GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-03 > 1174591316


From: Anders Pålsen <>
Subject: Re: [DNA] Using CODIS with STRUCTURE 2.1 to infer individualancestry
Date: Thu, 22 Mar 2007 20:21:56 +0100 (CET)
In-Reply-To: <f3f05ce80703221105s39fd4026w92a831bec3462a55@mail.gmail.com>


Yes, thats my experience too provided there is given enough calculation time. The clusters get weaker but are still there.

Anders

Dienekes Pontikos <> skrev:
I think the graphs look clean because of the setting to "Use
population information". If the unlabeled data are used, then the four
clusters still emerge but individuals appear to be more "mixed".

On 3/22/07, Doug McDonald wrote:
>
> One must always remember when comparing different uses of the Structure,
> or similar, programs. that the number of markers, and their suitability
> in the AIM department, is not the only determinant of how "clean"
> the resulting color bar graphs look. There is also how "clean",
> that is, how racially un-mixed, the dataset is.
>
> Perhaps the extremely unmixed appearance of the graphs from the CODIS
> database dataset is simply a result of the dataset itself having
> an extreme low amount of admixture. Perhaps the graphs in the
> referenced paper are from a set which has far more mixed people.
>
> I have played quite a bit with Structure and have seen this
> effect.
>
> Doug McDonald
>
> -------------------------------
> To unsubscribe from the list, please send an email to with the word 'unsubscribe' without the quotes in the subject and the body of the message
>


--

Dienekes' Anthropology Blog
http://dienekes.blogspot.com

-------------------------------
To unsubscribe from the list, please send an email to with the word 'unsubscribe' without the quotes in the subject and the body of the message



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

Alt i én. Få Yahoo! Mail med adressekartotek, kalender og notisblokk.

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

Alt i én. Få Yahoo! Mail med adressekartotek, kalender og notisblokk.

This thread: