john hawks weblog

There is a good case to be made that distinguishing neutrality from selection is now the central problem of molecular evolutionary biology. I don't intend to make the case, but I do want to discuss the problem. It arises for me because of a recent discussion of human-chimpanzee differences in gene expression, in a Science paper by Philipp Khaitovich and collaborators.

Reading this paper and some of its references has made me realize that one of the key aspects of the problem is that evolutionary biologists and molecular biologists often don't speak the same language. Sometimes the two groups may use exactly the same terms to mean different things --- yet, because they are mostly words borrowed from English, the meanings are similar enough to cause immense confusion.

These are my notes on gene expression differences between humans and chimpanzees. My focus is in pointing out things that might confuse, and attempting to determine the importance of the work to the project of uncovering the events and processes of human evolution. In that spirit, this is not a critique in any way, although I do include some critical comments; they are indications of the way this work differs from other kinds of evolutionary biology.

So, what do I mean when I say the meanings of words appear to be different? Consider the title of the paper:

Parallel Patterns of Evolution in the Genomes and Transcriptomes of Humans and Chimpanzees

This seems quite clear to an evolutionary biologist: humans and chimpanzees both evolved in a common direction from an ancestor that was different from both of them. At least, that's the usual meaning of "parallel evolution".

But in fact, the abstract makes clear that something very different is meant by the term "parallel" here:

The determination of the chimpanzee genome sequence provides a means to study both structural and functional aspects of the evolution of the human genome. Here we compare humans and chimpanzees with respect to differences in expression levels and protein-coding sequences for genes active in brain, heart, liver, kidney, and testis. We find that the patterns of differences in gene expression and gene sequences are markedly similar. In particular, there is a gradation of selective constraints among the tissues so that the brain shows the least differences between the species whereas liver shows the most. Furthermore, expression levels as well as amino acid sequences of genes active in more tissues have diverged less between the species than have genes active in fewer tissues. In general, these patterns are consistent with a model of neutral evolution with negative selection. However, for X-chromosomal genes expressed in testis, patterns suggestive of positive selection on sequence changes as well as expression changes are seen. Furthermore, although genes expressed in the brain have changed less than have genes expressed in other tissues, in agreement with previous work we find that genes active in brain have accumulated more changes on the human than on the chimpanzee lineage.

Now I'm confused. There is no evidence for parallel evolution here; rather, the expression profiles in the two species would seem to be parsimoniously explained by homology. In other words, there is nothing to suggest that the common ancestor of humans and chimpanzees had different expression levels in those tissues. Scanning the article, the word "parallel" is later used as a synonym for "similar"; and "the parallelism between sequence evolution and expression evolution" just means that both genetic divergence and expression differences depend on the tissue where genes are expressed. So "parallel patterns of evolution" is not about common pathways of selection, but instead a sort of tissue-dependent rate heterogeneity.

If that were the only thing confusing me, I wouldn't bother writing about it. But reading this stuff, I'm having constant Inigo Montoya moments: "You keep using that word. I do not think it means what you think it means."

Here's a passage from the conclusion:

In summary, we find that the patterns of evolutionary change in gene expression are largely compatible with a neutral model, in which different levels of constraints acting in different tissues add up for single genes (Khaitovich et al. 2005: 1853).

Of course, usually a neutral model is one in which there aren't any constraints...since the source of these constraints is selection.

But here, the "neutral model" is applied only to the changes in gene expression that did happen, not the changes that didn't. In a way, this distinction is analogous to Ohta's usage of "nearly-neutral" evolution: Lots of slightly deleterious changes in gene expression may have been precluded by selection, and only the truly neutral ones actually happened.

That's fair enough as it goes. And Khaitovich et al. (2004) make clear that the neutral model for gene expression is intended as a null hypothesis. It gives clear predictions for the differences we ought to expect, and in fact the differences that we observe do fit those predictions. So the burden is on an alternative explanation to explain the data better than drift. Until such an alternative surfaces, we are perfectly justified to say that humans and chimpanzees are different in gene expression mainly because of neutral evolution.

But we should recognize that it is a very particular interpretation of "neutrality": one that gives a substantial role to selection. This may help explain another bit of confusion; why Khaitovich et al. (2004) include this:

In fact, even at the level of morphology, it has been argued that many features are not adaptive, but instead result from physical constraints or historical accidents (Gould and Lewontin 1979) (Khaitovich et al. 2004:e132).

This is a citation to "The spandrels of San Marco", but one that is misapplied. Gould and Lewontin's (1979) arguments about the limits of adaptation concern the proper definition of adaptations and the logic for inferring past selection. In their paper, they consider ways that features of organisms may actually reflect selection on related features or structures, elements of architectural necessity, or genetic drift. But Gould and Lewontin did not give any examples of characters that evolved by drift, nor even mention the role of drift directly on morphological characters. Their most interesting argument is that for polygenic traits the level of selection per gene may be so slight as to allow many nonoptimal alleles to be fixed. The "physical constraints or historical accidents" parts of the paper may accord with the spirit of the gene expression work, but they are not about neutrality or genetic drift -- they are about interpreting the proper role of selection.

