Provide un normalized XPEHH

[hardingnj]

From Sabeti 2007:

We normalize the XP-EHH logratio such that the set of all such logratios has zero mean and unit variance. We denote these normalized XP-EHH logratios by " XP-EHH scores"

Would it be possible to output both? If you split your analysis by contig/chromosome, it's helpful to perform the normalization over all contigs, not by contig.

The new iHS format #25 makes this easy for iHS, so thanks.

[camaclean]

Right now, they're actually un-normalized. I should also output the normalized scores, but it looks like I forgot to get around to it.

[hardingnj]

Thanks. Maybe what precisely is being calculated could be made slightly clearer in the docs? Thanks for the tool + maintenance btw, great work.

[camaclean]

Yeah, I'll get that updated. Thanks.

[hardingnj]

Thinking about. re: iHS

for my use case it would be handy to add an allele freq column for the
data, so users could bin and calculate standardized iHS themselves
having combined several tables. Otherwise I have to go back and look for
the AF of those SNPs I have iHS values for. Just a suggestion though!

On 24/03/16 14:16, camaclean wrote:

Yeah, I'll get that updated. Thanks.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#27 (comment)

Nicholas J Harding
Principal Bioinformatician
Global Health and Malaria Programme
Kwiatkowski Group
The Wellcome Trust Centre for Human Genetics
Roosevelt Drive
Oxford
OX3 7BN
United Kingdom
+44 1865 287712

[camaclean]

Sorry! I had done this but for some reason I lost track of things and forgot to push the XPEHH standardization. I'll add in frequency columns, too.