ChatGPT won't pass my exams!

Here are a few questions for ChatGPT and its answers. The AI program takes the most common information on the web and spews it back at you. It cannot tell which information is correct or which information is more accurate.

It's easy to recognize that these answers were written by something that's not very good at critical thinking. I agree with other professors that they mimic typical undergraduate answers but I disagree that these answers would get them a passing grade.

ChatGPT shares one very important feature that's common in undergraduate answers to essay questions: it gives you lots of unecessary information that's not directly relevant to the question.

It's important to note that (lol) these ChatGPT answers share another important feature with many of the answers on my exams: they look very much like BS!

Read a short preview of What's in Your Genome

Check out this website, What's in Your Genome if you want to see a preview of my book (Preface, Prologue, part of Chapter 1).

The book will be released on May 16, 2023. We are currently working on the proofs prior to typesetting. You can preorder the hardcover version on Amazon.ca (Canada) for $39.95 (Cdn). I don't know when the electronic version will be available. You can also preorder on Amazon.co.uk for £26.99.

You can't yet preorder on Amazon.com and there's no information on that site about availability. I don't know about other Amazon sites.

---

Jupiter weighs two quettagrams

New names for very large and very small weights and sizes have been adopted.

Last November's meeting of the General Conference on Weights and Measures wasn't covered by the major media outlets so you probably don't know that the mass of an electron is now one rontogram and the diameter of the universe is about one ronnameter [SI units get new prefixes for huge and tiny numbers].¹

The official SI prefixes for very large things are now ronna (10²⁷) and quetta (10³⁰) and the prefixes for very small things are ronto (10^-27) and quecto (10^-30).

This is annoying because we've just gotten used to zetta, yotta, zepto, and yocto (adopted in 1991). I suspect that the change was prompted by the huge storage capacity of your latest smartphone (several yottabytes) and the wealth of the world's richest people (several zeptocents). Or maybe it was the price of houses in Toronto. Or something like that. In any case, we needed to prepare for kilo or mega increases.

The bad news is that the latest additions used up the last two available letters of the alphabet so if things get any bigger or smaller we may have to add a few more letters to the alphabet.

---

1. A friendly reader has pointed out that my title should have been "The mass of Jupiter is two quettagrams." My bad.

The function wars are over

In order to have a productive discussion about junk DNA we needed to agree on how to define "function" and "junk." Disagreements over the definitions spawned the Function Wars that became intense over the past decade. That war is over and now it's time to move beyond nitpicking about terminology.

The idea that most of the human genome is composed of junk DNA arose gradually in the late 1960s and early 1970s. The concept was based on a lot of evidence dating back to the 1940s and it gained support with the discovery of massive amounts of repetitive DNA.

Various classes of functional DNA were known back then including: regulatory sequences, protein-coding genes, noncoding genes, centromeres, and origins of replication. Other categories have been added since then but the total amount of functional DNA was not thought to be more than 10% of the genome. This was confirmed with the publication of the human genome sequence.

From the very beginning, the distinction between functional DNA and junk DNA was based on evolutionary principles. Functional DNA was the product of natural selection and junk DNA was not constrained by selection. The genetic load argument was a key feature of Susumu Ohno's conclusion that 90% of our genome is junk (Ohno, 1972a; Ohno, 1972b).

A multivalent mRNA flu vaccine

Scientists have recently developed an mRNA-lipid nanoparticle flu vaccine that protects against all known flu variants.

There are four types of influenza viruses but only the A and B viruses cause significant problems [Types of Influenza Viruses]. Influenza A viruses are the ones can have caused pandemics in the past and the current flu vaccinations are unlikely to offer a lot of protection since they are only directed toward the specific subtypes that are predicted to cause problems in the next flu season. The next global flu pandemic will probably come from a new influenza A virus that nobody predicted.

Junk DNA, TED talks, and the function of lncRNAs

Most of our genome is transcribed but so far only a small number of these transcripts have a well-established biological function.

