andreyshabalin
commented
3 months ago If you have infinite t-statistic, you have perfect correlation between the variables. The p-value is then essentially zero which you see here.
Open winnerHeyc opened 5 months ago
andreyshabalin
commented
3 months ago If you have infinite t-statistic, you have perfect correlation between the variables. The p-value is then essentially zero which you see here.
Why do extremely small p-values appear in the Matrix results? Are these due to calculation issues, and can they be used directly? head cis_eqtlout SNP gene beta t-stat p-value FDR Chr1_1025645 MSTRG.341.1 13.07 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1025645 MSTRG.343.1 4.09 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1307252 MSTRG.341.1 4.35666666666667 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1307252 MSTRG.343.1 1.36333333333333 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1505533 MSTRG.341.1 13.07 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1516057 MSTRG.341.1 13.07 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1505533 MSTRG.343.1 4.09 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1516057 MSTRG.343.1 4.09 Inf 2.2250738585072e-308 2.91956304784701e-304 Chr1_1573233 MSTRG.341.1 13.07 Inf 2.2250738585072e-308 2.91956304784701e-304