem hard for creationists to visualize and understand (or they're just dishonest and don't want to admit those equivocations).

Tom Mueller said...

Hello Photosynthesis

Notice that I did not insult you (I hope I didn't). I just noticed that there were some problems with mixed understandings and misunderstandings, so I pointed them out. I often use my own concepts loosely.

Not at all – I appreciate your patience and your indulgence for bringing me, an aging and non-current high school teacher, up to speed. Again, I thank you and John both for correcting any misconceptions on my part.

I took up pencil and paper and was about to do some rough calculations to contradict the IDiot rendition I just cited above, when it suddenly occurred to me that recombination in diploids greatly complicates the story. I then did some google-whacking and was chagrined that my efforts had already been better tackled elsewhere:

I direct your attention to Allan Miller’s insightful posts on

http://sandwalk.blogspot.ca/2012/01/understanding-mutation-rates-and.html

I am very grateful for this opportunity to refocus my thoughts and rewrite my exercises for the benefit of my students.

Basically, I want to explain to my students that the point mutation rate is astonishingly high (~100 mutations per generation) and that in the ~600 000 generations since our human lineage diverged from chimpanzees, the number of “collective mutation hits” in a population averaging ~1 000 000 would exceed the size of our genome, ergo the importance of distinguishing mutation rate from fixation rate for a variety of complicated reasons including Muller’s ratchet.

Going back to Muller’s ratchet, I reckon off hand that the number of mutation hits per human chromosome is roughly equal to the number of chiasmata per human chromosome. Is this just a coincidence or does this ratio hold across species and is less than coincidental from a Mullerian ratchet POV? It’s late at night and I am thinking out loud, I hope I am not embarrassing myself in public.

Meanwhile, given an already astonishingly high mutation rate that may in fact be highly variable (Susan Rosenberg’s findings jump to mind) and that for every recombination event that diminishes “mutational load” half of the reciprocal products are in fact accumulating “mutational load”; IDiot contentions that Evolution could not produce those highly divergent HARs (as cited above) is not even wrong any more (to paraphrase Wolfgang Pauli)

Ken Miller said...

I’m happy to be dragged into the discussion!

I exchanged a couple of emails with Tomkins last year, pointing out the errors in his papers (in creationist journals) on the Chromosome 2 fusion site. As you all know, he’s very excited that he’s found a “highly-expressed” gene that “spans” the fusion site. This means, he claims, that the fusion site couldn’t possibly be what it seems.

He’s wrong, and many of the reasons why have already been pointed out in this discussion. First and foremost, the gene in question (actually a pseudogene with no known function) is a member of a transcript family known as DDX11L. Tomkins pointedly ignores the Costa (2009) paper, which identified 18 members of this gene family. Each of them is next to a sequence known as “WASH,” which is transcribed in the opposite direction of the DDX11L pseudogene. And, more to the point, each and every one of them is located right next to a telomere – except for one. That’s the DDX11L2 sequence, which is parked right next to the fusion site. That alone is very strong evidence that site is exactly what it seems to be – the remnant of a telomere-to-telomere fusion.

Most of the genome databases show the DDX11L2 sequence as off to one side of the fusion site, so it really doesn’t span it. However, in some of the databases there are transcript variants that include the head-to-head telomere sequence motifs as one of the introns in the primary transcript. That is basis on which Tomkins claims that the gene spans the site. But the very same data are easily explained by variability in the termination of transcription so that occasionally a somewhat longer RNA is produced. This is exactly what I pointed out to Tomkins…. So that any claim that I “admitted” he was right about his interpretation is bogus.

And, if Larry permits me to rant on a bit:

• Tomkins said that “some” chromosome banding patterns were similar between humans and other great apes. Some? The matches were so extensive that Yunis & Prakash (1982) were able to align each and every chromosome from four different species!

• He said there were too few telomere repeats in the fusion site. A “pristine” site would have 20,000 – 30,000 bases. But the fact is that “pristine” telomeres would prevent fusion, and treatments that dramatically shorten telomeres actually cause fusion, which is why there are so few repeats in chromosome 2. The small number of telomere repeats is exactly what should be expected at a fusion site.

• Tomkins said that the DDX11L2 gene was unique to humans. Wrong. As Costa (2009) showed, there are members of this family in chimpanzees and gorillas. And, as you might expect, they are located right next to telomeres.

• Finally, Tomkins lays great emphasis on the observation that transcription factor binding has been found throughout this region. But simple binding says nothing about the specificity of binding or its biological importance.

• Tomkins has constructed a straw man in which an authentic fusion site would have to be an exact replica of the ends of two present day chimpanzee chromosomes to be valid. What he does not seem to realize is that the fusion occurred millions of years after our lineage separated from chimpanzees, and that both lines have continued to evolve along separate pathways. That accounts for the differences he regards as so significant.

