search

alachins / raisd

RAiSD: software to detect positive selection based on multiple signatures of a selective sweep and SNP vectors
33 stars 13 forks source link

Why are monomorphic SNPs removed? #9

Closed sergiopalmavera closed 4 years ago

sergiopalmavera commented 5 years ago

Hello! RAiSD removes monomorphic sites, however one of aspect of a selective sweep is that the region around the benefitial mutation can become monomorphic.

What is the purpose of removing such sites? and could this reduce the power of the analysis?

Thanks!

alachins commented 5 years ago

Hi!

Yes, removing monomorphic sites can reduce power, but also including them has the disadvantage that the tests are more sensitive to assumptions regarding mutation rates. You can read more about this here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1310644/

The statistic implemented in RAiSD is not designed to explicitly take monomorphic sites into consideration. The power of the analysis using RAiSD comes from the combination of three statistics, one for reduced variation, one for the shift of the site frequency spectrum, and one for LD. RAiSD accounts for the fact that the region around the benefitial mutation can become monomorphic by introducing the first statistic that receives high values in regions with reduced variation.

Best regards, Nikos

On Wed, Aug 7, 2019 at 11:34 AM sergiopalmavera notifications@github.com wrote:

Hello! RAiSD removes monomorphic sites, however one of aspect of a selective sweep is that the region around the benefitial mutation can become monomorphic.

What is the purpose of removing such sites? and could this reduce the power of the analysis?

Thanks!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/alachins/raisd/issues/9?email_source=notifications&email_token=AALKWCTGJ6KDOYMXQMG2BBLQDKCIJA5CNFSM4IJ5ZMO2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HD2NCZQ, or mute the thread https://github.com/notifications/unsubscribe-auth/AALKWCWEKJHJV3YNFWOO5B3QDKCIJANCNFSM4IJ5ZMOQ .

-- Nikolaos Alachiotis

sergiopalmavera commented 5 years ago

Hello Nikos! thank you very much for your response.

I was wondering if you would consider reasonable to integrate the results from RAiSD with results of a tests that takes haplotype homozyosity into account (i.e. iHS)?

Kind regards Sergio

EDIT: Would it be correct to state that RAiSD is oriented to identify ongoing selection and that if fixation of the selected allele is reached, another tests must be applied?

alachins commented 4 years ago

RAiSD detects complete sweeps, i.e., when fixation of the selected allele is reached.

We have compared RAiSD with iHS before, but note that its scope is different: iHS detects ongoing sweeps relying on extended haplotypes, and not complete sweeps.

https://jbiolres.biomedcentral.com/articles/10.1186/s40709-017-0064-0

Best regards, Nikos