Comments on Jonathan Bartlett's Office Hours

[Roy (30-year veteran)]

@John_Harshman

As an addendum to my original questions, I’ll add this: if we accept that some DNA is nonfunctional, i.e. junk, would we not expect that mutations in junk would be “random” by the standard meaning, i.e. with respect to fitness? And wouldn’t the distribution of mutations in junk make a for better comparison than starting with a uniform distribution?

I’m not an expert on DNA biochemistry, but AIUI even the ‘junk’ DNA is proteted from damage by DNA repair mechanisms - which (again AIUI) have no means of distinguishing sections of DNA. So even the distribution of mutations in ‘junk’ isn’t really a base level of mutation.

[Roy (30-year veteran)]

@johnnyb

@John_Harshman

Could you define “active information”? Why have you chosen that as the term?

I did not choose the term. The term was originally developed by William Dembski and Robert Marks in their analysis of evolutionary algorithms.

That was my first issue with the paper. It says the term “active information” is from the field of computer search optimisation. It doesn’t say that the term originates with and has only ever been used by a handful of ID advocates.

[Roy (30-year veteran)]

@johnnyb

Given that we have information that not all mutations are equally likely, we can and should replace the uniform assumption.

Then what would you be measuring? My method measures expected vs. actual. Yours is simply measuring actual vs. actual, which will always be zero. My goal is to create a test for the specific claim of whether or not a mutation is random with respect to function.

I am not at all sure whether “active information” would be such a test.

Using “WEASEL” programs as an example, changing the genetic algorithm so that it randomly mutates characters to one of AEHIKLMNRSTW_ instead of to any of A…Z_ would add approximately 1 bit of “active information”, but subsequent mutations would still be “random with respect to function”.[1] Similarly, limiting mutations (either completely or largely) to a subset of the genome or changing the frequency of types of mutation could also add “active information”, but the mutations could still be “random with respect to function".

[1] If you would argue that they are no longer “random with respect to function”, then mutations in the original algorithm aren’t “random with respect to function” either, since they can’t produce all of the characters that could be present in a Shakespearean quote (?:]'.![-";ï,) let alone any of the characters that Shakespeare might conceivably have used (but didn’t) or characters that he wouldn’t have used but could be included as possible mutations anyway.

[John_Harshman]

@Roy

So even the distribution of mutations in ‘junk’ isn’t really a base level of mutation.

I disagree. DNA repair is in fact a mechanism of mutation. Even as it returns some sites to their original conditions it results in change to others. At any rate, it’s upstream from observed mutations and must be considered part of the mutation process. The supposed pristine mutation rate you imagine does not exist.

[Roy (30-year veteran)]

I’m not imagining a “pristine mutation rate” - I’m arguing that there isn’t one, so there is no possible starting point for “active information”.

If DNA repair mechanisms vary, and I think they would, especially if ancient life is considered too (although this isn’t something I have expertise in, I’m presuming that DNA repair mechanisms have changed over time), is there really just one mutation distribution?

[John_Harshman]

@Roy

is there really just one mutation distribution?

Of course not. You don’t need ancient life; existing organisms have different mutation distributions. Your mitochondria have a different mutation distribution from your nuclei. I seem to recall that frequently transcribed sequences have a different distribution from rarely transcribed ones. There is not “a mutation rate”; there are “mutation rates”. If your point is that @johnnyb’s baseline is not biologically meaningful, then I think everyone but @johnnyb knows that.

[Roy (30-year veteran)]

@John_Harshman

@Roy

is there really just one mutation distribution?

Of course not.

Memo to self: skip the polite questions and return to arrogant know-it-all mode .

You don’t need ancient life; existing organisms have different mutation distributions. Your mitochondria have a different mutation distribution from your nuclei. I seem to recall that frequently transcribed sequences have a different distribution from rarely transcribed ones. There is not “a mutation rate”; there are “mutation rates”. If your point is that @johnnyb’s baseline is not biologically meaningful, then I think everyone but @johnnyb knows that.

Pretty much - my point is that @johnnyb hasn’t actually proposed a baseline, and there are many, many reasons why any baseline that’s proposed will not work.

[swamidass (S. Joshua Swamidass)]

@Roy

skip the polite questions and return to arrogant know-it-all mode .

Please don’t.

[Roy (30-year veteran)]

Having thought more about this over the weekend, my main substantial criticism is that there does not seem to be a way to determine what would qualify as a biological “pure random search”, which would be necessary to calculate IΩ, without which “active information” cannot be quantified.

It seems possible to calculate “active information” for some (but not all) GA mutation processes (but not necessarily across multiple generations with selection). But that’s because in GAs, the replication and mutation code and associated parameters (e.g. mutation types and frequencies) are typically separate from the pseudo-organisms, and do not change over a run of the GA. The size and structure of the “genome” is usually fixed[1]. The GA’s scope is known. IΩ can be calculated before the GA is run.

For biological organisms, none of this is true. The nature and structure of genomes evolves, and varies across organisms (DNA vs RNA, strands vs loops). Many organisms have more than one genome. The replication ‘machinery’, including replication error and deterioration control mechanisms, also evolves. We don’t know what other biopolymers might be suitable material for genomes, or how susceptible they would be to mutation; we don’t know what other replication systems might be possible; we only know what genomes look like now, after billions of years of evolution - we don’t even know much about how genetics and mutations worked in the past, let alone how it might have worked had events taken a different turn. We do not know the original form(s) of genomes, nor their potential variation. So we can’t determine IΩ for biological organisms, nor can we ever quantify biological “active information”.

[1] There are exceptions, such as tierra or Avida in which replication is performed by the pseudo-organisms; or my 'arms GA (and tierra again) where the genome may increase or decrease in length. There isn’t any reason why mutation rates can’t be included in a pseudo-organism’s genome, I’m just not aware of it being done - possibly because it’s an unnecessary complication. Maybe I’ll try it.

[swamidass (S. Joshua Swamidass)]

Not sure why @Dan_Eastwood closed this and moved it to the front porch (it shouldn’t be there). @roy it is better to post your thoughts here. I don’t want @johnnyb to get pilled on, and there are already 3 scientists discussing his ideas on the main thread.

[Roy (30-year veteran)]

My thoughts aren’t related to the thread title.

P.S. What criteria would convince you to change “3 scientists” to “3 other scientists”?

[Dan_Eastwood]

@swamidass

Not sure why @Dan_Eastwood closed this and moved it to the front porch (it shouldn’t be there).

At the time created it was more squabbling than useful comments. I must have set a timer too short.

[Mercer (John Mercer)]

@swamidass

Is your point that most mutations fall outside the intended target?

That seems like a complex question fallacy to me.

I don’t see any intention anywhere. Doesn’t intention require someone to do the intending?

[swamidass (S. Joshua Swamidass)]

I did not mean it that way. Most mutations fall outside the “useful” region?

[Mercer (John Mercer)]

Can you reliably identify the useful regions from the red bar graph?

@Mercer

[swamidass (S. Joshua Swamidass)]

That graph is not labeled for such a task. You may need to just directly say your scientific point.