search

choishingwan / PRS-Tutorial

A tutorial on how to run basic polygenic risk score analysis
MIT License
68 stars 104 forks source link

#mismatching-snps #28

Open DM087 opened 2 years ago

DM087 commented 2 years ago

Hi Sam,

I am following your plink PRS tutorial and I have a couple of questions re mismatching-snps section https://choishingwan.github.io/PRS-Tutorial/target/#mismatching-snps .

  1. You say in the 'Note' section that 'Most PRS software will perform strand-flipping automatically, thus this step is usually not required.' I am trying to use plink --score and was wondering about running this step. Do you know if plink takes care of strand-flipping when using --score ?
  2. Should I be changing anything in the cases where SNP1 in the base data is A/C, and C/A in the target dataset? Do I need to change the .bim file in this case (or the effect allele in base data)?

Thank you for your help.

choishingwan commented 2 years ago

PLINK does not do strand-flipping, though we have a full step-by-step guide on our PRS-Tutorial website (which you are linking to) where we show how to do that with PLINK + R

Sam

On Fri, Nov 19, 2021 at 10:58 AM DM087 @.***> wrote:

Hi Sam,

I am following your plink PRS tutorial and I have a couple of questions re mismatching-snps section https://choishingwan.github.io/PRS-Tutorial/target/#mismatching-snps http://url .

  1. You say in the 'Note' section that 'Most PRS software will perform strand-flipping automatically, thus this step is usually not required.' I am trying to use plink --score and was wondering about running this step. Do you know if plink takes care of strand-flipping?
  2. Should I be changing anything in the cases where SNP1 in the base data is A/C, and C/A in the target dataset? Do I need to change the .bim file in this case (or the effect allele in base data)?

Thank you for your help.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/choishingwan/PRS-Tutorial/issues/28, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJTRYX3TE7SGDOJI7C6BWTUMZX3HANCNFSM5IMO3KTA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

DM087 commented 2 years ago

Thank you for the prompt reply. Much appreciated! Regarding my 2nd question: what about A/C in base data and C/A in target data? Do I have to change the .bim file or change the effect allele and reverse the sign of beta so that both sets have A/C? How do you deal with it ?

Thank you.

choishingwan commented 2 years ago

It’s shown in the tutorial

In short, for PRSice and our trial, we flip the genotype encoding (short of updating the bim)

On Fri, 19 Nov 2021 at 11:47 AM, DM087 @.***> wrote:

Thank you for the prompt reply. Much appreciated! Regarding my 2nd question: what about A/C in base data and C/A in target data? Do I have to change the .bim file or change the effect allele and reverse the sign of beta so that both sets have A/C? How do you deal with it ?

Thank you.

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/choishingwan/PRS-Tutorial/issues/28#issuecomment-974234420, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJTRYUP6NRPUBXC36GSMSLUMZ5TXANCNFSM5IMO3KTA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

-- Dr Shing Wan Choi Instructor Genetics and Genomic Sciences Icahn School of Medicine, Mount Sinai, NYC

DM087 commented 2 years ago

Thanks Sam! Have a nice weekend!

DM087 commented 2 years ago

I am still not clear about it. My question is this. Would you reverse the sign of beta given the following

A1 in base GWAS summary statistics file is the same as A2 in target GWAS data?

Thanks

choishingwan commented 2 years ago

It's automated by PRSice

On Thu, Jul 14, 2022, 6:04 PM DM087 @.***> wrote:

I am still not clear about it. My question is this. Would you reverse the sign of beta given the following

A1 in base GWAS summary statistics file is the same as A2 in target GWAS data?

Thanks

— Reply to this email directly, view it on GitHub https://github.com/choishingwan/PRS-Tutorial/issues/28#issuecomment-1184941675, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJTRYRW7GMEGBPF5RKRVVDVUCFG7ANCNFSM5IMO3KTA . You are receiving this because you commented.Message ID: @.***>

DM087 commented 2 years ago

Thanks Sam. Sorry, I did not make it clear that I am talking about --score in plink. Does it mean that I will have to reverse the sign of beta in this case? Thanks

DM087 commented 2 years ago

@choishingwan any thoughts? thanks

choishingwan commented 2 years ago

Our tutorial has detailed how to do the flip. You can also flip the sign of the summary statistic I think.

On Mon, Jul 18, 2022, 11:29 AM rs7412 @.***> wrote:

@choishingwan https://github.com/choishingwan any thoughts? thanks

— Reply to this email directly, view it on GitHub https://github.com/choishingwan/PRS-Tutorial/issues/28#issuecomment-1187643975, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAJTRYUJGNFOPYBPBFHRQQ3VUVZ47ANCNFSM5IMO3KTA . You are receiving this because you were mentioned.Message ID: @.***>