search

szpiech / selscan

Haplotype based scans for selection
GNU General Public License v3.0
114 stars 33 forks source link

.OUT file blank #88

Open aramesh97 opened 2 years ago

aramesh97 commented 2 years ago

Hello,

I have successfully got the selscan to work with the --pmap flag and by providing a physical map of my chromosome. I have also used impute on my vcf to phase the file but my .out file is empty. In the log I get two types of errors:

  1. WARNING: Reached chromosome edge before EHH decayed below 0.05. Skipping calculation at position 13288828 id: chr21:13288828:G:A
  2. WARNING: Reached a gap of 229066bp > 200000bp. Skipping calculation at position 13850378 id: chr21:13850378:G:C

Can you please help me with this? Any explanation on what I can do to avoid this is much appreciated

szpiech commented 2 years ago

Hi there,

The default behavior of selscan is to skip calculations that are close to large gaps between loci or are close to chromosome end points. If your data is very sparse, you could change the maximum gap size with the --max-gap flag. If you are analyzing very small regions (e.g., perhaps you have simulated a 100kb region), you might want to use the --trunc-ok flag, which allows the program to compute the integration even if the EHH curve hasn't reached the default 0.05. I would caution against using --trunc-ok in very sparse data, as it could create weird biases. It is really best used for analyzing short regions of simulated data.

There is a small program for interpolating genetic map disances on my github (https://github.com/szpiech/predictGMAP) that does linear interpolation to get a genetic map file for all your sites (this also has a --max-gap flag, which you could choose to set very high if you are losing too many sites while processing). Eventually, I should add this functionality directly into selscan, but you can try this program if you are having trouble.

Let me know if you have any troubles.

Zachary

On Wed, Sep 21, 2022 at 2:54 PM aramesh97 @.***> wrote:

Hello,

I have successfully got the selscan to work with the --pmap flag and by providing a physical map of my chromosome. I have also used impute on my vcf to phase the file but my .out file is empty. In the log I get two types of errors:

  1. WARNING: Reached chromosome edge before EHH decayed below 0.05. Skipping calculation at position 13288828 id: chr21:13288828:G:A
  2. WARNING: Reached a gap of 229066bp > 200000bp. Skipping calculation at position 13850378 id: chr21:13850378:G:C

Can you please help me with this? Any explanation on what I can do to avoid this is much appreciated

— Reply to this email directly, view it on GitHub https://github.com/szpiech/selscan/issues/88, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABAKRQX3MKSIAT72AQRLZQLV7NKVBANCNFSM6AAAAAAQSLHP3E . You are receiving this because you are subscribed to this thread.Message ID: @.***>