review of YEC 'science' paper

[Aggie]

No, it doesn't include heterochromatin, since they're using aligned sequence. (Or at least that's my reading of it.)

So the aligned sequence includes some repetitive sequences, but not all of them? I'm curious how they decide which to include.

I don't see how to get a quantitative estimate of the amount of variation from that paper. What it shows is that there is lots of variation in heterochromatin that's big enough to be seen under a microscope.

Here's the part I was referring to:

The variation per chromosome pair remained rather consistent with the percentage of (+) C bands ranging from 2% to 15% and (-) C bands ranging from 3% to 16%. When both types of variants were considered, the average variation was 8%.

Am I misreading that?

 

[sfs]

So the aligned sequence includes some repetitive sequences, but not all of them? I'm curious how they decide which to include.

They include the ones that align, i.e. that an alignment program can find a unique placement for. I think this was sequence that could be assembled to some degree, but I can't check right now. In either case, these are fairly short repetitive regions that can be given a genomic context based on their flanking sequence. Most heterochromatin can't be treated that way, because the repetition occurs on too large a scale compared to sequencing reads (or the spacer between reads, in the case of paired-end sequencing).

Here's the part I was referring to:

That's telling you what fraction of cells have visibly larger or smaller segments, not the degree to which they change size.

 

[Aggie]

That's telling you what fraction of cells have visibly larger or smaller segments, not the degree to which they change size.

Do you think that's still significant in the context of Tomkins' point? If they vary by 8% between cells within a single individual, that would seem to indicate that an 8% variation between humans and chimpanzees isn't very meaningful.

 

[sfs]

Do you think that's still significant in the context of Tomkins' point? If they vary by 8% between cells within a single individual, that would seem to indicate that an 8% variation between humans and chimpanzees isn't very meaningful.

I suspect it's relevant, but it's really hard for me to evaluate. Partly because I don't know enough about the molecular biology involved. Could the visible differences represent not different amounts of chromatin, but different genetically determined degrees of condensation? I have no idea.

There basically no facts here that are well enough established for me to draw any conclusions from. I don't know whether humans and chimpanzees actually have different amounts of heterochromatin, or how much it varies within either species, or whether variation in overall heterochromatin amount has any biological significance at all. There is literally no element of his argument is understood well enough to take seriously, and perhaps that's the right response to this one.

There's a saying I've run into science: "Let's not waste clean thoughts on dirty data." There's simply no scientific reason to be trying to explain a phenomenon whose existence is doubtful. What Tomkins is doing isn't remotely scientific, of course. He's rummaging around in crappy data, cherry-picking anything that suits his thesis and make it look scientific.

 

[SLP]

Do you think that's still significant in the context of Tomkins' point? If they vary by 8% between cells within a single individual, that would seem to indicate that an 8% variation between humans and chimpanzees isn't very meaningful.

I would agree to that - it also raises the issue of how this affects the genetic comparisons among/between organisms creationists believe to be in-Kind variants. Oddly enough, naysayers like Tomkins seem to avoid doing sequence comparisons on such creatures at all costs. I suspect this is because they are truly afraid of the possible results, and how the requisite double-standards they would be forced to employ would make them look.

 

[SLP]

Most of the other creationist analyses that were limited to "ungapped" results produced a similarity in the 80-90% range. What I'm looking for is an explanation for why this particular one was so much lower. The 70% paper is important to address because at this point it's probably the best-known creationist paper comparing chimpanzee and human DNA. Google scholar shows it's been cited 32 times, which is above-average for a paper that's only two years old.

You did see who was citing, yes?

 

[SLP]

Same here, but I don't think appearances are the key factor:

Great insights... You've never seen sequence data, have you?

 

[Aggie]

You did see who was citing, yes?

Yes, but I'm not sure it matters. With respect to the viewpoint of the general public (at least in the United States), I suspect that creationist books and articles tend to have more influence than papers in scientific journals, especially those that are behind paywalls.

 

[SkyWriting]

Great insights... You've never seen sequence data, have you?

From where?