The evidence for a chromosomal fusion in the ancestry of our species is on very solid ground, and has been greatly strengthened by the very research he wishes to use against this idea.

Glenn said...

Hi Ken, thanks for your comments!

I also want to chime in on the "highly expressed" claim, which I can't help but think is a blatant lie. DDX11L2 is nowhere near as highly expressed as DDX11 itself, and secondly, the transcripts that cross the fusion site only make up about 20% of the DDX11L2 transcripts. It's demonstrably "lowly expressed" :D

Can you explain the evidence that the fusion occurred well after the chimpanzee split? I actually thought the fusion might have been one of the triggers for speciation ..

Ken Miller said...

Glenn - I'm pretty sure I've run across a study or two dating the fusion event to the last 2 or 3 million years. If I remember correctly, the fusion site is present in Neanderthal and Denisovan DNA. I can check this out if you like.

Gary Gaulin said...

3 million eh? The date works for me!

https://whyevolutionistrue.wordpress.com/2011/09/18/how-big-was-the-human-population-bottleneck-not-anything-close-to-2/#comment-135151

Thanks for the good news Ken.

Diogenes said...

Ken,
Thanks for joining the discussion. This thread is like a Christmas where I got everything on my list.

Did Tompkins really say DDX11L2 was "highly expressed"? Seems a bit brazen.

I don't know how you keep your cool arguing with these people.

Joe Felsenstein said...

If there was a fusion event 3 million years ago, that does not mean that the human-ancestor population at that time was as small as 2. It means that the ancestry for the fused region was a single chromosome. But because of recombination, all other regions of the genome were most likely descended from other individuals at that time.

That is widely misunderstood for mitochondria and Y-chromosomes (the other chromosomes come from ancestors other than Mitochondrial Eve and Y-chomosome Adam, in fact each region of the genome comes from a different ancestor, down to chunks the size of a gene). It is also true for the fusion region versus nearby regions of chromosome 2.

AllanMiller said...

Interesting that the corrected alignment returns towards the values given by the now-superseded hybridisation method. The 'gapping' parameter is particularly relevant to this. Complementary pairing will rapidly realign either side of an indel in a physical system, affecting the 'melting' temperature, which an ungapped search would miss.

Tom Mueller said...

Hi Allan

Your ears must have been ringing...

http://tinyurl.com/ofeyjd8

best

Faizal Ali said...

My understanding, though, is that the fusion could have played a role in speciation, in that it promoted a relative reproductive isolation among those with the fusion, who would be more likely to successfully reproduce with each other than with those who did not have the fusion . Am I correct on that?

AllanMiller said...

Glenn,

Can you explain the evidence that the fusion occurred well after the chimpanzee split? I actually thought the fusion might have been one of the triggers for speciation

I don't think there is a strong reason to associate such a translocation with speciation. True single biological species can be heterozygous for a fusion/break without any apparent phenotypic or meiotic fitness effect. eg http://www.nature.com/hdy/journal/v50/n3/abs/hdy198332a.html.

More generally, karyotype evolution on a broader scale appears to follow a regular switch in the polarity of female meiosis - http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1461872/pdf/11729161.pdf . Although the authors' first sentence is "Speciation is often accompanied by changes in chromosomal number or form even though such changes significantly reduce the fertility of hybrid intermediates", the data they report actually argues against hybrid-unfitness speciation being the typical consequence of translocation.

John Harshman said...

If I remember correctly, the dating of the fusion comes from counting substitutions in the inactive former telomeric repeats.

AllanMiller said...

Hi Tom - thanks!

Aceofspades said...

John, surely this would be misleading? Wouldn't many of these substitutions have happened as part of the same process that lead to the shortening of these telomeres in an individual prior to the fusion event?

According to Carl Zimmer in this article:

> So the small quantity of telomere DNA does not, in fact, raise grievous doubts about the evolution of fused chromosomes. Nor does the fact that the repeating DNA in the fused chromsome has mutations in it. Telomere DNA is just prone to mutation. In fact, if you look at the telomere on a chromosome, you’ll typically find that the newest pieces of repeating DNA are correct, but the older segments further from the loop’s end are slightly garbled. These errors arise in your own body. If a chromosome’s telomeres are damaged, you might well expect the new ones to be gone, and the garbled ones remain.

Unknown said...

The fusion apparently took place prior to the split of modern humans from Denisovans and Neanderthals. It is mentioned on pp. 64-65 of the supplementary information file for Prufer et al’s January 2014 paper in Nature on the Altai Neanderthal genome sequence. I believe it is still availbale here:
http://www.nature.com/nature/journal/v505/n7481/extref/nature12886-s1.pdf