Hi,
Thank you for developing X-Wing.
I had a quick question regarding its flexibility: Can the pipeline be run using more than two populations? Specifically, I’d like to evaluate summary statistics from, say, EUR, EAS, and AFR, and have LEOPARD generate an optimal linear combination of PRSs for a target population like EAS.
If this is supported, could you please advise on how to set it up — particularly how to modify flags or input structures when running LOGODetect, PANTHER and LEOPARD for more than two populations (as shown in the example)?
Additionally, suppose one has already run LOGODetect outside of the X-Wing pipeline, how should the inputs be adapted for PANTHER, considering that the standard LOGODetect implementation does not provide the annotation matrix required by PANTHER?
Thanks in advance for your help!
Hi,
Thank you for developing X-Wing.
I had a quick question regarding its flexibility: Can the pipeline be run using more than two populations? Specifically, I’d like to evaluate summary statistics from, say, EUR, EAS, and AFR, and have LEOPARD generate an optimal linear combination of PRSs for a target population like EAS.
If this is supported, could you please advise on how to set it up — particularly how to modify flags or input structures when running LOGODetect, PANTHER and LEOPARD for more than two populations (as shown in the example)?
Additionally, suppose one has already run LOGODetect outside of the X-Wing pipeline, how should the inputs be adapted for PANTHER, considering that the standard LOGODetect implementation does not provide the annotation matrix required by PANTHER?
Thanks in advance for your help!