The key is that many things from other fields may look alike, or seem to be analogous, but that doesn't mean that they are alike or analogous. That's why it is important to use terms precisely, because sometimes they don't mean what you think they mean.

Finding selective constraints

Delving into the papers, it is possible to find some facts about gene expression in humans and chimpanzees that may help clarify the role of expression in the evolution of human characteristics. But the statistical test of neutrality makes clear that the goals of the molecular biologist and the evolutionary biologist are often very different from each other. Both may be interested in evidence of constraints, and these are not too problematic to identify:

Our analyses show that each tissue is associated with a certain level of evolutionary constraints acting on the genes expressed in it -- for instance, brain imposes more constraints than liver. These constraints add up across tissues so that genes expressed in many tissues are subject to more constraints than are genes expressed in few tissues. The signatures of these constraints are seen both at the level of DNA sequence differences and at the level of expression differences (Khaitovich et al. 2005:1851).

No problem: the evolutionary change in both coding sequences and gene expression depends on what kind of tissue the genes are expressed in. The pattern of selection on genes acting in the brain has kept them more similar between chimpanzees and humans than genes that are active in the liver. This is true to an even greater extent for genes expressed in many tissues. These genes are much more similar between humans and chimpanzees than are genes expressed only in the liver.

Notice, I said "pattern of selection" instead of "level of evolutionary constraints". I think this distinction makes a bit of difference when considering the passage immediately following the last one:

We have recently suggested that the evolution of gene expression patterns largely conforms to the predictions of a neutral model of evolution (23), i.e., that most expression differences observed within and between species are selectively neutral or nearly neutral. Because most evolutionary changes in nucleotide sequences conform to a neutral theory (24), the parallelism between sequence evolution and expression evolution observed here supports the notion that most evolutionary changes in gene expression are similarly selectively neutral or nearly neutral (23) (ibid., references in original).

This doesn't follow. They have found that the pattern of selection is similar in humans and chimpanzees. They conclude that changes separating humans and chimpanzees are therefore not the result of selection. Again, I'm confused.

Citation number 23 from the passage above is a previous paper by Khaitovich and colleagues (2004). In it, the authors demonstrate that much evolution in gene expression does fit the expectations of neutral evolution. The tests they applied are listed in the abstract to the paper:

(1) expression differences between species accumulate approximately linearly with time; (2) gene expression variation among individuals within a species correlates positively with expression divergence between species; (3) rates of expression divergence between species do not differ significantly between intact genes and expressed pseudogenes; (4) expression differences between brain regions within a species have accumulated approximately linearly with time since these regions emerged during evolution. These results suggest that the majority of expression differences observed between species are selectively neutral or nearly neutral and likely to be of little or no functional significance (Khaitovich et al. 2004:e132).

Khaitovich et al. (2004) show all of these propositions to be approximately true among hominoids.

What does this mean? The motivation for the neutral theory was the discovery of abundant electrophoretic variation in natural populations. This appeared to contradict the established principles of genetics, which had been based on the assumption that most natural populations are monomorphic for a "wild-type" allele, and that polymorphism is rare. In contrast, electrophoretic studies found that polymorphism is ubiquitous. Under the assumption that this polymorphism is not reflected at the level of morphology or behavior, the neutral hypothesis proposed that the polymorphism has no phenotypic importance, and therefore no possible correlation with fitness.

Applied to gene expression, we might make a prediction: that the variability in gene expression (i.e. polymorphism) is great, and that many changes in gene expression have no fitness effect. The actual extent of gene expression differences in natural populations (of humans and chimpanzees) is not obvious in these studies -- for the most part, differences within and between these species is concealed within cluster diagrams, which are based on distance formulae.

But if we look at the diagram presented in Khaitovich et al. (2005:1851), a conclusion emerges:

Figure 1 from Khaitovich et al. (2005:1851)

The study concludes that genes expressed in testis were likely under positive selection because of the large between-species divergence compared to the small within-species divergence.

But notice that only in this case does the ratio of within-species to between-species divergence look anything like that for gene sequences! For the other tissues, the between-species divergence is incredibly short compared to the amount of divergence between individuals within species. In contrast, most gene sequences show around a tenfold greater branch length between humans and chimpanzees than within either species.

This problem is discussed by Khaitovich et al. (2004):

The fact that the overall accumulation of expression differences conforms to a selectively neutral model does not mean, of course, that all expression differences between species are selectively neutral. As for nucleotide changes, some changes in gene expression will have had phenotypic consequences and some of these will have become fixed due to positive selection. To identify such gene expression differences, we propose to use the ratio of divergence between species to diversity within species, akin to the tests suggested for quantitative genetic traits (Charlesworth 1984; Lynch and Hill 1986; Turelli et al. 1988) and in agreement with recent suggestions by Rifkin et al. (2003) or Hsieh et al. (2003). However, to do this it is necessary for each gene considered to distinguish the gene expression diversity caused by genetic differences between individuals from the diversity caused by environmental factors. This is crucial since the environmental component is likely to be much larger than the genetic component. For example, under strict neutrality and no environmental influence, we expect a divergence to diversity ratio that is equal to the ratio of time of divergence of the species to the average time to the common ancestors of the individuals sampled within a species. This would be about 1:10 for humans and chimpanzees (Chen and Li 2001; Lander et al. 2001). However, the observed ratio is approximately 1:3, suggesting that the environmental component is on the order of three times bigger than the genetic component. Studies of gene expression differences among individuals with different genetic relatedness will eventually allow an estimation of the genetic component of expression variation.

