Increase in genomic novelty at the origin of animals

Hi everyone,

Evolution News and Views has an interesting article titled, Groundbreaking Paper Shows Thousands of New Genes Needed for the Origin of Animals, which reports on a recent paper by Jordi Paps and Peter W. H. Holland in Nature Communications that attempts to infer the minimal protein-coding genome of the first animal and estimate the proportion of “novelties” in the ancestral genomes. Surprisingly, the paper uncovers "an unprecedented increase in the extent of genomic novelty during the origin of metazoans."

Highlights:

Genomic novelty in the origin of animals. Concerning gene novelty, we infer the ancestral metazoan genome included a remarkable 1189 Novel HG [homology groups]; this number is similar to that inferred by previous study centred on the genome of a demosponge. Our analyses indicate a threefold increase compared to novelties in the previous nodes (389, 340, and 399 novel HG in the older holozoan nodes; Fig. 1, Supplementary Data 3). The Novel HG comprise 19% of the total HG in the first metazoan, compared to only 8–10% in most older nodes examined; Planulozoa and Bilateria LCA [last common ancestor] nodes also have high proportions of Novel HG (Figs. 1, 2a, Supplementary Note 3, Supplementary Data 5)… The 1189 metazoan Novel HG contain a large number of regulatory functions compared to the metazoan Ancestral HG set (e.g., 23 vs 6% transcription factors, 11 vs 4% signalling), and are depauperate in enzymatic and metabolic functions (Supplementary Data 4 and 6). Comparing the Novel HG of each phylogenetic node, the number genes for several protein classes displays a peak in the animal ancestor (Fig. 2c); the novel functions which are more abundant in the LCA of animals (Supplementary Data 6) are nucleic acid binding (23%), transcription factors (23%), and signalling molecules (11%). Thus, the first animal genome was not only showing a higher proportion of Novel HG, but these also perform major multicellular functions in the modern fruit fly genome. The implication is that the transition was accompanied by an increase of genomic innovation, including many new, divergent, and subsequently ubiquitous genes encoding regulatory functions associated with animal multicellularity.

Twenty five novel core groups of genes in animals. We identified which novel gene functions were more retained through evolution of animals. We find a total of 25 Novel Core HG: protein groups emerging in the genome of the first animal and still present in at least 43 of the 44 metazoan genomes examined (Table 1); these are independent of alternative phylogenetic scenarios at the root of animals. For these 25 HG, we give examples of modern bilaterian gene families contained in these HG. Together they cover the spectrum of classical functions linked to animal multicellularity: gene regulation, signalling, cell adhesion and cell cycle. Earlier opisthokont LCA have much lower numbers of Novel Core HG (Fig. 1, Supplementary Data 3 and 7), supporting the importance of genetic innovation in the emergence of animals.

The paper concludes:

There are two alternative scenarios that could explain these patterns depending on the length of the branch leading to the metazoan LCA [last common ancestor]. The first assumes that the birth rate of new genes was constant over time, thus the branch leading to the first metazoan was longer than other opisthokont internodes. This would imply an extended ‘stew’ in which the molecular components of animal biology were assembled. However, we note that molecular phylogenetic analyses do not generally show longer branches in the stem lineage of animals, contrary to this scenario. The second possibility involves many new genes emerging during a short ‘popcorn’ stage, caused either by a higher gene birth rate (perhaps produced by environmental factors elevating mutation rates, or due to whole-genome duplications), and/or a lower gene death rate (due to high integration of new genes into regulatory networks). In this scenario, the acquisition of multicellularity would quickly stabilise new molecular systems for cell adhesion, cell communication and the control of differential gene expression, as shown by the increase in proportion of Novel Core HG [homology group] seen in the metazoan ancestor. These include genes previously hypothesized to be pivotal in the emergence of Metazoa, with additional genes singled out here for the first time as agents involved in the transition. This scenario is also consistent with enhanced rates of gene novelty in the ancestors of Planulozoa [ctenophores, placozoans, cnidarians + bilaterians] and Bilateria when embryonic patterning systems were being elaborated. Further data and analyses are needed to discriminate between the two scenarios.

In short: 1189 homology groups are necessary for the origin of Metazoa, 494 novel HGs are required for the origin of Eumetazoa (sponges + Planulozoa + Bilateria), 1201 novel HGs are needed for the origin of Planulozoa (ctenophores, placozoans, cnidarians + bilaterians), and an additional 1580 HGs are required for the origin of Bilateria. I’m a layperson, but these numbers strike me as quite high. From an evolutionary standpoint, is there anything particularly astonishing about them?