The fact that most of our genome is transcribed has been known for 50 years but that fact only became widely known with the publication of ENCODE's preliminary results in 2007 (ENCODE, 2007). The ENOCDE scientists referred to this as "pervasive transription" and this label has stuck.

By the end of the 1970s we knew that much of this transcription was due to introns. The latest data shows that protein coding genes and known noncoding genes occupy about 45% of the genome and most of that is intron sequences that are mostly junk. That leaves 30-40% of the genome that is transcribed at some point producing something like one million transcripts of unknown function.

A University of Chicago history graduate student's perspective on junk DNA

A new master's thesis on the history of junk DNA has been posted. It's from the Department of History at the University of Chicago.

My routine scan for articles on junk DNA turned up the abstract of an M.A. thesis on the history of junk DNA: Requiem for a Gene: The Problem of Junk DNA for the Molecular Paradigm. The supervisor is Professor Emily Kern in the Department of History at the University of Chicago. I've written to her to ask for a copy of the thesis and for permission to ask her, and her student, some questions about the thesis. No reply so far.

Here's the abstract of the thesis.

“Junk DNA” has been at the center of several high-profile scientific controversies over the past four decades, most recently in the disputes over the ENCODE Project. Despite its prominence in these debates, the concept has yet to be properly historicized. In this thesis, I seek to redress this oversight, inaugurating the study of junk DNA as a historical object and establishing the need for an earlier genesis for the concept than scholars have previously recognized. In search of a new origin story for junk, I chronicle developments in the recognition and characterization of noncoding DNA sequences, positioning them within existing historiographical narratives. Ultimately, I trace the origin of junk to 1958, when a series of unexpected findings in bacteria revealed the existence of significant stretches of DNA that did not encode protein. I show that the discovery of noncoding DNA sequences undermined molecular biologists’ vision of a gene as a line of one-dimensional code and, in turn, provoked the first major crisis in their nascent field. It is from this crisis, I argue, that the concept of junk DNA emerged. Moreover, I challenge the received narrative of junk DNA as an uncritical reification of the burgeoning molecular paradigm. By separating the history of junk DNA from its mythology, I demonstrate that the conceptualization of junk DNA reveals not the strength of molecular biological authority but its fragility.

It looks like it might be a history of noncoding DNA but I won't know for certain until I see the entire thesis. It's only available to students and staff at the University of Chicago.

---

Protein concentrations in E. coli are mostly controlled at the level of transcription initiation

The most important step in the regulation of protein-coding genes in E. coli is the rate of binding of RNA polymerase to the promoter region.

A group of scientists at the University of California at San Diego and their European collaborators looked at the concentrations of proteins and mRNAs of about 2000 genes in E. coli. They catalogued these concentrations under several different growth conditions in order to determine whether the level of protein being expressed from each of these genes correlated with transcription rate, translation rate, mRNA stability or other levels of gene expression.

The paper is very difficult to understand because the authors are primarily interested in developing mathematical formulae to describe their results. They expect you to understand equations like,

even though they don't explain the parameters very well. A lot of important information is in the supplements and I couldn't be bothered to download and read them. I don't think the math is anywhere near as important as the data and the conclusions.

Publishing a science book - Lesson #1: The publisher is always right about everything

Don't bother trying to reason with a publisher. All of them have different views on proper style and every single one of them is absolutely certain that their style is the only correct one.

I'm in the middle of the copyedit stage of my book. This is the stage where a copyeditor goes through your manuscript and makes any corrections to spelling and grammar. This is a lot of work for any copyeditor having to deal with one of my manuscripts and I greatly appreciate the effort. My book is a lot better now than it was a few weeks ago. (Who knew that there was only one l in canceled?)

It's also the stage where the publisher imposes their particular style on the manusript and that can be a problem. I'll document some of the issues in subsequent posts but to give you an example, consider the titles of books in the reference list. I wrote it like this: The Selfish Gene and Molecular and Genome Evolution. This is not in line with my publisher's handbook of style so the titles were converted to lowercase as in: The selfish gene and Molecular and genome evolution. I objected, pointing to numerous other science books that used the same titles that are on the covers of the books and suggesting that my readers were more familiar with The Selfish Gene than with The selfish gene.