Khaitovich et al. (2004) attempted to use pseudogene expression as a neutral test of the ratio of between-species divergence to within-species diversity. But this is clearly an error, since the expression of these pseudogenes must depend on other factors (e.g. promoters and inhibitors, which are coded by active genes). It makes no difference that the expressed pseudogenes themselves do not affect fitness; whatever controls their expression may well affect many other (non-neutral) things, including the expression of active genes. Thus, pseudogene regulation might well be expected to be very much like the expression of active genes -- if indeed the pseudogenes' expression were controlled by unique-to-pseudogene promoters, then we might not expect them to still be expressed at all.

This improper use of pseudogene comparisons may really have led them astray on the interpretation of testis-specific expression changes. If the ratio of between-species to within-species divergence is closer to 10:1, then it might mean that the environmental influence on testis-expressed genes is lower.

In any event, it would appear that the extent of gene expression differences within species is very great indeed. The similarity in between-to-within species divergence ratios among tissue types shows that the pattern of selection among tissue types may be similar. But the relatively low between-species divergence shows that far fewer gene expression changes become fixed within species than occur within species. Some of this difference may be the consequence of environmental difference --- but it should be noted that environments vary between species as well as within them. From the great environmental differences between humans and chimpanzees, we might well expect between-species gene expression differences to be disproportionately great, not disproportionately minor. So a better explanation would appear to be strong selective constraints on most within-species variants.

Of course, many of the differences between humans and chimpanzees in gene expression may have no fitness consequences. If this number of neutral changes is very large, in fact, it will make it very difficult to find evidence for adaptive changes in gene expression.

Does a long interspecies branch length indicate positive selection? If we could control for environmental effects, it might. But this would be a very course test of selection --- it would require that the number of positively selected (i.e. adaptive) changes be very great -- perhaps more than the number of neutral changes. Here, we are looking at the expression levels of all genes (or at least a substantial proportion of them) within a tissue. Even if the tissue underwent substantial changes during human evolution, it is probable that these changes involved a relatively small proportion of all the genes that are active in a tissue.

How do gene expression differences relate?

At this point, it's worth thinking about what we aim to explain. I am interested mainly in three things: how selection caused the evolution of humans from an ape ancestor; whether the absolute number of adaptive changes in the human genome was high enough to substantially affect nucleotide diversity; and the extent to which human anatomical and behavioral evolution may have depended upon regulatory changes versus protein structural changes.

As far as the number of selected changes in expression, the proportion of selected changes is not that informative. If the total number of changes (selected and neutral) changes is very large -- as it appears to be -- then a very small proportion of expression changes would still be a substantial number of changes. And it still remains possible that a relatively high proportion of differences between species were selected, even if there is a high level of within-species diversity, because of the unknown role of environment. And the same gene might easily have been under selection for its expression multiple times, or different times in different tissues. So there is far to go to figure out the role of adaptive change in expression on overall gene diversity.

The extensive diversity in gene expression within species is very interesting from the perspective of regulatory vs. protein evolution. It may be that there is more extensive diversity among people in gene expression than in protein structure, and the fact that chimpanzees show a similar pattern may suggest that it is generally true. The role of selection depends on how heritable such variation in expression actually is, but considering the extensive variation and its continuous nature, it may be that gene regulation forms a much more facile substrate for adaptive evolution than does amino acid sequence change.

One question is whether gene expression differences say very much about what we are interested in explaining about human evolution. For example, there is this passage from Khaitovich et al. (2004), concerning expression differences in different regions of the brain:

Our data show that tissues that diverged recently have very similar gene expression profiles irrespective of the differences in function. For instance, the transcriptome of Brodmann's area 44 in the left hemisphere (Broca's area) is very similar to that of the prefrontal cortex in both humans and chimpanzees, although it is known to be involved in speech processing in humans while it must have another function in chimpanzees (Kandel et al. 2000). This is what we would expect if the time since divergence rather than the extent of functional differences determined the magnitude of transcriptome change. Thus, although a number of expression differences between brain regions surely correspond to functional differences, our findings suggest that a sizeable proportion of the differences are functionally neutral.

Of course, they are not saying that there are no differences in gene expression between humans and chimpanzees in Broca's area; they are merely saying that the scale of differences in this area does not differ from other areas of the brain. So there clearly could be differences in gene expression that have functional importance in this region.

But what if there weren't? A real possibility is that the important differences between humans and chimpanzees lie in the circuitry of this region, and not in the function of the neurons themselves. Indeed, the expression profiles of the neurons might be entirely identical for all we know, and the key differences might lie in the embryology of the developing neural circuits. These embryological differences themselves would be the product of differences in gene expression, but only at a particular stage of ontogeny.

Clearly we need more than a one-dimensional account of expression differences. The evolutionary differences between humans and chimpanzees are determined by gene interactions that have a time component. What's worse, these depend on the interactions among developing (and differentiating) tissues, so that the in vivo differences in expression may not be easily modeled with in vitro methods.

