Comment on Dr. Jason Rosenhouse “Among the Creationists” by Sean Pitman.

Just came across this. Short-short version:

1. Re: sequence space, Rosenhouse is right and Pitman is wrong. We can see this by the fact that virtually any protein can tolerate a lot of different point mutations, and retain function. Thus, every protein has a lot of stepping stones around it, and those stepping stones have stepping stones around them, etc. It is true that a lot of point mutations are “bad”, but all evolution needs is some narrow paths through sequence space, not big highways. Natural selection eliminates the bad mutations and keeps the population on the “ridges” (or paths, if you will). John Maynard Smith pointed this out back in the 1970s.

Confirming evidence: (1) virtually any protein falls into a family with a large number of other proteins, variously diverged from less than 1%, to over 50%, which often all have about the same function, and in many cases can substitute for each other. (2) These sequences have strong phylogenetic tree structure, and their change is well-modeled by a Markov chain modeling a stepwise point-substitution process. The entirety of modern bioinformatics and genomics rests on these models (BLAST, gene identification, genome annotation, etc.), so a few half-baked analogies that assume the problem is “finding a sequence all-at-once by random generation from scratch” fail to deal with the data, and will not convince anyone who knows anything about the modern field.

No one is asking for “evolution from scratch”. All that I’m asking for here are the odds of evolving something new at a various levels of functional complexity.

As far as your claim that narrow paths exist within sequence space between various steppingstones, this simply isn’t true when it comes to steppingstones that have qualitatively novel functionality. You have long been confused on this point. While it is true that there are in fact “islands clusters” of function within sequence space, where all the proteins that make up a particular island carry the same type of functionality, it is simply not true that these island clusters are connected to other island clusters with qualitatively unique functionality along narrow “ridges” between islands in sequence space (i.e., different “steppingstones”) beyond very low levels of functional complexity.

Now, is it in fact true that your argument that most proteins are flexible so that most positions can be mutated, one at a time, without a complete loss of function. However, it isn’t quite like you claim. After all, pretty much all of the letters in this paragraph could be mutated, one at a time, without a significant qualitatively loss in the intended meaning of the paragraph. However, the more and more mutations that occur at the same time, the exponentially less and less likely it is that the sequence will maintain beneficial functionality. The reason for this is because there is an “experimentally observed exponential decline in the fraction of functional proteins with increasing numbers of mutations (Bloom et al. 2005).”

“Our theory predicts that for large numbers of substitutions the probability that a protein retains its structure will decline exponentially with the number of substitutions, with the severity of this decline determined by properties of the structure. . . Our work unifies observations about the clustering of functional proteins in sequence space. . . ”

Bloom goes on to point out that,

“Experiments have demonstrated that proteins can be extremely tolerant to single substitutions; for example, 84% of single-residue mutants of T4 lysozyme and 65% of single-residue mutants of lac repressor were scored as functional. However, for multiple substitutions, the fraction of functional proteins decreases roughly exponentially with the number of substitutions, although the severity of this decline varies among proteins.”

In short, most mutations that affect a region or island cluster of thermodynamically stable sequences in sequence space are destabilizing in such a way that each additional mutation has an exponentially destabilizing effect. Obviously, this means that the vast majority of sequences in sequence space would not produce viably stable proteins. It also suggests that as sequence space increases in size by 20^N, the ratio of viable vs. non-viable sequences, not just systems, decreases exponentially.

“This is the reason why “simulations (Taverna and Goldstein 2002a) and experiments (Davidson et al. 1995; Keefe and Szostak 2001) clearly show that the vast majority of protein sequences do not stably fold into any structure (meaning the least stable folded protein is still far more stable than the typical random sequence).”

Consider also the observations of Nelson and Onuchic regarding these islands within sequence space:

Sequence space as being populated by families, each folding to a particular coarse grained structure and each surrounded by a shell of increasingly frustrated sequences. . . This produces a frustration barrier, e.g., a region of frustrated sequences between each pair of minimally frustrated families. Any stepwise mutational path between one minimally frustrated sequence family and another must then visit a region of slow or nonfolding sequences. . . In the case of real proteins, the sequences in these high frustration regions are much less likely to meet physiological requirements on foldability. . . If the requirement is sufficient, the region between two families will be completely excluded, which cuts sequence space into separate fast-folding, stable parts. This provides a mechanism for partitioning protein sequence information into evolutionarily stable, biochemically useful (foldable) subsets.”