I was overruled by my publisher who noted that they make their style choices for good reasons—it's for "consistency, clarity, and ease of reading." I assume that publishers, such as Oxford, would make the same argument while insisting that the title should be The Selfish Gene.

In case you ever find yourself in this position, you should keep in mind that your contract will almost certainly say that the publisher has complete control of your book and they can make any changes they want as long as it doesn't affect the meaning of what you wrote.

Here's what it says in my contract, "The Publisher shall publish the Author's work in whatever style and format it thinks most suitable ... While the Publisher may, in its sole discretion, consult the Author with respect to said style and format, the Publisher retains the right to make all final decisions on matters of format, design, selling price and marketing."

I was aware of some issues with inappropriate covers and tiles in the past so I had an extra sentence added to the contract that said, "The Publisher and Author will discuss and agree upon the title and cover design." It's a good thing I put that in because the publisher was pressuring me to change the title of the book and I was able to resist.

Authors can't win most fights over style and format. I've been discussing the publishing of science books with a number of other authors over the past few months and several of them told me not to bother trying to argue with a publisher because they will never give in. They have a set style for all books and they won't make an exception for an individual author no matter how good an argument you make.

I didn't listen to those other authors. Silly me.

I'm thinking of trying to write a standard set of guidelines that scientists could put into their contracts to cover the most egregious style restrictions. It might be helpful if all science writers would insist on inserting these guidelines into their contracts.

---

Can the AI program ChatGPT pass my exam?

There's a lot of talk about ChatGPT and how it can prepare lectures and get good grades on undergraduate exams. However, ChatGPT is only as good as the information that's popular on the internet and that's not always enough to get a good grade on my exam.

ChatGPT is an artificial intelligence (AI) program that's designed to answer questions using a style and language that's very much like the responses you would get from a real person. It was developed by OpenAI, a tech company in San Francisco. You can create an account and log in to ask any question you want.

Several professors have challenged it with exam questions and they report that ChatGPT would easily pass their exams. I was skeptical, especially when it came to answering questions on controversial topics where there was no clear answer. I also suspected that ChatGPT would get it's answers from the internet and this means that popular, but incorrect, views would likely be part of ChatGPT's response.

Here are my questions and the AI program's answers. It did quite well in some cases but not so well in others. My main concern is that programs like this might be judged to be reliable sources of information despite the fact that the real source is suspect.

Did molecular biology make any contribution to evolutionary theory?

Some evolutionary biologists think—incorrectly, in my opinion—that molecular biology has made no contributions to our understanding of evolution.

PNAS published a series of articles on Gregor Mendel and one of them caught my eye. Here's what Nicholas Barton wrote in his article The "New Synthesis".

During the 1960s and 1970s, there were further conceptual developments—largely independent of the birth of molecular biology during the previous two decades (15). First, there was an understanding that adaptations cannot be explained simply as being “for the good of the species” (16, 17). One must explain how the genetic system (including sexual reproduction, recombination, and a fair meiosis, with each copy of a gene propagating with the same probability) is maintained through selection on individual genes, and remains stable despite mutations that would disrupt the system (17, 19, 20). Second, and related to this, there was an increased awareness of genetic conflicts that arise through sexual reproduction; selfish elements may spread through biased inheritance, even if they reduce individual fitness (19, 21, 22). In the decade following the discovery that DNA carries genetic information, all the fundamental principles of molecular biology were established: the flow of information from sequences of DNA through RNA to protein, the regulation of genes by binding to specific sequences in promoters, and the importance of allostery in allowing arbitrary regulatory networks (23, 24). Yet, the extraordinary achievements of molecular biology had little effect on the conceptual development of evolutionary biology. Conversely, although evolutionary arguments were crucial in the founding of molecular biology, they have had rather little influence in the half-century since (e.g., ref. 25). Of course, molecular biology has revealed an astonishing range of adaptations that demand explanation—for example, the diversity of biochemical pathways, that allow exploitation of almost any conceivable resource, or the efficiency of molecular machines such as the ribosome, which translates the genetic code. Technical advances have brought an accelerating flood of data, most recently, giving us complete genome sequences and expression patterns from any species. Yet, arguably, no fundamentally new principles have been established in molecular biology, and, in evolutionary biology, despite sophisticated theoretical advances and abundant data, we still grapple with the same questions as a century or more ago.

