My fish has eaten your fish

Methinks Axe is f**ed in the head.

OMG. Seriously. Like.

OK, this is not some kind of TRICK!

The "paper" is actually comparing protein structures to Chinese characters?

Seriously. This is a "paper?"

What if you turn the proteins 90 degrees in Y and 90 degrees in Z? Do they still cast Chinese character shadows?

Uh, no.

So this is the best of ID.

You're kidding, right? This is a joke. Ha ha. Got me. Right? Say it's a joke.

Wow. When I saw this on the Biologic Institute's web page I thought they were just using a really stupid analogy and couldn't possibly be fitting proteins to Chinese characters.

Awesome! Somehow they manage to make themselves look more and more stupid with every action.

Hah! I've got that 'Reality Bites' fish on the back of my car, along with my Darwin fish. I get a lot of cars honking their horns at me on a Sunday morning...

On topic: I understand that the DI's reality filter is screwy, but do they also not possess basic embarrassment filters? When you first blogged this I thought it was funny but fantastical - it's still hard to believe that people can be this willfully stupid.

Quick blogwhoring: click my name, come visit me.

So next, the Chinese are going to patent all these proteins and then what do we do?! I think Axe is just trying to yank ze kiang

I presume Axe intended that comment as a subtle reference to this stupid thing:

http://content.answers.com/main/content/wp/en/thumb/8/84/180px-Truth_fi…

I've read the paper after coming across your post here this morning, and as a protein person I think I can say that there are definitely many problems I see with this paper, many of which are found in the underlying assumptions and statements made by Axe as if they are fact.

While it has been a common assumption (with lots of evidence to back it up) that protein fold space is discrete, at least in the way that we classify it (SCOP, CATH) this is by no means settled and there have been some arguments made for fold space being continuous. Axe cherry picks his references with regards to mutation and protein stability and ignores examples of relatively simple mutational paths that have altered the functions of homologous proteins.

There is also the huge problem of whether this comparison by analogy bears any resemblance what so ever to reality, and I would charge that it doesn't. While it does try and improve upon lattice models by incorporating function, it ignores to many well known aspects of protein function that differ markedly from the "function" of structures like Han Chinese characters. For one proteins are not rigid characters, they move and flex in biological reality and the modeling we do of them as rigid structures (or with limited flexibility) is a simplification borne of computational necessity. However we do try and account for it, or minimize it as much as possible. In my opinion Stylus does no such thing. It also ignores the reality of multi-functional proteins and multi-functional folds. This ideas has been around for a long time and it is widely held that many ancestral proteins likely carried out more then one reaction but imperfectly, duplicated descendants specialized on one of the tasks with the other reactions becoming increasingly spurious. We know of proteins that exist in nature that do this.

I think there are plenty of assumptions that make the analogy here very poor but of course, like Axe's previous paper it will be trumpeted by the DI as a shining example of Creationist Research that shows evidence for Design, which of course it does no such thing.

Yeah, well, my fish drinks your fish's milkshake!

OK, if you don't mind, what, exactly, is a "spannable set"?

a subtle reference

I haven't read the paper that thoroughly, but that jumps out rather than being subtle IMO.

However we do try and account for it, or minimize it as much as possible. In my opinion Stylus does no such thing. It also ignores the reality of multi-functional proteins and multi-functional folds.

I'm no biologist, but I have to question a simulation that purports to be between homology, but then assumes that any deviation from the target means decreased fitness. (Which is hidden by a small fuzzification factor purportedly due to 'change in environment' to give a few folds with higher fitness.) It is as if the whole protein is an active site. But doesn't other parts of the protein contribute to homology as well?

Also I don't understand the part of the fitness function that has a "cost" function for codons, presumably to avoid that proteins gets longer. (Which sucks for creationists if proteins can be multifunctional, but also presumably sucks for the simulation.)

The "cost" is AFAIU referenced to a growth energy cost, not a primary selection cost. And it also constrain the functional fold space that they purport to (fully?) explore. But are there "costs" involved in selection, or is it exclusively function that counts as I've been reading it?

I'm in ur chineez charakters, disprovin ur evilooshunz.

if you don't mind, what, exactly, is a "spannable set"?

I have never heard of such a math concept.

But I can intuit what Wes means, he means that you can pass from protein to protein in sequence space by evolutionary processes.

This is an confusing terminology as a "span" in math is a range, in best case a distance for a set of points in geometry. (And here you could envision implementing a geometry or topology with a measure such as distance in folding space.)

But in fact there is already a set span, which is the (value) range of the set.

This is impressive. I can't believe this ever even saw the light of day somehow. Waiting to get home till I actually read what's been written, but based on what's already been posted, it's hard to believe that even someone from the DI would be *that* stupid.

Also, maybe it's just my own personal preference but I think it would sound better with an "Om" before the "nomnomnom" in your victory screw. I just find the noms lacking something without the Om, like the noms are just chewing but the Om is taking a bite or something...

I'm no biologist, but I have to question a simulation that purports to be between homology, but then assumes that any deviation from the target means decreased fitness. (Which is hidden by a small fuzzification factor purportedly due to 'change in environment' to give a few folds with higher fitness.) It is as if the whole protein is an active site. But doesn't other parts of the protein contribute to homology as well?

Exactly, the thing with Protein Fold Families, etc is that we have proteins with structure similar on some level (families are more similar to one another structurally then super families for instance)
but may carry out different functions. So you get situations where the majority of the "important" mutations are in or near the active site, a binding surface, etc. This simulation seems to treat the whole shape as being absolutely critical for function, which may or may not be the case.

This simulation software also ignores things like disordered proteins and sections of proteins that are inherently disordered.

The whole simulation software, and the analogy, just seem weak to me and not very relevant.

I'm working my way through Campbell and Reece -so I'm the layman ERV refers to. Although the professionals will no doubt find this paper full of miss it really appeals to me. The most difficult part of trying to become scientifically literate is concept comprehension. The remedy (to me) seems to be new/different/ or as Hopkins says: "counter, original, spare, strange" ways to discuss the pied beauty of well... proteins.

As performance art, it works.

@bunnycatc3r
In fact, ID publications in the mainstream scientific press are aimed squarely at laymen - working scientists will usually recognize the nonsense, but the mere fact of publication looks impressive to non-scientists. It's just like the BS (Behe & Snoke) paper that came out in Protein Science back in '05. Michael Lynch tore it a new one in a response article, but the Discovery Institute got to add it to their list of citations anyway.

Also - "pied beauty"?

If this paper is so bad how did it get through peer review on PLOS One?

Just breezing through the article, I had some thoughts:

God is chinese, we need to learn mandarin.

This article could be served as fish food, but an hour later, the fish will be hungry again.

I need to read it in more detail later, or maybe not. The assertions in the abstract are killing me. With laughter.

Hee!

What made me laugh is that the "Reality Bites" fish is on the back of my Jeep.

Now I love it even more.

Hee!

#19

As Coturnix explained,

This paper was reviewed just like any other, it is technically sound, but at PLoS ONE the reviewers are not supposed to judge the novelty, utility or media-worthiness (or politics) of a paper - just its technical soundness and proper writing.

If I may editorialize, papers at PLoS ONE were either previously rejected from a competitive journal (often another PLoS journal) or were too off the beaten path to be published in a typical biology journal. I appreciate not bowing to "novelty," but relevance to science ought to be considered. I wonder if Axe even bothered to send it to PLoS Computational Biology.

Seriously. This is a "paper?"

Maybe of the toilet variety........