search

omerwe / polyfun

PolyFun (POLYgenic FUNctionally-informed fine-mapping)
MIT License
85 stars 21 forks source link

All PIPs were zero #160

Closed poppy-yingying closed 1 year ago

poppy-yingying commented 1 year ago

Hi Omer, I'm doing the fine mapping analyses with Polyfun and Susie, using genotypes from the Plink file of 1KGP within a 500kb window. The finemap result shows that both the PIP, BETA_SD and the CREDIBLE_SET columns contain only 0, BETA_MEAN column contains 0 and -0.

python finemapper.py --geno 1KGP --sumstat snps_with_var.gz --n 598000 --chr 4 --start 26530403 --end 27030403 --method susie --max-num-causal 5 --cache-dir LD_cache --out snp --allow-missing

Is that because the PIPs are too small, or are there some other problems? Any help is much appreciated!

omerwe commented 1 year ago

@poppy-yingying Unfortunately I don't know what's the problem. Probably one of the input files has constant data. I can look into it if you can prepare a small reproducible example.

Also, I should caution that we show in the table that using 1000 genomes LD leads to wrong fine-mappig results (Table 3 from the paper), so I would avoid using it.