This does not seem fair to me. I think that neutral theory, nearly neutral theory, and the importance of random genetic drift relied heavily on work done by molecular biologists. Similarly, the development of dating techniques using DNA and protein sequences is largely the work of molecular biologists. It wasn't the adaptationists or the paleontologists who discovered that humans and chimpanzees shared a common ancestor 5-7 million years ago and it wasn't either of those groups who discovered the origin of mitochondria.

And some of us are grappling with the idea that most of our genome is junk DNA, a question that never would have occurred to evolutionary biologists from a century ago.

Barton knows all about modern population genetics and the importance of neutral theory because later on he says,

If we consider a single allele, then we can see it as “effectively neutral” if its effect on fitness is less than ∼1/2Ne. This idea was used by Ohta (54) in a modification of the neutral theory, to suggest why larger populations might be less diverse than expected (because a smaller fraction of mutations would be effectively neutral), and why rates of substitution might be constant per year rather than per generation (because species with shorter generation times might tend to have large populations, and have a smaller fraction of effectively neutral mutations that contribute to long-term evolution). Lynch (21) has applied this concept to argue that molecular adaptations that are under weak selection cannot be established or maintained in (relatively) smaller populations, imposing a “drift barrier” to adaptation. Along the same lines, Kondrashov (55) has argued that deleterious mutations with Nes ≈ 1 will accumulate, steadily degrading the population. Both ideas seem problematic if we view adaptation as due to optimization of polygenic traits: Organisms can be well adapted even if drift dominates selection on individual alleles, and, under a model of stabilizing selection on very many traits, any change that degrades fitness can be compensated.

Barton may think that the drift-barrier hypothesis is "problematic" but it certainly seems like a significant advance that owes something to molecular biology.

What do you think? Do you agree with Barton that, "... the extraordinary achievements of molecular biology had little effect on the conceptual development of evolutionary biology."

---

Sequencing both copies of your diploid genome

New techniques are being developed to obtain the complete sequences of both copies (maternal and paternal) of a typical diploid individual.

The first two sequences of the human genome were published twenty years ago by the International Human Genome Project and by a company called Celera Genomics. The published sequences were a consensus using DNA from multiple indivduals so the final result didn't represent the sequence of any one person. Furthermore, since each of us has inherited separate genomes from our mother and father, our DNA is actually a mixture of two different haploid genomes. Most published genome sequences are an average of these two separate genomes where the choice of nucleotide at any one position is arbitrary.

The first person to have a complete genome sequence was James Watson in 2007 but that was a composite genome sequence. Craig Venter's genome sequence was published a few months later and it was the first complete genome sequence containing separate sequences of each of his 46 chromosomes. (One chromosome from each of his parents.) In today's language, we refer to this as a diploid sequence.

The current reference sequence is based on the data published by the public consortium (International Humand Genome Project)—nobody cares about the Celera sequence. Over the years, more and more sequencing data has been published and this has been incorporated into the standard human reference genome in order to close most gaps and improve the accuracy. The current version is called GRCh38.p14 from February 3, 2022. It's only 95% complete because it's missing large stretches of repetitive DNA, especially in the centromere regions and at the ends of each chromosome (telomeric region).

The important point for this discussion is that CRCh38 is not representative of the genomes of most people on Earth because there has been a bias in favor of sequencing European genomes. (Some variants are annotated in the reference genome but this can't continue.) Many scientists are interested the different kinds of variants present in the human population so they would like to create databases of genomes from diverse populations.