All this is to say we still have a lot to learn about gene expression in human evolution. Also, it is clear that different kinds of biologists need to read more of each other's work. The lack of familiarity with the use of common words really has the potential to lead to confusion. Fortunately, that sounds the same in all kinds of biology.

References:

Khaitovich P, Hellmann I, Enard W, Nowick K, Leinweber M, Franz H, Weiss G, Lachmann M, Pääbo S. 2005. Parallel patterns of evolution in the genomes and transcriptomes of humans and chimpanzees. Science 309:1850-1854. Full text online

Khaitovich P et al. 2004. A neutral model of transcriptome evolution. PLoS Biol 2:e132. Full text (free)

I'm reading through the recent review article on "The use of racial, ethnic, and ancestral categories in human genetics research," by the "Race, Ethnicity, and Genetics Working Group" of the National Human Genome Research Institute. Very official sounding, or anonymous sounding, although the participants in the working group are listed.

How is it? Here's an indication: I count 205 references. By my count, 11 of these are earlier than 1990.

Hmm...does that mean it's more up-to-date?

References:

Race, Ethnicity, and Genetics Working Group. 2005. The use of racial, ethnic, and ancestral categories in human genetics research. Am J Hum Genet 77:519-532. Full text

A short paper in PLoS Biology by Thomas Breuer and colleagues describes the first two observed instances of tool use in wild gorillas. Reuters has reported on the research as well.

From the Reuters report:

They describe the two instances in the northern rain forests of the Republic of Congo.

"We first observed an adult female gorilla using a branch as a walking stick to test water deepness and to aid in her attempt to cross a pool of water at Mbeli Bai, a swampy forest clearing in northern Congo," Breuer and his international colleagues wrote.

In the second case, they saw another pull up a dead shrub.

"She forcefully pushed it into the ground with both hands and held the tool for support with her left hand over her head for two minutes while dredging food with the other hand," they wrote. "Efi then took the trunk with both hands and placed it on the swampy ground in front of her, crossed bipedally on this self-made bridge, and walked quadrupedally towards the middle of the clearing."

Since gorillas use tools in captivity, they have long been known to have the cognitive capacity to do so; not unexpected since chimpanzees and orangutans both use tools in the wild and captivity. It seems clear that at least the basic cognitive potential for tool use is very ancient in hominoids, dating at least to the last common great ape ancestor, and quite possibly much earlier.

So the evolutionary question is what causes some lineages to develop this potential in practice, and what causes other lineages to retain more latent abilities. How active a role does ecology have, versus social dynamics? And how much tool use is facilitated by dedicated brain circuitry, versus more generalized cognitive processes that may be adapted to other behaviors?

Observing anything in wild gorillas helps to answer these questions. If gorillas use tools with good facility in captivity, but never, ever, used them in the wild, then it would be a good reason to think that their "cognitive adaptations" for tool use were not particularly adaptations to tool use at all. In this case, the cognitive underpinnings of tool use in great apes would likely reflect adaptations involving other behaviors.

More observations of tool use in wild gorillas reopens the question. Perhaps hominoid brains do have adaptations specific to tool use (or seeing external objects as possible tools), and those come into adaptive importance from time to time.

In Nature this week there is a short but interesting review by political scientist Diane Paul (University of Massachusetts) of the new book Race to the Finish: Identity and Governance in an Age of Genomics by Jenny Reardon. The book is about the rise and fall of the Human Genome Diversity Project.

As Reardon tells it in this engrossing and even-handed book, the scientists never knew what hit them, and so were unable to mount a response to the project's detractors. The scientists involved believed they were in a race against time to answer compelling questions about human origins and migrations. But the peoples on whose cooperation the project depended -- or at least those claiming to speak for them -- were not interested in the scientists' questions about human origins (to which they already had satisfying answers), disliked being thought of as a resource, took umbrage at the assumption that they were vanishing, mistrusted the project leaders' motives, especially in regard to patent issues and, in general, did not see what was in it for them (Paul 2005:621).

The review is generally positive, ending on this note:

In the event, the critics stopped the project in its tracks. Reardon sees little to celebrate in this victory. The project's proponents correctly predicted from the start that, if they failed, the research would continue but in a much less public and organized way. The study of human genetic variation is now fashionable, but it is being pursued without scrutiny of the deeper issues that Reardon believes essential to the pursuit of both a more reflective science and a more sensitive society. Funders have understandably tried to avoid the controversies that sank the Diversity Project. But the ironic result has been to narrow discussion of the issues at stake even further (Paul 2005:622).

That's certainly the case: the same research is still happening, just outside the attention zone of HGDP opponents. But some of it has returned to the surface, especially the Genographic Project, which has already seen objections from indigenous peoples. Every time the cycle comes back around, the technology is faster and cheaper, it is easier for individuals to defect from the decisions of tribal elders, and there is a greater genomic context to compare results from different populations. Especially now that that Human Genome Project has achieved its result, sampling human diversity is the next logical step. If geneticists weren't doing it, they'd just be spinning their wheels.

