search

zhqingit / giremi

GIREMI is a method that can identify RNA editing sites using one RNA-seq data set without requiring genome sequence data.
42 stars 15 forks source link

Homozygous SNV's #36

Closed steveners closed 7 years ago

steveners commented 7 years ago

Hello! I have some question: in previous issues you reccomend to remove homozygous SNV's from SNV list. Now I have a question: should I remove homozygous SNV's or I should remove homozygous SNP's known from dbSNP? I found in article that this requirement is necessary only for GLM stage: you train this model assuming that all SNP's from list are heterozygous. So why should I remove all homozygous SNV's? Thank you!

zhqingit commented 7 years ago

Hi Stevener,

I suggest to remove all homozygous SNVs. Specifically, the SNVs with extreme frequencies. The MI distribution is also affected by the homozygous SNVs. Thanks.

Best, Qing

2017-06-20 14:00 GMT-07:00 steveners notifications@github.com:

Hello! I have some question: in previous issues you reccomend to remove homozygous SNV's from SNV list. Now I have a question: should I remove homozygous SNV's or I should remove homozygous SNP's known from dbSNP? I found in article that this requirement is necessary only for GLM stage: you train this model assuming that all SNP's from list are heterozygous. So why should I remove all homozygous SNV's? Thank you!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/zhqingit/giremi/issues/36, or mute the thread https://github.com/notifications/unsubscribe-auth/ALL2zX-pqhXYM4pfLTH-8TWmizmTbjhgks5sGDLegaJpZM4OAJov .

steveners commented 7 years ago

Hi Qing,

Thank you! It is just strange that there is no information about it in paper.