The first complete, telomere-to-telomere (T2T), human genome sequence was published last year [A complete human genome sequence (2022). It was made possible by advances in sequencing technology that generated long reads of 10,000 bp and ultra-long reads of up to 1,000,000 bp [Telomere-to-telomere sequencing of a complete human genome]. The DNA is from a CHM13 cell line that has identical copies of each chromosome so there's no ambiguity due to differences in the maternal and paternal copies. The full name of this sequence is CHM13-T2T.

The two genomes (CRCh38 and CHM13) can't be easily merged so right now there are competing reference genomes [What do we do with two different human genome reference sequences?].

The techniques used to sequence the CHM13 genome make it possible to routinely obtain diploid genome sequences from a large number of individuals because overlapping long reads can link markers on the same chromosome and distinguish between the maternal and paternal chromosomes. However, in practice, the error rate of long read sequencing made assembly of separate chromosomes quite difficult. Recent advances in the accuracy of long read sequencing have been developed by PacBio, and this high fidelity sequencing (PacBio HiFi sequencing) promises to change the game.

The Human Pangene Reference Consortium has tackled the problem by sequencing the genome of an Ashkenazi man (HG002) and his parents (HG002-father and HG004-mother) using the latest sequencing techniques. They then asked the genome community to submit their assemblies using their best software in a kind of "assembly bakeoff." They got 23 responses.

Jarvis, E. D., Formenti, G., Rhie, A., Guarracino, A., Yang, C., Wood, J., et al. (2022) Semi-automated assembly of high-quality diploid human reference genomes. Nature, 611:519-531. [doi: 10.1038/s41586-022-05325-5]

The current human reference genome, GRCh38, represents over 20 years of effort to generate a high-quality assembly, which has benefitted society. However, it still has many gaps and errors, and does not represent a biological genome as it is a blend of multiple individuals. Recently, a high-quality telomere-to-telomere reference, CHM13, was generated with the latest long-read technologies, but it was derived from a hydatidiform mole cell line with a nearly homozygous genome. To address these limitations, the Human Pangenome Reference Consortium formed with the goal of creating high-quality, cost-effective, diploid genome assemblies for a pangenome reference that represents human genetic diversity. Here, in our first scientific report, we determined which combination of current genome sequencing and assembly approaches yield the most complete and accurate diploid genome assembly with minimal manual curation. Approaches that used highly accurate long reads and parent–child data with graph-based haplotype phasing during assembly outperformed those that did not. Developing a combination of the top-performing methods, we generated our first high-quality diploid reference assembly, containing only approximately four gaps per chromosome on average, with most chromosomes within ±1% of the length of CHM13. Nearly 48% of protein-coding genes have non-synonymous amino acid changes between haplotypes, and centromeric regions showed the highest diversity. Our findings serve as a foundation for assembling near-complete diploid human genomes at scale for a pangenome reference to capture global genetic variation from single nucleotides to structural rearrangements.

We don't need to get into all the details but there are a few observations of interest.

