Response to Feschotte

Before I went away last week, I wrote a post that was critical of a recent paper in Trends in Ecology and Evolution for essentially equating the term 'Eukaryotes' with animals. The corresponding author of that paper, Cedric Feschotte, came by to comment on the post, but because I was away at the time I am only getting to the response now.

I'll start by saying I have nothing against Cedric or his lab and only have a vague familiarity with their work, so I did not pick this paper for any other reason than it was the most recent paper that I have read, which made this mistake. The title of the paper is “Promiscuous DNA: horizontal transfer of transposable elements and why it matters for eukaryotic evolution”, and my issue with it is simply that the entire focus is on animals - a very small portion of eukaryotes. I take issue with this particular offense because I think it is disingenuous to claim to have discovered something in all of complex life, when in fact the findings are limited to a small subset of that group.

I am actually happy that Cedric came by to share his opinion and hopefully he will continue to. In the interest of discussion I'll include the points of comment by Cedric in italics and my response below each.

As the corresponding author on the TREE paper you are referring to, I am feeling obligated to rectify several inaccuracies in your post.

First, I have to say I am quite offended, not to say shocked, by your post. Because my group is actually one of the very few, along with my wife’s Ellen Pritham, to work on transposable elements in several ‘protist’ genomes, including apicomplexans, trichomonads, Entamoeba rhizarians…etc. We are collaborating with a number of eminent protist biologists and hence we are well aware that the diversity of eukaryotes is far from limited to metazoans.

Being aware of the difference is not the same as structure your writing to reflect it. I'm reading one paper without background knowledge of what Cedric's group has done in the past, nor who their collaborators are. IMO, this knowledge should not be required when evaluating an article. If you're aware of the diversity, then make that clear in a paper on animals by not claiming it is research that 'matters for eukaryotic evolution'.

Unlike most others in the TE field, we always try to be very careful in our wording to avoid generalizing some of our findings in metazoans to other eukaryotic groups. Granted, it is not often easy to do, it is a fine line between oversimplification and overspecialization – especially in journals that have tight constraints in title, abstract, references and overall text length. For simplicity and brevity, we have to use general terms such as ‘eukaryotes’ in titles and abstract, to make it clear that prokaryotes will not be covered and also, in this particular piece, to emphasize the differences with what is known on HT of mobile elements in prokaryotes. I guess we could have used ‘multicellular eukaryotes’, although I suspect the editor would have demanded to delete ‘multicellular’!!

This is strongly reminiscent of a conversation going on over at DrugMonkey's blog about the veracity of statements by scientists. It is easy to blame the editor and say that that they wouldn't allow a different title, but the people at TREE are good and certainly value accuracy over hype. And perhaps that is the root of the problem when we are talking about cases where the authors are fully aware that they are grossly over-stating their findings in a title and abstract. Are we getting too far into the journalistic mindset when it comes to titles? Is it all about grabbing a reader with a title, even at the cost of accuracy? I certainly hope not, because I think that undermines the central believe that scientists should strive for accuracy to the best of their ability at that time.

The telling statement is that it is Cedric's assumption that the more accurate title would have been turned down, but it was never considered in the first place.

More importantly, however, is that even the revised title proposed by Cedric would have been incorrect. In what way do animals represent 'multicellular eukaryotes' any better than they do eukaryotes? Off the top of my head I can think of 4 other independent lineages that have evolved multicellularity. None of these are discussed in any detail in the paper in question.

Secondly, and perhaps more importantly, before accusing us to cite papers (Keeling and Palmer) after only reading their abstract, you should be reading more carefully the ones you are writing about in your blog. In the Introduction of our paper, which you are paraphrasing partially and inaccurately in your post, we were careful to say that we are placing emphasis in this review on transfer of genetic material between the NUCLEUS of MULTICELLULAR eukaryotes. Indeed we state in the very first sentence of the paper:

“Horizontal transfer has long been recognized as a crucial
mechanism driving bacterial evolution [1]. In contrast, the
evolutionary significance of horizontal transfer between
the nuclear genomes of multicellular eukaryotes has
remained more obscure [2].”

We are perfectly aware of the growing number of HGT events reported in unicellular eukaryotes, a topic well covered in the Keeling and Palmer review. But note that almost all of these are transfers from bacteria or plastids to nucleus. There are still very few clear instances of HGT reported between the nuclear genomes of eukaryotes, even for unicellular organisms (but see the remarkable case of bdelloid rotifers). As for plants, until last month or so, there had been not a single report of nucleus-to-nucleus HGT between plants. All the cases of HGT reported by Palmer and others involved organelle genes.

