HyPhy message board - Print Page

HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl Methodology Questions >> How to >> basic population genetic statistics http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1351200601 Message started by rbjmax on Oct 25^th, 2012 at 2:30pm

Title: basic population genetic statistics
Post by rbjmax on Oct 25^th, 2012 at 2:30pm

hi all-
I would like to calculate 'pi' for my virus populations- mean number of pairwise nucleotide differences weighted by numbers of individuals per haplotype. is this available in HyPhy?

thanks!
Rachel

Title: Re: basic population genetic statistics
Post by Sergei on Oct 26^th, 2012 at 11:18am

Hi Rachel,

How do you want to supply the # of individuals per haplotype? You could encode that in the sequence name, e.g.

>seq1_100
act...
>seq2_500
agt...

or by providing a separate file, such as

seq_name, haplotype_count
seq1,100
seq2,500

Sergei

Title: Re: basic population genetic statistics
Post by rbjmax on Nov 7^th, 2012 at 11:18am

Sergei-
I have both file types prepared. the first option is the most cohesive with my overall data management strategy.

Rachel

Title: Re: basic population genetic statistics
Post by Sergei on Nov 13^th, 2012 at 4:34pm

Hi Rachel,

Use the attached. It expects an alignment where each sequence id is like "tag_count", e.g. seq1_500

Sergei

http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=PiPopulation.bf (1 KB | 5 )

Title: Re: basic population genetic statistics
Post by rbjmax on Jan 4^th, 2013 at 2:48pm

thank you, Sergei!

that .bf ran beautifully, thank you. does it utilize the Nei 1987 method like Arlequin used to? also, the values were about 100 times higher than I expected- are they percentage values?

Rachel

Title: Re: basic population genetic statistics
Post by Sergei on Jan 4^th, 2013 at 3:41pm

Hi Rachel,

The method is as simple as it gets: just compute the NUMBER of pairwise differences between sequences. It doesn't for example take sequence length into account, which is why you may be seeing numbers 100* larger that expected. The attached code implements the Nei and Li (1979) method (Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=PiPopulation_001.bf (1 KB | 1 )

Title: Re: basic population genetic statistics
Post by rbjmax on Jan 7^th, 2013 at 10:46am

perfect, thanks Sergei