LD reference panel for other populations

[swvanderlaan]

Hi,

Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories?

Specifically, I am looking for an 'ALL' reference panel to use with my GWAS results.

I noticed this in the README of the eur_w_ld_chr panel:

LD Scores computed with ldsc and the following parameters

• 1000 Genomes
• sum r^2 taken over HapMap3 SNPs
• HLA removed in 6.* but not in 6_old.*

For use with the --w-ld-chr flag in the tutorial.

But if you have a script to share to make the panel?

Thanks

Sander

[paturley]

Hello Sander,

The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need.

Best, Patrick

On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < [email protected]> wrote:

Hi,

Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories?

I am looking for an 'ALL' reference panel to use with my GWAS results.

Thanks

Sander

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/118, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ .

[wanluanluan]

Hello Sander, The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need. Best, Patrick … On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < @.***> wrote: Hi, Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories? I am looking for an 'ALL' reference panel to use with my GWAS results. Thanks Sander — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#118>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ .

Hi, How can I read local LD scores data，chicken was the experimental subject. I got "After merging with reference panel LD, 0 SNPs remain." for my MTAG analysis，when using the default LD reference panel in the ldsc/mtag package. I think the default LD reference panel only only apply to human data. Thanks， Zhou

[paturley]

Hello Zhou,

I'm a bit confused about your comment. Are you using chicken summary statistics but a human LD reference panel?

Patrick

On Mon, Jan 18, 2021 at 9:49 PM wanluanluan [email protected] wrote:

Hello Sander, The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need. Best, Patrick … <#m_8220116198107694072_> On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < @.***> wrote: Hi, Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories? I am looking for an 'ALL' reference panel to use with my GWAS results. Thanks Sander — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#118 https://github.com/JonJala/mtag/issues/118>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ .

Hi, How can I read local LD scores data，chicken was the experimental subject. I got "After merging with reference panel LD, 0 SNPs remain." for my MTAG analysis，when using the default LD reference panel in the ldsc/mtag package. I think the default LD reference panel only only apply to human data. Thanks， Zhou

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/118#issuecomment-762566581, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PP6AJLHI7XZHOYSZDS2TXNDANCNFSM4TS7XWBQ .

[wanluanluan]

Hello Zhou, I'm a bit confused about your comment. Are you using chicken summary statistics but a human LD reference panel? Patrick On Mon, Jan 18, 2021 at 9:49 PM wanluanluan [email protected] wrote: … Hello Sander, The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need. Best, Patrick … <#m_8220116198107694072_> On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < @.***> wrote: Hi, Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories? I am looking for an 'ALL' reference panel to use with my GWAS results. Thanks Sander — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#118 <#118>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ . Hi, How can I read local LD scores data，chicken was the experimental subject. I got "After merging with reference panel LD, 0 SNPs remain." for my MTAG analysis，when using the default LD reference panel in the ldsc/mtag package. I think the default LD reference panel only only apply to human data. Thanks， Zhou — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#118 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PP6AJLHI7XZHOYSZDS2TXNDANCNFSM4TS7XWBQ .

Hi，

yes，so，0 SNPs remain.If I have chicken LD scores data,how can I using chicken LD reference？

Thanks, Zhou

[paturley]

Do you have a chicken reference panel? If so, you should be able to follow the directions at the LDSC wiki to generate new LD scores. https://github.com/bulik/ldsc/wiki/LD-Score-Estimation-Tutorial

Once you have those, you can just direct MTAG to the path of your new LD scores, and it should work fine.