In terms of TEs, the focus of our review, we still no very little on unicellular eukaryotes (aside from a handful of yeasts).

This is a non-sequitur. Just because the data on TEs aren't readily available in published genomes and you are unwilling to analyze newly released non-animal genomes for TEs in a review (which, BTW, I don't have a problem with), how does that excuse the title and abstract? I still don't get how it is okay to use a small subset of an enormous group and claim it as representative. If the data from the rest of the group is unavailable, then say so, but don't claim to speak for it by using the term eukaryote (or even multicellular eukaryotes) when referring to animals.

Most of the protist genomes sequenced so far are from species with very small genomes (e.g. apicomplexans) which seem to harbor very few, if any TEs.

This is blatantly false. Oomycetes (5 available)? Diatoms (3 available)? Ectocarpus (multicellular, but released after the paper in question was likely in press)? Many green algae? Emiliania? Dictyostelium? There are many more even before we get into the multiple fungal genomes and things like Monosiga, which would be a particularly good comparison against animals. Again, being aware of the diversity and actually using it are two very different issues.

Out of the >200 solid cases of HT of TEs reported to date, to the best of our knowledge, none involve a unicellular or protist species. It means we know virtually nothing about the phenomenon of HTT in these organisms. So I don’t see what else we could have possibly said about these organisms in our review. Clearly a lot of work remains to be done in this area. As I mentioned above, we are actively working on characterizing TEs in several, deeply diverged protist lineages.

In several places in our paper, we warn against the temptation to generalize the patterns inferred in this review based on a very limited sample of (mostly multicellular) eukaryotic species. For example, page 540, we wrote:

” Currently, the taxonomic sampling bias of whole-genome sequencing projects
among eukaryotes leans strongly towards animals
and fungi, making it difficult to infer patterns among
under-represented groups, for example unicellular eukaryotes
and plants. Thus, it is important to be cautious when
making generalizations about patterns of HTT based on
1.5 million extant eukaryotes become
available.”

I think we couldn’t be more explicit and more cautious. Don’t you think?

Agreed. I think this is the most explicit and cautious statement in the whole paper. Too bad it is buried on page four of an eight page paper with the title "Promiscuous DNA: horizontal transfer of transposable elements and why it matters for eukaryotic evolution". It hardly counters such statements, such as "We argue that the introduction of transposable elements by horizontal transfer in eukaryotic genomes has been a major force propelling genomic variation and biological innovation.", which can be found in the abstract.

For all these reasons, I found your post completely unfair and misleading. I am very sensitive to the issue of ignorance of the general public and lack of attention given to the wonderful diversity of unicellular eukaryotes, and I found myself regularly advocating and educating my students and colleagues about it.

In my opinion, it was poor judgment from your part to use our paper to illustrate a real and serious issue. It is not serving the cause eloquently. I hope you’ll realize this when you read it again, this time more carefully. And I hope people reading your blog will do the same.

I appreciate the sentiment here and I think it is great that Cedric is both aware of non-animal eukaryotic diversity and is interested in preachin' the word to others, but this was hardly a poor paper to use to illustrate my point. The intent and background of a author is not part of a paper, just the words on the page. If you want to talk about a large group of organisms, then do it, but talking the talk and then pulling a bait-`n-switch on the reader is not the same thing. This is ESPECIALLY true when the objective is hyping the title at the expense of accuracy. It is important to 'sell' one's papers in this day and age, but I think it is more constructive to sell them on their content and not default to the New York Post Handbook on Headlines.

3 responses so far

  • tideliar says:

    *grabs popcorn*

    I'm only a neuroscientist, albeit one with a PhD in MolGen, so I'm gonna leave thew debate to the grow-ups, but this seems like a very fair "rebuttal" of their defence. I do very much hope Prof. Feschotte or his collaborators return to continue the discussion though.

  • Nat says:

    This is another instance of science blogosphere for the win. Consider how long it would have taken to have this discussion in the bounds of a journal "Letters to Editor" and Reply. It wouldn't have been worth it.

    I love this stuff!

  • gnuma says:

    The Feschotte and Pritham labs have a pretty impressive track record of analyzing TE dynamics in genomes -- maybe next they will conquer the taxa you list above!

Leave a Reply