saigegit / SAIGE

Development for SAIGE and SAIGE-GENE(+)
GNU General Public License v3.0
64 stars 27 forks source link

Discrepancy between reported P value and P value calculated from Z statistic #145

Open AhmedEdris opened 3 months ago

AhmedEdris commented 3 months ago

Hi, I am conducting a GWAS Meta-analysis, and while conducting QC, some of the cohorts I am including have shown a discrepancy between the reported SAIGE P value, and the P value calculated from Z statistic. (Please see attached plot for pattern, showing -log10P for observed P value vs -log10 of P value calculated from Z statistic) We were wondering what could be the cause of this discrepancy.

Thank you! image