Finally, here’s ENV’s take on the paper:

Whether you’re an evolutionary biologist or a proponent of intelligent design, the notion that the origin of animals required new genes — even numerous new genes — might strike you as uncontroversial. But this claim was strongly challenged by UC Berkeley evolutionary paleontologist Charles Marshall who reviewed Darwin’s Doubt in the journal Science. It actually became a centerpiece of the debate between Marshall and Meyer about the Cambrian explosion… Marshall didn’t stop there. He went further, saying that Meyer has an “idiosyncratic fixation with new protein folds” and “an outdated understanding of morphogenesis” — all due to Meyer’s supposedly inaccurate claims that the Cambrian explosion would have required the origin of many new genes. Now this new paper, “Reconstruction of the ancestral metazoan genome reveals an increase in genomic novelty,” provides a direct refutation of Marshall’s insistence that the origin of animals didn’t require lots of new genes.

In the opinion of readers, does the new paper at least partially vindicate Meyer? What say you?

1 Like

Animals didn’t originate in the Cambrian. So bringing up Marshall and the Cambrian Radiation is irrelevant

You can accept this papers conclusions and still say the Cambrian didn’t require hardly any new genes

Here’s Marshall:


It’s clear he was talking about the origin of new body plans and not the origin of Metazoa

@T.j_Runyon

There is a new article here in the system which explains how a single protein “unlocked” the opportunities of a high oxygen environment.

In any case, whether a few or a lot, this is a god-guided system we are discussing.

@T.j_Runyon

I hope this link copied right:

It depends on precisely what happened in that debate. Do you have a transcript where we can direct quotes?

I’m pretty sure most biologists would think that the Cambrian explosion would require new genes, but that it is not that hard to evolve new genes. Keep in mind that this “explosion” takes place over 40 million years. If Marshal really did claim otherwise, it might be a case of disconfirmation bias.

Though, looking at the quote that @T.j_Runyon points out, Marshal does not seem to say there were no new genes. He is pointing out that new forms would arise by rewiring regulatory networks. He does NOT say that there were no new genes. Marshall is also correct that it is not clear that new folds are required (and the referenced paper does not delve into this).

He is also correct that we expect additional proteins to be accrued over time, so the number of genes in this new paper is an overestimate.


So yes, Meyer is vindicated if @vjtorley summarized Marshal correctly. However, if that quote is correct by @T.j_Runyon, I’m not sure Meyer is vindicated. I have not listened to that debate, and can’t say for sure. I would certainly say there is high disconfirmation bias against ID. However, @T.j_Runyon is the one presenting a direct quote. Can you clarify @vjtorley?

Here’s a little more from the debate, which is good and you should listen, where Marshall seems to make the point that the Cambrian didn’t required tons of new genes. Heres Marshall:

[Charles Marshall, timestamp 1:22:20] In his criticism of my review and in his book, when he talks about the origin of new genes, he talks about lysyl oxidase and points out that that’s a very important gene that arthropods… need to make their skeletons. So it looks like a brand new gene comes into existence in that phylum. And so a scientist would have said, hmm, that’s seem to be a problem, let’s explore that a little bit further… and so I did that… and what I found is that it’s four copies in humans, all animals, good heavens! It’s also in yeast, fungi, mushrooms. So it turns out that this key gene that seems so important to Stephen in the establishment of an arthropod body plan, it preexists animals by a long shot, it’s been simply co-opted to stiffen the collagen to lead to the chitin."

I don’t think this paper has anything to do with Meyer or the Cambrian.

Note: Meyer would accuse Marshall of begging the question because he is assuming the gene is homologous.

1 Like

@T.j_Runyon

It’s important to remember the difference between ordinary scientific methodology… and the methods of those who can explain rapid mutations based on “theology-related metaphysics”. In other words, if God is the one invoking changes as needed, God-guided evolution can be as fast as it NEEDS to be.

Hi everyone,

Thank you all for your comments.

@swamidass:
I’m pretty sure most biologists would think that the Cambrian explosion would require new genes, but that it is not that hard to evolve new genes. Keep in mind that this “explosion” takes place over 40 million years. If Marshall really did claim otherwise, it might be a case of disconfirmation bias.

I don’t have a transcript of the debate, but it appears that Marshall did say that. In his 2013 review of Meyer’s book, Marshall wrote:

His [Meyer’s] case against current scientific explanations of the relatively rapid appearance of the animal phyla rests on the claim that the origin of new animal body plans requires vast amounts of novel genetic information coupled with the unsubstantiated assertion that this new genetic information must include many new protein folds. In fact, our present understanding of morphogenesis indicates that new phyla were not made by new genes but largely emerged through the rewiring of the gene regulatory networks (GRNs) of already existing genes (1).

Footnote (1) refers to D. H. Erwin, J. W. Valentine, The Cambrian Explosion: The Construction of Animal Biodiversity (Roberts and Company, Greenwood Village, CO, 2013).

So according to Marshall, “new phyla were not made by new genes.” On that point, he’s at least partly wrong. It seems that Meyer is right on one thing: lots of new genes were required to make new phyla of animals. Obviously, that alone wasn’t enough, of course.

I’m interested in your statement that “it is not that hard to evolve new genes.” According to the new paper by Paps and Holland, the new genes would have covered “the spectrum of classical functions linked to animal multicellularity: gene regulation, signalling, cell adhesion and cell cycle” (p. 5).