References

Paul D. 2005. Diversity and controversy. Nature 437:621-622. Full text (subscription)

I didn't have much to add last week. Although I did have the same question as this week: Doesn't this fictional D.C. have a medical examiner? I mean, both of these (the bomb victim; the mummified hanging victim) are cases that a forensic anthropologist might well get called in on, but not until the ME (or coroner) had performed the legalities (like signing a death certificate) and released the body to the Jeffersonian Medico-legal lab. I guess skipping that part helps tighten up the plot (and gives a good reason for Brennan to pal around with a hunky FBI agent.

I can appreciate the catch-22 between declaring a homicide or not, but this is exactly the sort of thing that will get you thrown out of a lot of future court cases. I guess it's lucky that Brennan hasn't had to go to court yet!

OK, now it seems to me that actually watching all the tapes is something you could have the minor characters do, to give them a reason for being.

Another week without much for the anthropologists. Maybe it's time for a Civil War crime to surface.

The AP is reporting on the discovery of a double newborn burial near Krems, Austria. The remains are estimated at 27,000 years old, and were buried directly side-by-side along with a string of 31 beads and mammoth bones.

Not much detail in the story, although there is this:

Archaeologists are combing the area to see if the infants' mother is nearby, as giving birth to twins in that era would have been extremely difficult and potentially fatal.

Not to mention for the babies!

The paper by Guatelli-Steinberg et al. (2005), earlier referred to here, is now available online from PNAS.

The results are basically as reported by National Geographic, finding that Neandertal anterior teeth have perikymata counts within the range of living human populations. Perikymata are microscopic ridges on the enamel surface of teeth; they mark the incremental growth of the teeth over small periods of time. The idea has been that these ridges work a bit like tree rings; they mark the amount of time that the tooth took to grow. However, as this study indicates, the formation of perikymata is not quite so simple as the addition of tree rings, and human populations actually vary substantially in the number of perikymata on their teeth.

What makes this different from earlier work (like Ramirez Rossi and Bermudez de Castro 2004) is the inclusion of an African sample. The very low perikymata count of the recent Africans significantly extends the range, which had previously been assessed in Europeans only. Thus, the conclusion here is that there is no evidence from perikymata to indicate that Neandertal development was any different from that within living human populations.

Now we can wait for the other shoe to drop:

The finding from the African population sampled here shows that some developmentally normal humans have much lower perikymata counts than others. This varies by tooth (since they don't all develop for the same time): the lower canines have the highest counts, with a mean over 150; the lower incisors have the lowest counts with a mean down near 100. Remember that these values are means; individuals in the sample must have scored lower, although the range of the sample is not reported in the paper.

With this sample, the human range encompasses the Neandertals. It encompasses all the earlier European hominids (chiefly from Atapuerca) sampled by Ramirez Rossi and Bermudez de Castro (2004), because these hominids had counts higher than Neandertals.

Let's take a look at Dean et al. (2001:628), who give values for earlier hominids. Here's a table including some earlier hominids along with the South African values from Guatelli-Steinberg et al. (2005). The current paper does not include the numbers, so I am reading estimates off the figure, but considering they are means and the important aspect is the total range, the numbers aren't critical. Lower numbers are less like the recent Europeans that were the standard before the new comparative work.

From these numbers, Sangiran 4 and SK 27 are within the range of modern human population means. So are three of the teeth of Australopithecus (i.e. A. africanus), and the remaining three teeth are pretty close, so that it seems likely the A. africanus dentition wasn't very different in its perikymata number from the range of living Africans.

The standouts are KNM-WT 15000 and Paranthropus (i.e. A. robustus). A. robustus is easy to explain: its anterior teeth are a lot smaller than ours. A lot smaller. If enamel formation rates were similar, then they ought to have taken less time to form, regardless of other aspects of somatic development.

The puzzle is KNM-WT 15000, the famous Nariokotome skeleton. Is this skeleton a normal representative of early human populations? Is it at the extremely low end of a normal range including others like KNM-ER 1590 (also a bit smaller than the mean, although probably not outside the range of living Africans)? Is it pathological?

The other shoe is the research paper that will cover all these questions.

Now it could be that these numbers really aren't comparable for some reason; I don't do perikymata, but I can tell that the counts depend on estimates of crown height and packing density, so it's not obvious that they were derived in the same way (although the papers do share one author).

But the Neandertals are far from the most interesting part of this perikymata problem. Can we tell a human from an australopithecine from these data? If so, why do some of the earliest humans have the lowest (i.e. sub-australopithecine) counts?

I think we can disregard the idea that their somatic development rates were "highly derived" in a non-human-like direction. It's not like they're Neandertals, after all.

References:

Guatelli-Steinberg D, Reid DJ, Bishop TA, Larsen CS. 2005. Anterior tooth growth periods in Neandertals were comparable to those of modern humans. Proc Nat Acad Sci USA 102:14197-14202. Abstract

Dean C, Leakey MG, Reid D, Schrenk F, Schwartz GT, Stringer C, Walker A. 2001. Growth processes in teeth distinguish modern humans from Homo erectus and earlier hominins. Nature 414:628-631.

Ramirez Rossi FV, Bermudez de Castro JM. 2004. Surprisingly rapid growth in Neanderthals. Nature 428:936-939. Full text (subscription)

A weblog is a monologue. I try to keep this in mind whenever I post on a topic where I have an opinion. If I wanted to, I could just browbeat ideas I don't like. But that wouldn't be very useful to people trying to get a general idea of the science, and it certainly wouldn't be very interesting.

