GWAS finishing effort - 691 missing variants

[yk-tanigawa]

It turned out that the summary statistics generated in array-combined/gwas/current does not match the number of expected lines (1,080,969).

There were 407 files with 1080278 lines (meaning that 691 lines are missing in each file).

[yk-tanigawa]

We identified that of those 691 variants,

• 369 variants are on both arrays
• 322 variants are on one array

and performed association analysis.

For some phenotypes, we saw

Skipping ... since all samples are controls

[yk-tanigawa]

Summary of counts

• We applied association analysis for 407 (pop, GBE_ID)
  • 180 have some results
    • 176 are finished with full results (691 variants)
    • 4 have 322 variants (369 variants are missing)
      • # samples <= # predictor columns
  • 227 were skipped
    • 2: all cases
    • 196: all controls
    • 17: constant quantitative phe
    • 12: # samples <= # predictor columns

[yk-tanigawa]

180 files were merged.

https://github.com/rivas-lab/ukbb-tools/tree/master/04_gwas/extras/202006-GWAS-finish/gwas691

From this analysis, we know that

• 4 files have 1,080,599 lines
  • 4: # samples <= # predictor columns
• 227 files have 1,080,278 lines
  • 2: all cases
  • 196: all controls
  • 17: constant quantitative phe
  • 12: # samples <= # predictor columns