World Families Forums - hidden mutations

Welcome, Guest. Please login or register.
August 01, 2014, 06:39:47 PM
Home Help Search Login Register

+  World Families Forums
|-+  General Forums - Note: You must Be Logged In to post. Anyone can browse.
| |-+  R1b General (Moderator: rms2)
| | |-+  hidden mutations
« previous next »
Pages: [1] Go Down Print
Author Topic: hidden mutations  (Read 326 times)
ironroad41
Old Hand
****
Offline Offline

Posts: 219


« on: May 24, 2012, 07:46:59 AM »

This is a copy of a post on rootsweb by stan:








Welcome to RootsWeb.com Sign in  













DISCOVER MORE >














HomeSearchesFamily TreesMailing ListsMessage BoardsWeb SitesPasswordsHelp







GENEALOGY-DNA-L Archives
Archiver > GENEALOGY-DNA > 2012-05 > 1337852326
--------------------------------------------------------------------------------


 
From: didier.vernade@Safe-mail.net
Subject: Re: [DNA] Major new paper on Haplogroup G
Date: Thu, 24 May 2012 05:38:46 -0400

Hi Stan,
 
I saw the post by John Chandler and it may look as a warning for you since the "expectation" is much less than 300 000 years...
 Anyway, I understand the principle of the corrections but not the modalities. Could you take 2 different examples. One would be for a "simple" and very well know case such as R1b-M222. I posted a long while ago, when this group was knowwn as R1b1c7, on this list about R1b1c7 as a calibration tool (for TMRCA). This group is a recent one with very few corrections, if any, needed. This is for me to see what's going on when no correction is needed.
Then, let's look at the haplotypes G-L497, from http://www.familytreedna.com/public/G-YDNA ; only those said L497+ , all subclades. I'll compare with other methods.
 
How do you count mutations ? DYS490=12 ---> DYS490=10 ; is this one mutation or 2 ?
 What do you do with unsual results like "null" results and extra copies (DYS464) ?
 What about RecLOH events ? (ex : DYS385=11,14 ---> DYS385=11,11 in one generation).
 