Now don't get me wrong: I have certain pet peeves that lower my impulse control on the "post" button. And long experience teaching had taught me that hiding my opinion isn't honest or interesting, and is only rarely useful.

However, the presence of an opinion isn't a difference between the weblog and a journal article. Research articles certainly have an opinion, and they almost always further an agenda. What's worse, that agenda attains a privileged status in a journal, because opposing it requires more than simply pointing out its flaws --- it requires putting together a research article to overturn some part of the result. In other words, the publication process exerts a cost on scientific logic.

This cost is analogous to the cost of parliamentary process on legislation: essential items are combined with each other into "omnibus" bills that include lots of nonessentials put in to make certain legislators happy. In American tradition, these nonessential costs of doing legislative business are called "pork". This combination of stuff makes it easier to pass bills through legislatures, but it is not the most efficient use of public money.

Science is the same way. Like legislation, research articles are larded well with nonessential items: people cite their friends, try to make their limited conclusions apply to broader questions, and add irrelevant references to pet ideas. But also, research articles deliberately lack things that --- in the ideal scientific argument --- ought to have been there, like discussion of alternate hypotheses that may explain the results, relevant work in analogous fields, and (most importantly) conceptual weaknesses in Sometimes these omissions and extras are caught by peer review and amended. But it's too much trouble to catch them all, especially if they follow a well-worn agenda from prior papers. So science is full of pork.

Paleoanthropology is among the worst, because so many different stakeholders contribute to the work: anthropologists, archaeologists, geneticists, geologists, and so on. Much of the process of training our students is getting them to understand which parts of research articles are essential and which parts are pork. Just like a legislative intern, they have to learn which people are served by which concessions --- a tangled mess that is fully as ideosyncratic as bridge projects in congressional districts. So we have stories about "lumpers" and "splitters", and we keep track of who advised whom, and who worked at whose field site (or wants to in the future) and who went to which meeting. Students quake in fear about who might be on grant review panels and who might review our papers, and they include things in grant proposals and articles that they don't really think in the hopes of getting accepted. And some of all that quaking is really good for the science, because sometimes students ought to consider things they don't really think, or do tests they really don't want to do. On the other hand, sometimes it's just in the service of pork.

But for the student who doesn't have the time or inclination to work out all these intricate relationships, there is an alternative strategy: just don't read the articles at all! This is by far the most common way of working in paleoanthropology; reading nothing at all in a research article except for the abstract and the conclusion. The field is small enough that this strategy doesn't work very well --- we are pretty strongly K-selected, and students who ignore certain things do so at peril to their careers. But in genetics, there are a lot more students, a lower proportion of them will stay in the field for long, and there is a much shorter research cycle. For the most part, even the reviewers haven't read very much about human evolution. So the things that do get cited and the arguments from anthropology that are presented are highly idiosyncratic. It can be hard for an anthropologist to understand why what seem like essential issues have been completely ignored. When this research comes back into anthropology, to support ideas in an anthropological article, the context of all that pork barrel thinking is completely lost. All that is left is the impression that genetic data "support" some hypothesis about human evolution, without consideration of all the alternative hypotheses that have been left out of the discussion.

The long publication process in journals blunts criticism. Some journals don't even accept letters to the editor (!), and those that do so tend to limit their publication to methodological comments rather than smaller issues. So it is effectively impossible to cut the pork. This has certain advantages for authors, in that it lets them publish things they can't really support. It has clear disadvantages for comprehension, because people outside the process cannot evaluate the basis for conclusions. There is no transparency, in other words --- almost as if paleoanthropology were a priesthood instead of a science.

Don't misunderstand me; if I didn't think paleoanthropology was science, I wouldn't do it. And I think we are, if anything, more critical of certain kinds of logical inference than most other fields --- mainly because the limited nature of our evidence necessitates it. How many other parts of biology have had as detailed an exploration of the nature of species as paleoanthropologists?

But certain kinds of inferences go a long time in this field without being challenged, even when the evidence that supports them is weak, absent, or (worse) circular. How many articles have cited the mtDNA coalescence time as evidence about human origins without any mention of the word, "selection"? After six decades of research on the species problem, how many still make conclusions about reproductive isolation after applying species concepts not based on reproduction?

In a year or so of writing the weblog, I've learned that I can't get away with pork barrel thinking. When I write things that I can't support, my readers let me know, usually within a few hours. This is as true of minor errors as it is of major misinterpretations. Sometimes you will see rapid follow-up posts here, or updates on posts. These almost always come from reader comments. As critically as I treat ideas that I discuss, you can believe that readers are treating me just as critically. No one in science holds opinions lightly, but mine are beholden to a very large sample of critical readers -- many professionals -- who rejoice in letting me know when I'm sloppy. That means that I have to be transparent: I quote actual text, and link to original articles, so anyone can compare what I write to what the original article says.

I'm not writing up new research results here, or breaking any news. What I'm doing is reading things with due diligence, reflecting on results, and pointing out logical pork where I find it. It is already having an effect: it has informed my research, brought attention to papers in tangent fields that bear on human evolution, and has rattled a few cages. I think that this kind of work will only grow in importance: as paleoanthropology spreads to encompass the genome as a whole, there will need to be much more coordination of different fields with each other.

