search

zhanxw / rvtests

Rare variant test software for next generation sequencing data
131 stars 41 forks source link

still negative AF, for binary trait #27

Open jielab opened 7 years ago

jielab commented 7 years ago

Hi, I just noticed that there are still negative AF value generated by RVTESTS, when my phenotype is a binary trait.

Can you please look into this?

Best regards, Jie

zhanxw commented 7 years ago

I have examined codes and have not found obvious problems. Are the negative AF close to zero? Are those variant sites singletons?

jielab commented 7 years ago

Dear Xiaowei:

Please see the attached screenshot below. These abnormal AF only occurred for my binary trait based analysis. There is no problem with the continuous trait based analysis. I see AF values higher than 10000 and lower than -100. So, there must be some thing very wrong.

Best regards, Jie

capture

zhanxw commented 7 years ago

@jiehuang001 is it possible you can post the log file here? I have tried many ways to debug this issue, but am still stuck.

jielab commented 7 years ago

Dear Xiaowei:

please see below, the first one shows that there are AF with values below -100. The second is the actual command and the RVTESTS log file. Hopefully you will be able to replicate the error and find a solution.

Best regards, Jie

1

2

zhanxw commented 6 years ago

The default frequency calculation method is changed in this version (https://github.com/zhanxw/rvtests/releases/tag/v2.0.6). The AF is just half of the mean genotypes/dosages, and it should be positive all the time now.

jielab commented 6 years ago

Thanks, Xiaowei!

Best regards,

Jie

From: zhanxw [mailto:notifications@github.com] Sent: 2017年10月10日 13:11 To: zhanxw/rvtests rvtests@noreply.github.com Cc: jiehuang001 jiehuang001@gmail.com; Mention mention@noreply.github.com Subject: Re: [zhanxw/rvtests] still negative AF, for binary trait (#27)

The default frequency calculation method is changed in this version (https://github.com/zhanxw/rvtests/releases/tag/v2.0.6). The AF is just half of the mean genotypes/dosages, and it should be positive all the time now.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/zhanxw/rvtests/issues/27#issuecomment-335543413 , or mute the thread https://github.com/notifications/unsubscribe-auth/AZsvfzu0BHAlLMwZaZOWzy2ehfmH6lPMks5sq6UagaJpZM4NUlir . https://github.com/notifications/beacon/AZsvf5lHISUmel93mm1v5IFgpVnCe8arks5sq6UagaJpZM4NUlir.gif