• All of the attempted assemblies were reasonably good but the best ones had to make use of the parental genomes to resolve discrepancies.
• Some assemblies began by separating the HG002 (child) sequences into two separate groups based on their similarity to one of the parents. Others generated assemblies without using the parental data then fixed any problems by using the parental genomes and a technique called "graph-based phasing." The second approach was better.
• All of the final assemblies were contaminated with varying amounts of E. coli and yeast DNA or and/or various adaptor DNA sequences that were not removed by filters. All of them were contaminated with mitochondrial DNA that did not belong in the assembled chromosomes.
• The most common sources of assembly errors were: (1) missing joins where large stretches of DNA should have been brought together, (2) misjoins where two large stretches (contigs) were inappropriately joined, (3) incorrect inversions, and (4) false duplications.
• The overall accuracy of the best assemblies was one base pair error in 100,000 bp (10^-5).
• Using the RefSeq database of 27,225 genes, most assemblies captured almost all of these confirmed and probable genes but several hundred were not complete and many were missing.
• No chromosome was complete telomere-telomere (T2T) but most were nearly complete including the complicated centromere and telomere regions.
• The two genomes (parental and maternal) differed at 2.6 million SNPs (single nucleotides), 631,000 small structural variations (<50 bp), and 11,600 large structural variations (>50 bp).
• The consortium used the best assembly algorithm to analyze the genomes of an additional 47 individuals. They began with the same coverage used for HG002; namely, 35X coverage. (Each stretch of DNA was sequenced 35 times on average - about equal amounts in both directions.) This was not successful so they had to increase the coverage to 130X to get good assemblies. They estimate that each additional diploid sequence will reguire 50-60X coverage. This kind of coverage would have been impossible in the 1990s when the first human genome was assembled but now it's fairly easy as long as you have the computer power and storage to deal with it.

---

University of Michigan biochemistry students edit Wikipedia

Students in a special topics course at the University of Michigan were taught how to edit a Wikipedia article in order to promote function in repetitive DNA and downplay junk.

The Wikipedia article on Repeated sequence (DNA) was heavily edited today by students who were taking an undergraduate course at the University of Michgan. One of the student leaders, Rasberry Neuron, left the following message on the "Talk" page.

This page was edited for a course assignment at the University of Michigan. The editing process included peer review by four students, the Chemistry librarian at the University of Michigan, and course instructors. The edits published on 12/01/2022 reflect improvements guided by the original editing team and the peer review feedback. See the article's History page for information about what changes were made from the previous version.

References to junk DNA were removed by the students but quickly added back by Paul Gardner who is currently fixing other errors that the students have made.

I checked out the webpage for the course at CHEM 455_505 Special Topics in Biochemistry - Nucleic Acids Biochemistry. The course description is quite revealing.

We now realize that the human genome contains at least 80,000 non-redundant non-coding RNA genes, outnumbering protein-coding genes by at least 4-fold, a revolutionary insight that has led some researchers to dub the eukaryotic cell an “RNA machine”. How exactly these ncRNAs guide every cellular function – from the maintenance and processing to the regulated expression of all genetic information – lies at the leading edge of the modern biosciences, from stem cell to cancer research. This course will provide an equally broad as deep overview of the structure, function and biology of DNA and particularly RNA. We will explore important examples from the current literature and the course content will evolve accordingly.

The class will be taught from a chemical/molecular perspective and will bring modern interdisciplinary concepts from biochemistry, biophysics and molecular biology to the fore.

Most of you will recognize right away that there are factually incorrect statements (i.e. misinformation) in that description. It is not true that there are at least 80,000 noncoding genes in the human genome. At some point in the future that may turn out to be true but it's highly unlikely. Right now, there are at most 5,000 proven noncoding genes. There are many scientists who claim that the mere existence of a noncoding transcript is proof that a corresponding gene must exist but that's not how science works. Before declaring that a gene exists you must present solid evidence that it produces a biologically relevant product [Most lncRNAs are junk] [Wikipedia blocks any mention of junk DNA in the "Human genome" article] [Editing the Wikipedia article on non-coding DNA] [On the misrepresentation of facts about lncRNAs] [The "standard" view of junk DNA is completely wrong] [What's In Your Genome? - The Pie Chart] [How many lncRNAs are functional?].

I'm going to email a link to this post to the course instructors and some of the students. Let's see if we can get them to discuss junk DNA.

---

How not to write a Nature abstract

A friend recently posted a figure on Facebook that instructs authors in the correct way to prepare a summary paragraph (abstract) for publication in Nature. It uses a specific example and the advice is excellent [How to construct a Nature summary paragraph].

I thought it might be fun to annotate a different example so I randomly selected a paper on genomics to see how it compared. The one that popped up was An integrated encyclopedia of DNA elements in the human genome.

---

What I'm reading these days

So many books ... so little time.

---