– Erik David Nelson and Jose Nelson Onuchic, Proposed mechanism for stability of proteins to evolutionary mutations, Evolution, Vol. 95, pp. 10682-10686, September 1998

Consider also that even for much smaller proteins with sizes of only 18aa (very very low level of functional complexity) having average minimum Hamming gap distances of less than half a dozen or so between one novel island cluster and the next, consider that Cui et. al., suggest that such distances are essentially beyond the realm of crossability by point mutations.

Evolutionary explorations by point mutations may be likened to diffusion. Their extent is limited on a fragmented mortality landscape, because sequences belonging to different networks (Table 1) are beyond reach.

Yan Cui, Wing Hung Wong, Erich Bornberg-Bauer, Hue Sun Chan, Recombinatoric exploration of novel folded structures: A heteropolymer-based model of protein evolutionary landscapes., PNAS, Vol. 99, Issue 2, 809-814, January 22, 2002.

And, this effect only gets exponentially worse and worse with each step up the ladder of functional complexity.

2. Junk DNA. Genome sizes vary a huge amount between very similar organisms. Most of this difference is in repetitive sequences. Finding bits and pieces of function here and there doesn’t change this big picture. Any discussion of whether or not most of the genome is functional that fails to deal with the huge variability in genome sizes is worthless. I.e., the ENCODE claims, and the creationist claims, and Pitman’s discussion here.

It isn’t just bits and pieces of function, it is a significant proportion of non-coding sequences that appear to have functionality to one degree or another. Does this mean that all functional aspects are therefore vital to life? No, not at all. However, the notion that non-coding sequences are largely or almost entirely evolutionary garbage doesn’t make much sense. Even from an evolutionary perspective, it makes no sense to maintain junk that is expensive to reproduce and maintain, from one generation to the next.

3. Flagellum: google Panda’s Thumb Matzke flagellum. Pitman doesn’t even have his homology numbers right.

I’ve discussed this topic with you before, quite extensively. In my fairly lengthy discussion of your arguments (Link), where have I gone significantly off base with my homology numbers? – or any other key argument for that matter? The key problem with your argument is that you assume that the novel steppingstones that you list off are actually close enough in sequence space to be easily found from the perspective of the previous steppingstone. This simply isn’t true. The steppingstones that you yourself list off are far too far away from each other for random mutations to get from one to the other in what anyone would consider to be a reasonable amount of time. There simply are no known “narrow ridges” from one of your proposed steppingstones to the next to make evolutionary progress statistically feasible. Beyond this, you don’t have a single example of evolution in action between any one of your proposed steppingstones. If they are in fact so close together, as you imagine, why is there no demonstration of this?

And even if T3SS are derived from flagella, (a) this assertion rests entirely on phylogenetics and evolutionary assumptions, which Pitman et al. deny; (b) it still shows that the reduced system retains function; and (c) Type 3 secretion is just one of five or six known types of cross-membrane secretion, and at least one other one has evolved into a different sorts of motility (Type 4 secretion is related to archaeal flagella, and is apparently ancestral).

In short, you guys have to try harder to even get started with having a real argument.

The devolution of TTSS secretory systems is generally accepted in the literature. I don’t think there’s any real argument on this point. Also, I personally have no problem with arguments that evolution can remove parts from a system while maintaining subsystem functionality – like removing the engine and tires from a car while having the radio still work. This sort of thing is statistically likely. The problem is when one assumes, as you do, that this situation can be reversed – that it is just as easy to go uphill as it is to go downhill. That’s a false assumption. It is much much harder to co-opt subsystems to work together to produce a more complex system that requires a larger number of parts before its function can be realized to any useful advantage. The reason for this is because independently functioning subsystems would require numerous modifications to work together in a qualitatively new way. And, as the minimum part requirement for the larger system increases, so do the minimum number of required modifications to the subsystems at hand. And, with each linear increase in the minimum number of required non-selectable modifications, the average number of random mutations required to achieve these modifications increases exponentially.

And, that’s the problem with your argument in a nutshell. It simply isn’t statistically reasonable and it hasn’t been demonstrated or observed beyond very very low levels of functional complexity.

Sean Pitman Also Commented

Recent Comments by Sean Pitman