tag:blogger.com,1999:blog-37148773.post5339606949595936667..comments2024-03-27T14:50:47.345-04:00Comments on <center>Sandwalk</center>: Bacteria Phylogeny: Facing Up to the ProblemsLarry Moranhttp://www.blogger.com/profile/05756598746605455848noreply@blogger.comBlogger17125tag:blogger.com,1999:blog-37148773.post-32009626746235125982011-04-12T08:39:37.590-04:002011-04-12T08:39:37.590-04:00This is a very helpful approach for metagenomic an...This is a very helpful approach for metagenomic and phylogenetic research. I am a master student in Greece and I would die to try the AMPHORA pipeline in my master thesis and an article we prepare here in our lab at the Biomedical Research Foundation (BRF) of the Academy of Athens. Unfortunately I can not make it work, even after a lot of effort from friends and my shelf. The main problem is that I am not sure about the format of the input file I use (txt in Fasta format of 104 bacterial proteoms). But even when I try to run AMPHORA/Phylotyping.pl with a reference file, the application runs but it stops without printing or saving the tree. Please help!Joannahttps://www.blogger.com/profile/04265374885055345313noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-60175341974628304082008-10-19T21:17:00.000-04:002008-10-19T21:17:00.000-04:00Jud - as pointed out, the tree as presented is not...Jud - as pointed out, the tree as presented is not meant to be read as rooted, so the base of the tree could just as readily be between Gammaproteobacteria and all other lineages as between Deinococci and other lineages as shown. That said, the arrangement shown has obviously been chosen for its similarity to the arrangement found in many rDNA trees, which find Aquificae, Thermotogae and Deinococci quite low on the tree. Whether or not this arrangement represents the actual evolutionary history is still very debatable.<BR/><BR/>Ultrastructurally, eubacteria can be divided into two broad groups. Monodermata include mostly Gram-positive bacteria with a single cell membrane inside the cell wall, and would include Firmicutes, Actinobacteria and Thermotogae. Didermata are mostly Gram-negative bacteria with two cell membranes, one on either side of the cell wall, and would include everything to the right of Firmicutes in Wu and Eisen's tree, as well as Cyanobacteria, Aquificae and Deinococci. Either one of these two groups could be paraphyletic with regard to each other. The separation of Deinococci and Aquificae from other didermates might indicate a long-branch effect with those two groups appearing in the wrong part of the tree. Alternatively, Monodermata could have arisen polyphyletically through multiple losses of the outer membrane, or Didermata could be polyphyletic with multiple monodermate ancestors developing an outer membrane.<BR/><BR/>I suppose my central point is that the higher-level bacterial phylogeny is far from settled (and if researchers like Doolittle are correct, may not even be identifiable if time and LGT have eroded all the reliable signals).Christopher Taylorhttps://www.blogger.com/profile/11075565866351612441noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-78301467493763836972008-10-17T13:41:00.000-04:002008-10-17T13:41:00.000-04:00anonymous wrote: [T]he tree in the figure is an un...anonymous wrote: <I>[T]he tree in the figure is an unrooted tree so it does not really say anything about the base.</I><BR/><BR/>Yup, sorry, my fault for incorrectly phrasing the question. An attempt to do better: Does it appear possible or likely that bacteria such as Thermus thermophilus and Deinococcus geothermalis were ancestral to the other bacteria species shown in the figure?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-64488066409249745222008-10-17T13:16:00.000-04:002008-10-17T13:16:00.000-04:00Hi Jud, the tree in the figure is an unrooted tree...Hi Jud, the tree in the figure is an unrooted tree so it does not really say anything about the baseAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-65466995710876931062008-10-17T07:09:00.000-04:002008-10-17T07:09:00.000-04:00It's wonderful for a layperson like me to be able ...It's wonderful for a layperson like me to be able to read about research like this.<BR/><BR/>I've looked at the figure reproduced in the post in the PDF version at Genome Biology, at a high enough zoom level for my middle-aged eyes to read text. What I see there leads me to the following question: Do the results tend to show bacteria such as Thermus thermophilus and Deinococcus geothermalis at the base of the phylogenetic tree? If so, this may be something well-understood in the field, but I hadn't known it and find it interesting.