On Sat, Jan 23, 2021 at 2:48 AM wanluanluan [email protected] wrote:

Hello Zhou, I'm a bit confused about your comment. Are you using chicken summary statistics but a human LD reference panel? Patrick On Mon, Jan 18, 2021 at 9:49 PM wanluanluan [email protected] wrote: … <#m_-8930198164004317449_> Hello Sander, The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need. Best, Patrick … <#m_8220116198107694072_> On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < @.***> wrote: Hi, Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories? I am looking for an 'ALL' reference panel to use with my GWAS results. Thanks Sander — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#118 https://github.com/JonJala/mtag/issues/118 <#118 https://github.com/JonJala/mtag/issues/118>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ . Hi, How can I read local LD scores data，chicken was the experimental subject. I got "After merging with reference panel LD, 0 SNPs remain." for my MTAG analysis，when using the default LD reference panel in the ldsc/mtag package. I think the default LD reference panel only only apply to human data. Thanks， Zhou — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#118 (comment) https://github.com/JonJala/mtag/issues/118#issuecomment-762566581>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PP6AJLHI7XZHOYSZDS2TXNDANCNFSM4TS7XWBQ .

Hi，

yes，so，0 SNPs remain.If I have chicken LD scores data,how can I using chicken LD reference？

Thanks, Zhou

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/118#issuecomment-765884411, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5P7RG72OKERLUSCEILS3J5LRANCNFSM4TS7XWBQ .

[wanluanluan]

Do you have a chicken reference panel? If so, you should be able to follow the directions at the LDSC wiki to generate new LD scores. https://github.com/bulik/ldsc/wiki/LD-Score-Estimation-Tutorial Once you have those, you can just direct MTAG to the path of your new LD scores, and it should work fine. On Sat, Jan 23, 2021 at 2:48 AM wanluanluan [email protected] wrote: … Hello Zhou, I'm a bit confused about your comment. Are you using chicken summary statistics but a human LD reference panel? Patrick On Mon, Jan 18, 2021 at 9:49 PM wanluanluan @.*** wrote: … <#m_-8930198164004317449_> Hello Sander, The LD scores used by MTAG are the same as those used by LD Score Regression. So you should be able to use any LD scores that are posted online. For example, the Price lab has a bunch here I think. ( https://alkesgroup.broadinstitute.org/LDSCORE/) You could also use the 1000G data to make your own LD scores if none of the scores there are what you need. Best, Patrick … <#m_8220116198107694072_> On Thu, Nov 12, 2020 at 3:49 AM Sander W. van der Laan < @.***> wrote: Hi, Any chance that you have made panels for other populations too? For instance, one for the 'EAS', one for the 'AFR', 'AMR', 'SAS', and perhaps a combined 'ALL' panel similar to the 1000G populations categories? I am looking for an 'ALL' reference panel to use with my GWAS results. Thanks Sander — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#118 <#118> <#118 <#118>>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PSCQJYFKRK6AYNXL3SPOORJANCNFSM4TS7XWBQ . Hi, How can I read local LD scores data，chicken was the experimental subject. I got "After merging with reference panel LD, 0 SNPs remain." for my MTAG analysis，when using the default LD reference panel in the ldsc/mtag package. I think the default LD reference panel only only apply to human data. Thanks， Zhou — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#118 (comment) <#118 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5PP6AJLHI7XZHOYSZDS2TXNDANCNFSM4TS7XWBQ . Hi， yes，so，0 SNPs remain.If I have chicken LD scores data,how can I using chicken LD reference？ Thanks, Zhou — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#118 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5P7RG72OKERLUSCEILS3J5LRANCNFSM4TS7XWBQ . Thank you so much for your reply.

[harleyi]

At the link: https://alkesgroup.broadinstitute.org/LDSCORE/ There is a text file: https://storage.googleapis.com/broad-alkesgroup-public/LDSCORE/README_new_data_location.txt That states:

"A minimal set of files necessary to run LDSC are available for free acecss at https://doi.org/10.5281/zenodo.7768714."

At the zenodo link, I believe the eas_ldscores.tar.bz2 are found in the file: [1000G_Phase3_EAS_weights_hm3_no_MHC.tgz] Please correct me if I am mistaken. Best, --Isaac Harley