With your help, and especially your continued reading, I've created a new kind of resource. I just want to say I feel incredibly privileged to have such an active and interested readership.

The BBC ran a show tonight (Thursday Sept. 22) on the Liang Bua discoveries from Flores; meanwhile BBC News is reporting a few more details about the pathology claims:

Jacob was soon joined by a handful of researchers in the belief that the discovery team had happened upon nothing more than a member of our own species with a rare disease.

Professor Bob Martin, one of the team that is set to publish new evidence challenging the discovery team's original interpretation, says the Hobbit's brain is "worryingly" small and contradicts a fundamental law of biology.

...

Ann MacLarnon of Roehampton University, UK, has discovered the skull of a microcephalic in the vaults of London's Royal College of Surgeons with a brain that matches that of the Hobbit for size.

"It showed that we really could demonstrate with a specimen that [microephaly] could explain the Hobbit's small brain," she told Horizon.

Along with the others who have already come out publicly, like Maciej Henneberg and Alan Thorne --- and of course Teuku Jacob --- this is starting to seem like a rather large team of experts arrayed on the anti-floresiensis side.

"Set to publish new evidence" sounds good; we should see this coming out soon.

Meanwhile, there is the problem of the second mandible:

"Let's buy into [the sceptics'] argument just for a bit of fun," said Professor Bert Roberts of the University of Wollongong, Australia, a member of the discovery team.

"We've got a complete lower jaw that's identical to the first so there we have a situation where we've now got to have two really badly diseased individuals.

"We've got a diseased population like some sort of leper colony, living in Liang Bua 18,000 years ago. The probabilities have got to be vanishingly small."

This may become the most intensely studied pair of jaws ever. Are they really so similar? Remember that the second jaw hasn't yet been published. From the pictures, it looks if anything a bit smaller than LB1, and strange --- although not in precisely the same way. Are they both "badly diseased individuals"?

I'd say it's at least as likely as that skeleton being normal.

Modularity is a property of biological organization: organisms are composed of subunits that perform different functions. At the cellular level, the cell is composed of organelles that have different functions in protein assembly, metabolism, growth, and homeostasis. This organization is reflected at the level of DNA, which consists of sequences organized into functional subunits: coding regions, introns, promoter regions, multigene complexes, and ultimately chromosomes. It characterizes the anatomy of multicellular organisms, which are divided into organs such as hearts, lungs, ganglia, and eyes, or leaves, stems, flowers, and roots.

And modularity underlies the organization of the brain. Mammalian brains are divided into different parts --- neocortex, cerebellum, thalamus, etc. These parts contribute differently to different functions, as do the subcomponents of each of the parts. The neocortex is comprised of tissues that contribute to different tasks, such as Broca's area for speech and the

Brains do a lot of things that are analogous to computers, in terms of information processing. Indeed, the things that our brains are really good at are increasingly being done by computers, from adding and subtracting to face recognition. Our brains certainly have talents that are a challenge for computers, and vice versa. For one thing, our brains are well-motivated for the most part while computers have no motivation at all. But they are broadly similar in their ability to take in and manipulate information: a computer is more similar to our brains than, say, to our muscles.

But today's computers are not modular in the same way our brains are modular. Computers have different components, that is true --- they have memory chips and disk drives and power supplies and one or more central processing units (CPUs). But the great strength of computers is that the same CPU can be programmed to perform any algorithmic task. That's the principle of the universal Turing machine: a certain kind of simple device can --- by providing different instructions --- perform any transformation on data that we could think of. Indeed, if we had the right software we could cause today's computers to act like humans; they would just be really, really, really slow.

This provides a hint about why we would want the brain to be modular, rather than to provide different instructions to a single small but fast CPU. Differrent brain tissues can perform different functions at the same time, so that we can very quickly recognize someone's face, remember where we know her from, contort our faces to an pleasant expression of recognition, and say "Hello." Specializing each of these tasks into different tissues allows us to perform them much more quickly.

But that doesn't answer the question of how mental modularity evolved in the first place. After all, at the time that face recognition became important in primate evolution, our primate ancestors weren't doing all the things that we do now. Why couldn't they have simply evolved a single system that added new capabilities as it went along? This kind of mega-system would seem to be the natural consequence of evolution: tinkering slightly with a pre-existing function ought to be much easier than adding whole new modules from scratch.

Now, I should point out that we really don't have a lot of detail about how fine-grained the modularity of the brain actually is. It might very well be that a lot of what we do actually is cobbled together into a few mega-systems. But there is abundant evidence that a lot of cognition is actually very domain-specific: different neural circuits exist to perform discrete tasks.

