I would like to get advice on how to deal with the following scenario.
Scenario
-
I have GWAS summary statistics from a large meta-analysis of mixed-ancestry cohorts (multi-ancestry meta-GWAS).
-
I do not have access to cohort-specific GWAS summary statistics.
-
I also do not have a suitable LD reference panel that matches the effective LD structure of the meta-analysis (e.g. no per-cohort LD, no per-ancestry LD, and no large external biobank with a similar ancestry mix).
Question
- What are realistic options for fine-mapping in this situation, if any?
- Is it acceptable to use an imperfect external LD panel (e.g. a partially mismatched LD or a weighted mix of ancestry-specific LD panels) as an approximation, or should I skip fine-mapping and focus on alternative prioritization strategies instead?
Thank you!
Best regards,
Jeongah