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-71507355011129210002008-10-16T23:00:00.000-04:002008-10-16T23:00:00.000-04:00Oh, and I note, for the Venter Sargasso Sea paper,...Oh, and I note, for the Venter Sargasso Sea paper, I used HSP70, EF-TU, EFG, RpoB and RecA as our protein markers, so maybe you would like those more?Jonathan Eisenhttps://www.blogger.com/profile/07953790938128734305noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-26367403838065855292008-10-16T19:53:00.000-04:002008-10-16T19:53:00.000-04:00Well #1 thanks for writing about our paper #2 ... ...Well #1 thanks for writing about our paper <BR/><BR/>#2 ... I just want to chime in on the species issue. I think there is no reason why we cannot use the term species for bacterial groups. Sure, they are not quite the same groupings as we would see in eukaryotic species, but there really do seem to be true groupings where on average many of the genes in the genome are more similar/related among members of a group than with other groups. To me, that is good enough to call things a species as it indicates some higher rate of gene flow within the group than between. Sure, lateral transfer messes things up, but it seems to not have eliminated consistent groupings (whether you believe Ford Doolittles argument about what this means or not).Jonathan Eisenhttps://www.blogger.com/profile/07953790938128734305noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-26677143333599506672008-10-16T09:42:00.000-04:002008-10-16T09:42:00.000-04:00Allen MacNeill asks,How, precisely, is the stateme...Allen MacNeill asks,<BR/><BR/><I>How, precisely, is the statement "dead wrong"? Seems to me there are several possibilities:<BR/><BR/>... to be members of the same species, two individuals must be capable of interbreeding and producing fertile offspring under "natural" conditions. Prokaryotes simply don't do this at all.</I><BR/><BR/>This part is dead wrong. Many bacteria "species" have something akin to sex where individuals can exchange alleles,Larry Moranhttps://www.blogger.com/profile/05756598746605455848noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-70061891845026441702008-10-16T02:42:00.000-04:002008-10-16T02:42:00.000-04:00Allen MacNeill said: "Species", in other words, ar...Allen MacNeill said:<BR/><I> "Species", in other words, are a figment of the human imagination. </I><BR/><BR/>And so are colors. The whole idea of color "green" is a holdover from Platonic typological thinking and has done little or nothing to advance our understanding of how electromagnetic waves are perceived by photoreceptors. <BR/><BR/>Strangely enough, I still find it useful. :-)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-37549426808297523592008-10-16T00:40:00.000-04:002008-10-16T00:40:00.000-04:00The reason I don't like to use the term "species" ...The reason I don't like to use the term "species" to refer to the termini of prokaryote phylogenies is that it conveys all the wrong ideas about what phylogenies are all about. Focusing on reproductive incompatibility has little or no bearing on diversification among prokaryotes, and is hopelessly muddled by the problem of horizontal gene transfer. Indeed, I would go further: I think the whole idea of "species" is a holdover from Platonic typological thinking, and has done little or nothing to advance our understanding of how genetic and phenotypic diversification has proceeded in most phylogenies, including animals (the only group in which reproductive incompatibility plays a crucial role in phylogenetic diversification).<BR/><BR/>"Species", in other words, are a figment of the human imagination.Allen MacNeillhttps://www.blogger.com/profile/13692148273564872787noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-63860970034721211502008-10-16T00:33:00.000-04:002008-10-16T00:33:00.000-04:00How, precisely, is the statement "dead wrong"? See...How, precisely, is the statement "dead wrong"? Seems to me there are several possibilities:<BR/><BR/>1) That's not what Darwin, Dobzhansky, and Mayr said<BR/><BR/>2) The "biological species concept" isn't based on reproductive isolation<BR/><BR/>3) Reproduction in prokaryotes is indeed coupled with sexual recombination in essentially the same way it is in eukaryotes (especially animals)<BR/><BR/>4) Prokaryotes can be "fertile" or "infertile" in the same way that eukaryotes can<BR/><BR/>I'm curious; which of these assertions do you agree with, and on what evidence?Allen MacNeillhttps://www.blogger.com/profile/13692148273564872787noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-22872572903633866132008-10-15T16:31:00.000-04:002008-10-15T16:31:00.000-04:00Allen MacNeill,What is the point of calling the te...Allen MacNeill,<BR/><BR/><I>What is the point of calling the terminal taxa in prokaryotic phylogenies "species", beyond a wholly spurious analogy with eukaryotes?</I><BR/><BR/>You have to call them something and "species" is as good as anything on a blog such as this.<BR/><BR/>I'm well aware of the problems with defining "species."