A new paper by Nadav Kashtan and Uri Alon of the Weizmann Institute finds a reason why modularity might commonly evolve. Here's the abstract:

Biological networks have an inherent simplicity: they are modular with a design that can be separated into units that perform almost independently. Furthermore, they show reuse of recurring patterns termed network motifs. Little is known about the evolutionary origin of these properties. Current models of biological evolution typically produce networks that are highly nonmodular and lack understandable motifs. Here, we suggest a possible explanation for the origin of modularity and network motifs in biology. We use standard evolutionary algorithms to evolve networks. A key feature in this study is evolution under an environment (evolutionary goal) that changes in a modular fashion. That is, we repeatedly switch between several goals, each made of a different combination of subgoals. We find that such "modularly varying goals" lead to the spontaneous evolution of modular network structure and network motifs. The resulting networks rapidly evolve to satisfy each of the different goals. Such switching between related goals may represent biological evolution in a changing environment that requires different combinations of a set of basic biological functions. The present study may shed light on the evolutionary forces that promote structural simplicity in biological networks and offers ways to improve the evolutionary design of engineered systems (Kashtan and Alon 2005:13773).

The study simulated the evolutionary process by allowing different kinds of circuits to compete with each other. The circuits that best solved the range of problems in the "environment" were replicated into a new generation (i.e. "selection"), with a small likelihood of random changes (i.e. "mutations") to their function. They also performed a similar simulation using neural networks instead of circuits.

These experiments resulted in many different designs that satisfied the same computational goal. So the evolution in the different simulations was contingent: random factors led different simulations to different optimal solutions.

But this experiment added a novel component: sometimes the computational goals changed slightly over time. Each of the computational goals might involve several subtasks. One possibility is that these different subtasks might remain constant over time. But an additional possibility is that they might change slightly in importance relative to each other: in other words, the circuits might be presented with slightly different problems at some times than others.

Now, if this "environment" of problems to solve changed in a consistent direction over time, then we would expect the circuits likewise to evolve to solve the newer problems more efficiently.

But instead of changing the task from the beginning to the end of the simulations, Kashtan and Alon (2005) caused the tasks to oscillate over time. Sometimes one subtask might be more important, sometimes another, but there was no long-term directional change, just a steady variability in what kinds of tasks were important.

In this fluctuating environment, the circuits evolved to be modular. Changing different requirements at different times caused different subsystems to arise to solve each of these subtasks. These modular systems were slightly less efficient than the nonmodular systems that evolved to solve fixed tasks: they required additional logic gates to do the same job. But it took them a much shorter time to make the adjustment to solving slightly different problems.

You might think that this evolution was the result of the division of the computational goal into perceivable subtasks. But interestingly, when the environment encompassed goals that could logically be divided into subtasks but did not vary over time, the circuits did not evolve toward modular solutions. Here's an example:

A human engineer would easily notice the modularity in this problem and design a network that is made of two modules, one that analyzes the left side of the retina, and the other for the right side of the retina. In contrast, the structure of the evolved networks was not modular (Fig. 5b)(Qm = 0.15 0.02). As in the case of electronic circuits, fixed-goal evolution produces a nonmodular network even though the problem itself is readily decomposable into separate subgoals (Kashtan and Alon 2005:13776).

Nor was the effect simply the result of variation over time: when tasks were made to vary randomly, instead of by the emphasis of different subtasks, no modular structure resulted in the evolving circuits.

The authors discuss their results:

Why do modularly varying goals speed up evolution (in terms of the number of generations to reach perfect solution) when compared with evolution under a fixed goal? One reason that fixed-goal evolution is often slow is that the population becomes stuck in local fitness maxima. Because the fitness landscape changes each time that the goal changes, modularly varying goals can help move the population from these local traps. Over the course of many goal changes, modularly varying goals seem to guide the population toward a region of network space that contains fitness peaks for each of the goals in close proximity. This region seems to correspond to modular networks (ibid:13777).

The other element of the study is the demonstration that these kinds of modular circuits consist of similar structures repeated to comprise different modules. Such repeated elements are here called "network motifs". This is another characteristic of some biological organization, so it is very interesting:

In addition to their modular structure, the networks evolved under modularly varying goals display significant network motifs. The same motifs are reused throughout each network in different modules. Some of these motifs are also found in biological information processing networks. For example, feed-forward loops and bifans are found in transcription networks (7). Feed-forward loops, bifans, and diamonds are found in signal transduction and synaptic neuronal networks (7). In signal transduction networks (34) and the neuronal network of C. elegans (39), multilayered feed-forward patterns similar to those in Fig. 5c, are strong network motifs. An example is multilayered protein kinase cascades, in which families of kinases in each layer activate families of kinases in the next layer (34, 40, 41).

One possible explanation for the origin of the motifs in the olved networks is that modular networks are locally denser than nonmodular networks of the same size and connectivity. This local density tends to increase the number of subgraphs (42). To test this possibility, we evolved networks to reach the same modularity measure Q as the networks evolved under modularly varying goals, but with no information-processing goal (see Supporting Text). We find that these modular networks have no significant network motifs (Fig. 9). They show relatively abundant feedback loops that are antimotifs in the networks evolved under modularly varying goals. It therefore seems that the specific network motifs found in the evolved networks are not merely caused by local density, but may be useful building blocks for information processing (ibid:13777-13778).

The authors end with a discussion of how their results may apply to biological evolution, with specific biological examples, although not drawn from the brain.

References:

Kashtan N, Alon U. 2005. Spontaneous evolution of modularity and network motifs. Proc Nat Acad Sci USA 102:13773-13778.