Would you regard the appearance of thousands of new genes in the animal lineage over a period of 40 million years (say) as particularly remarkable, Joshua? I’d like to hear your professional opinion on this point, because as a layperson, I don’t have any real idea how fast one would normally expect these genes to appear, anyway.

@swamidass:
Marshall is also correct that it is not clear that new folds are required (and the referenced paper does not delve into this).

I agree with you here. Meyer needs to make a solid case for this claim.

@T.j_Runyon:
Animals didn’t originate in the Cambrian. So bringing up Marshall and the Cambrian Radiation is irrelevant.

It’s true that the very first animals appear to predate the Cambrian by about 100 million years. However, as I pointed out above, new genes didn’t only appear with the origin of animals: “1189 homology groups are necessary for the origin of Metazoa, 494 novel HGs are required for the origin of Eumetazoa (sponges + Planulozoa + Bilateria), 1201 novel HGs are needed for the origin of Planulozoa (ctenophores, placozoans, cnidarians + bilaterians), and an additional 1580 HGs are required for the origin of Bilateria.” Bilateria seem to have appeared about 555 million years ago, right before the Cambrian. And I wouldn’t mind betting that new genes appeared with the origin of animal phyla in the Cambrian, too. The key question, however, is: are new genes hard to get? @swamidass appears to think not, so I’d like to hear his explanation why.

@gbrooks9:
In other words, if God is the one invoking changes as needed, God-guided evolution can be as fast as it NEEDS to be.

Good point.

By the way, can anyone provide me with a good definition of “homology group” in biology? I haven’t been able to find one online. The nearest I can find is the following statement in Paps and Holland’s 2018 paper:

Hence, each HG [homology group] defines a set of proteins that have distinctly diverged from others; it may contain one ortholog (gene family) or multiple paralogs (gene classes or superfamilies), regardless of their mechanism of origin and traditional classification. We believe this assignment is relevant from an evolutionary perspective as it highlights groups of proteins that are highly diverged from others.

This is a subjective term on several levels. What is “a lot” vs. “a little”? What is a “new” gene vs. a modified version of an existing gene?

Except it clarifies the next adverb with “largely”. As good biologists do, he is hedging, talking about the rule, while verbally acknowledging the exception. That is how biologists talk. He is saying the dominant process is not new genes, though there may be new genes formed.

Honestly, I do not find this remarkable. I mainly find it poorly specified. How is “new” being defined here?

That critical ambiguity aside, we are talking about a whole friggen planet of single celled organisms and small multicellular organisms, it is hard to understand how there wouldn’t be new proteins coming into existence, both de novo an by modifying existing proteins. I’m just not sure what the objection is. 40 million years is a long time, and has already been pointed out, there were animals before the Cambrian explosion. I’m not seeing the problem here.

The paper you are quoting is using squishy laungage. Properly speaking, we do not know how many HGs are required for early Bilateria. We expect early bilateria are going to be much less complex than modern bilateria. You do not get to minimum changes required from analyzing extant species. That is an equivocation.

I’d expect that there are some new genes. It is not clear how many are required.

That is a squishy definition. Homology is a real concept, but it is a fuzzy concept, which includes a large range of qualitative and quantitative definitions. It is usually not clear from looking at data where one protein family begins and the other starts. That is why one should never build an important argument on a qualitative understanding of “homology groups” or folds. Those arguments are all underspecified. We need quantitative definitions to make sense of this.

Meyer does not provide that level of rigor to his work, so there is no way to know to what extent his definition maps to the new paper or not.


That being said, there is a lot of discomfirmation bias against ID. Sometimes, however, they forget that the public performance of a debate is beside the point. They need to better couch their statements in quantitation to even understand what they mean. Otherwise, it seems like:

  1. For measuring the number of de novo HG’s that did arise in history, use a definition that maximizes the number.
  2. For measuring the number of de novo HG’s that that natural processes can produce, use a definition that minimizes the number.

That is just an equivocation in the end. You have to compare apples to apples to have a valid argument. However, that squishiness is useful if we are making a rhetorical argument. Scientists pick up on the difference quickly, and we do not like the duplicity in definitions like that. I’m not saying that Meyer did that here, but that certainly is a pattern in ID arguments.

1 Like

I don’t even see the point of discussing this if we are all committed to the idea that God leads mutations (either from an I.D. perspective, or from a @swamidass perspective). This is not where we should be focusing our efforts.

@vjtorley, What would you consider to be the key differentiation that we should be examining?

im sure some new genes were required for new phyla. But it was primarily driven by the rewiring of GRNs. Marshall is dealing with the origin of new body plans. Not the origin of the groups dealt with in this paper. Which all predate the Cambrian (molecular and fossil evidence). So I’ll say again this paper is pretty irrelevant when it comes to the Cambrian radiation and Meyer’s and Marshall’s viewpoints on it. Dealing with two different things. Also, I’m not too surprised these groups required lots of new genes.

1 Like