<BR/><BR/>You must have a better word or you wouldn't have raised the issue. Perhaps you could share it with us?<BR/><BR/>BTW, this part of your comment is dead wrong.<BR/><BR/><I>The Dobzhansky-Mayr "biological species concept" is (like Darwin's) based on reproductive isolation; to be members of the same species, two individuals must be capable of interbreeding and producing fertile offspring under "natural" conditions. Prokaryotes simply don't do this at all.</I>Larry Moranhttps://www.blogger.com/profile/05756598746605455848noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-72395680760679328832008-10-15T11:52:00.000-04:002008-10-15T11:52:00.000-04:00You could also check out this recent paper on defi...You could also check out <A HREF="http://dx.doi.org/10.1371/journal.pbio.0050087" REL="nofollow">this recent paper</A> on defining species in strictly asexual rotifers for a way to treat speciation where recombination and reproduction have been decoupled.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-54926148027136427242008-10-15T09:53:00.000-04:002008-10-15T09:53:00.000-04:00There is no problem of "species" in evolutionary b...There is no problem of "species" in evolutionary biology. None. There are a lot of confused people who think that "species" has to be something definable in absolute terms and applicable in 100% cases. Which isn't obviously always possible because Nature is more complex than we wish it to be. ANY classification is oversimplification of reality. <BR/><BR/>And so yes, taxonomy in prokaryotes is much more difficult but is not fundamentally different from eukaryotes. We still need to call different bacteria different names, right? So why not "species"? There would have to be *some* name anyway. If species don't exist in procaryotes then what does exist? - Some different *word*, obviously. But it's just a word.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37148773.post-17389563473955984002008-10-15T08:55:00.000-04:002008-10-15T08:55:00.000-04:00What is the point of calling the terminal taxa in ...What is the point of calling the terminal taxa in prokaryotic phylogenies "species", beyond a wholly spurious analogy with eukaryotes? None of the definitions of "species" that are generally applied to eukaryotes can be applied to prokaryotes for the simple reason that the latter do not couple reproduction with sexual recombination. The Dobzhansky-Mayr "biological species concept" is (like Darwin's) based on reproductive isolation; to be members of the same species, two individuals must be capable of interbreeding and producing fertile offspring under "natural" conditions. Prokaryotes simply don't do this at all.<BR/><BR/>Lynn Margulis (taking her cue from Sorin Sonea Lucien Mathieu, of the Université de Montreal) asserts that <BR/><BR/>"...bacteria do not have species at all (or, which amounts to the same thing, all of them together constitute one single cosmopolitan species). Speciation is a property only of nucleated organisms."<BR/><BR/>For more on the whole problem of "species" in evolutionary biology, see:<BR/>http://evolutionlist.blogspot.com/2006/03/origin-of-specious.htmlAllen MacNeillhttps://www.blogger.com/profile/13692148273564872787noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-16273100472988563122008-10-14T23:51:00.000-04:002008-10-14T23:51:00.000-04:00Thirty years ago Woese was proposing to spin off t...Thirty years ago Woese was proposing to spin off the Archaebacteria, but he was not sequencing rRNA genes. He was isolating the rRNA itself and using RNA fingerprinting techniques based on Sanger's 1965 2-D electrophoresis method to compile the oligonucleotides produce from RNase digestion. He was measuring homologies and sequences differences, but not actually producing any full rRNA sequences.<BR/><BR/>PCR, nominally invented in 1984, doesn't open the doors to easy sequencing until after about 1989.Sagehttps://www.blogger.com/profile/04259090314712198514noreply@blogger.comtag:blogger.com,1999:blog-37148773.post-6599934631975247372008-10-14T21:20:00.000-04:002008-10-14T21:20:00.000-04:00Very nice (the DOI link to the paper doesn't work ...Very nice (the DOI link to the paper doesn't work at the moment, by the way, but the other one does). The main problem that I can imagine with using concatenated sequences for prokaryotes (without my being an expert, of course) would be the Horizontal Gene Transfer issue - if different gene sections are giving you different but both perfectly accurate estimates of phylogeny, what effect will that have on your result? I've wondered if prokaryotes might be one group of organisms where a supertree approach might be more effective - making separate trees from separate genes, then combining them for your end phylogeny.Christopher Taylorhttps://www.blogger.com/profile/11075565866351612441noreply@blogger.com