search

MRCIEU / opengwas-requests

A place to request or contribute data to the IEU GWAS database
3 stars 1 forks source link

ieu-b-110 (LDL) has strange-looking plots and ldsc errored out #18

Open yfarjoun opened 3 years ago

yfarjoun commented 3 years ago

in the report for https://gwas.mrcieu.ac.uk/datasets/ieu-b-110/ the QQ plot is empty, as is the beta vs. std plot. the LDSC log indicated that 0 snps were found:

Beginning analysis at Mon Mar 29 16:40:49 2021
Reading summary statistics from /data/cromwell-executions/qc/6910f09b-8ff3-478e-adcc-b2440bf4a56e/call-ldsc/inputs/268670770/ieu-b-110.vcf.gz ...
Read summary statistics for 0 SNPs.

and it completes with an error:

Traceback (most recent call last):
  File "./ldsc/ldsc.py", line 647, in <module>
    sumstats.estimate_h2(args, log)
  File "/ldsc/ldscore/sumstats.py", line 330, in estimate_h2
    args, log, args.h2)
  File "/ldsc/ldscore/sumstats.py", line 252, in _read_ld_sumstats
    sumstats = _merge_and_log(ref_ld, sumstats, 'reference panel LD', log)
  File "/ldsc/ldscore/sumstats.py", line 238, in _merge_and_log
    raise ValueError(msg.format(N=len(sumstats), F=noun))
ValueError: After merging with reference panel LD, 0 SNPs remain.

This is quite an important GWAS (LDL), is the problem in the source, or in the processing?

Thanks!