Didier
 
> Hi Didier,
 > Willingly I will answer your questions, but unfortunately, in the Chinese.
 > I didn't want to learn English, because I believed in the reason of masters
> of world, that would choose neutral and international language Esperanto.
 > I was even an active activist of the international movement of Esperantists
 > and a co-organizer of international E-conferences.
 > But this idea lost. Therefore now I am writing in the “Chinese” English...
 >
> What is the multiplier of the exponential sequence?
 >
> So along with the height of places already mutate in the haplotype,
> proportionally a chance is growing for repeat mutations, practically and
> statistically invisible, unobserved; e.g. in haplotype 67 mark. from 1/66
> through 33/33 to 66/1.
 >
> So the increase in invisible mutations isn't linear (arithmetical), but
> exponential (geometrical). Proportionally a time of visible mutations is
> being extended.To this mutation rate slowed down visible in next markers,
> delays in lazily mutating markers are overlapping each other.
 > So, if we will resign from the laziest DYS472 marker (one mutation on about
> 1.700000 years in pairs fater-son, according to the table J. Chandler's), to
> one slowest DYS578 marker (1 mutation on 307125 years) in this haplotype
> 2755 mutations are falling.
 >
> If for one mutation in pairs father - son (in 67 mark. haplotype) 111 years
> are falling, and on 66 mutations - 7326 years, so it is necessary to
> multiply next numbers of this arithmetical row with numbers of such an
> exponential row in order to get the final result for 66 visible mutations -
> the time 307125 years.
 > With method of the simulation I am finding such a row, of which a multiplier
> 1,058236053^1...66.
 >
> Results are effective in practice in calculations of the time of recent
> mutations ("genealogical"), as well as former ("evolutionary", e.g. Adam’s),
> that is with numerous repeat, invisible mutations.
 >
> To needs of estimating the mutation in haplotypes Y-STR I am not finding the
> model in textbooks of genetics.
 > But for the SNP mutations they drew the analogous model up Jukes-Kantor
 > (see: Daniel Hartl, Andrew Clark, Principlesof Population Genetics, ed. 4,
> 2007; chap.7.2; fog. 7.4).
 > Although in the history of the man we aren't expecting repeat SNP mutations
> in nucleotides, however for gradual long-lasting processes up to the
> attention a possibility of repeat mutations is stemming.
 > Exponential graph Jukes-Kantor is pointing, that observed mutation rate
> systematically and exponentially is decreasing, or differently - in spite of
> the permanent passage of time the number of observed mutations is
> decreasing, and above 75 per cent of the mutate of loci in the "haplotype"
> the increase in visible mutations is even zero.
 >
> Stan.
 >
>
>
> ----- Original Message -----
>
> From: <didier.vernade@safe-mail.net>
 > To: <genealogy-dna@rootsweb.com>
 > Sent: Wednesday, May 23, 2012 9:07 AM
 > Subject: Re: [DNA] Major new paper on Haplogroup G
 >
>
> > Stan,
 > >
 > > Where did you describe you method ? I need to understand "multiplier of
> > the exponential sequence"
 > > What's the principle : “everyone with everyone” ?
 > > "all second-rate GD out which don't reach MRCA." : this is chinese to
> > me....
 > >
 > > Didier
 > >
 > > Anatole Klyosov wrote :
 > >> > My response:
 > >> > It is awfully incorrect.
 > >> > First, you should compare "average GD", not haplotypes "standing out
> >> > with
 > >> > highest GD".
 > >> > Second, the table is just a disaster. Its basis (explained there) is
 > >> > totally wrong.
 > >> > For example, 33 mutations correspond to 9400 years (that is, they
 > >> > accumulated on average during 9400 years). In your table it is 23,700
 > >> > years (!).
 > >> > 26 mutations correspond to 6400 years, not 12,600 years, as you have
 > >> > obtained. However, since you operate with "standing out with highest
> >> > GD",
 > >> > it should apparently be lower than 6400 years.
 > >> > It is sad that with such an error, of about 200-1100% (for higher "GD"
> >> > in
 > >> > your table), you operate with a coefficient of 1.058236043 (!). What a
 > >> > precision!
 > >> > Anatole Klyosov
 > >> ..............
 > > Stan answer :
 > >> I am continuing my response.
 > >> 1) The 1.058236043 number isn't a coefficient of the mutation!
 > >> It is only multiplier of the exponential sequence.
 > >> It is established based on the simulation, for joining a point of
> >> departure
 > >> and final.
 > >> Starting and final point – are result from the table of mutation rates
> >> (e.g.
 > >> J. Chandler’s)
 > >>
 > >> 2) When we are calculating genetic distances among two separate groups,
 > >> averaging of all results is necessary.
 > >> But when I am calculating GD inside one assigned group (intraclade) on
> >> the
 > >> principle “everyone with everyone”,
 > >> are needed to rule all second-rate GD out which don't reach MRCA.
 > >> I am only taking the ones GD which clearly reach MRCA.
 > >> It is group having mainly a GD maximum value.
 > >> it is possible then also to divide them in two branches.
 > >> Their GD I am averaging.
 > >>
 > >> 3) Achieved with this way result of about 12600 years Bp for MRCA of the
 > >> P303 group, was confirmed through putting together the XK97J sample,
> >> which
 > >> is the separate oldest branch under the P303 mutation, with remaining 125
 > >> haplotypes with it mutation.
 > >> Their GD to MRCA amounts to (39-65)/2, on average 51/2 = 12000 years Bp.
 > >>
 > >> 4) “awfully incorrect...”, “is just a disaster...” “it is sad...”
 > >> Are these emotions language of the learning?
 > >> All the best! Stan
 > >
 

--------------------------------------------------------------------------------

This thread:
DNA] Major new paper on Haplogroup G by Thomas Krahn <thomas@familytreedna.com>■Re: [DNA] Major new paper on Haplogroup G by "Brian P. Swann" <bps@norvic8.force9.co.uk>■Re: [DNA] Major new paper on Haplogroup G by mtDNA H Project <hmtdna@gmail.com>■Re: [DNA] Major new paper on Haplogroup G by Andrew Mceachern <andrewmceachern@ymail.com>



