sriramlab / SCORE

SCalable genetic CORrelation Estimator
8 stars 1 forks source link

Documentation #7

Open jerome-f opened 2 years ago

jerome-f commented 2 years ago

Hi,

This is a nice tool, but the documentation is quite poor (RHE-mc as well) I am familiar with the documentation in LDSC it is quite comprehensive and explains some of the theoretical background. I have questions regarding using imputed snps (in the 10s of millions)/WGS snps for the analysis how do you go about that, can I perform the analysis chromosome by chromosome ? if so how to reconcile that for whole genome. The paper has good amount of theory but the readme is quite minimalistic. I am new to the field and feel like shooting in the dark. I am sure this will help the community at large.

sriramsr commented 2 years ago

Hi,

Thank you for the feedback -- we will improve the documentation.

Analyzing chromosome by chromosome is possible mathematically but is not supported by either software as the estimates will need to be combined across the chromosomes. These are features that we are looking to build into the software.

Let us know if you have any further questions or comments.

Sriram

On Thu, Aug 11, 2022 at 7:56 PM Jerome @.***> wrote:

Hi,

This is a nice tool, but the documentation is quite poor (RHE-mc as well) I am familiar with the documentation in LDSC it is quite comprehensive and explains some of the theoretical background. I have questions regarding using imputed snps (in the 10s of millions)/WGS snps for the analysis how do you go about that, can I perform the analysis chromosome by chromosome ? if so how to reconcile that for whole genome. The paper has good amount of theory but the readme is quite minimalistic. I am new to the field and feel like shooting in the dark. I am sure this will help the community at large.

— Reply to this email directly, view it on GitHub https://github.com/sriramlab/SCORE/issues/7, or unsubscribe https://github.com/notifications/unsubscribe-auth/AENU7P6Q4O7LJL3IVQT2R2LVYW4OHANCNFSM56KG2MTA . You are receiving this because you are subscribed to this thread.Message ID: @.***>