◦Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net
◦Re: [DNA] Major new paper on Haplogroup G by aklyosov@comcast.net
◦Re: [DNA] Major new paper on Haplogroup G by Robert Hughes <baddog.nobisquit@gmail.com>
◦Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net
◦Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net■Re: [DNA] Major new paper on Haplogroup G by Andrew Mceachern <andrewmceachern@ymail.com>

DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.netDNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net■Re: [DNA] Major new paper on Haplogroup G by "Bernard SECHER" <secher.bernard@orange.fr>

◦Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net■Re: [DNA] Major new paper on Haplogroup G by <pietrzakstan@poczta.onet.pl>

◦Re: [DNA] Major new paper on Haplogroup G by aklyosov@comcast.net■Re: [DNA] Major new paper on Haplogroup G by <pietrzakstan@poczta.onet.pl>
■Re: [DNA] Major new paper on Haplogroup G by <pietrzakstan@poczta.onet.pl>■Re: [DNA] Major new paper on Haplogroup G by "Anatole Klyosov" <aklyosov@comcast.net>


◦Re: [DNA] Major new paper on Haplogroup G by aklyosov@comcast.net
◦Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net■Re: [DNA] Major new paper on Haplogroup G by <pietrzakstan@poczta.onet.pl>■Re: [DNA] Major new paper on Haplogroup G by john.chandler@alum.mit.edu (John Chandler)





Re: [DNA] Major new paper on Haplogroup G by didier.vernade@Safe-mail.net

 



--------------------------------------------------------------------------------


RootsWeb is funded and supported by Ancestry.com and our loyal RootsWeb community. Learn more.

About Us | Contact Us | Acceptable Use Policy | PRIVACY STATEMENT | Copyright
© 2012 Ancestry.com

I will admit that this is a tough post to interpret.  That said, his comments about hidden mutations and their frequency is very important, Iff he is right.  I have previously commented about the mutation about the modal (and Machiavelli too).  The idea being, that without multisteps, the probabililty of a dys loci mutating beyond +/- 1 is very small for essentially two thirds of the first 67 dys loci.  Chandler is getting involved here and it will be interesting to read his replies(rebuttals)?
 
« Last Edit: May 24, 2012, 07:53:34 AM by ironroad41 » Logged
Maliclavelli
Guru
*****
Offline Offline

Posts: 2147


« Reply #1 on: May 24, 2012, 09:26:11 AM »

I have written tons of letters about the "hidden mutations", to Nordtvedt, Klyosov and all the others. What did mean my principles (mutations around the modal, convergence to the modal as time passes, sometime a mutation goes for the tangent and we have an outlier). Upon this I based my theories and I haven't ever believed to the times of Vizachero and all the others and I am waiting trustful the aDNA.
Logged

Maliclavelli


YDNA: R-S12460


MtDNA: K1a1b1e

JeanL
Old Hand
****
Offline Offline

Posts: 425


« Reply #2 on: May 24, 2012, 09:28:14 AM »

The lovely and humble Anatole Klyosov never ceases to amaze me.

Cough*Sarcasm*Cough
Logged
ironroad41
Old Hand
****
Offline Offline

Posts: 219


« Reply #3 on: May 24, 2012, 09:43:55 AM »

There may be a simpler way to get at this question using CDYa,b.  We have a pretty fair handle on the rate of mutation of this dys loci from father son-pairs.  What we should (?) be able to do, for specific sets of data with known TMRCA's, e.g. Ian Cam of Clan Gregor is count the number of "visible mutations" and compare that to the number of expected mutations based on the rate data.  It should show up short?

I am sure there are other sets of data available to do this.

BTW, I believe this was the gist of Chandlers criticism that the real TMRCA for the set of data being investigated was assumed, possibly incorrectly.
Logged
Pages: [1] Go Up Print 
« previous next »
Jump to:  


SEO light theme by © Mustang forums. Powered by SMF 1.1.13 | SMF © 2006-2011, Simple Machines LLC

Page created in 0.133 seconds with 18 queries.