Why does BUGWAS calculate another p-values instead of using the raw GEMMA output p-values?

[Owen-haha]

Hi developer,
I realized that BUGWAS converts GEMMA output to another p-value instead of using the original one. Can you explain why do we need to do this conversion? Why do it in different ways for biallelic and tritetra allelic SNPs?

The relevant codes are as below.
############################################

For biallelic SNPs

LH1 <- lmm$logl_H1
D <- sapply(LH1, get_deviance, lognull=lambda.lognull["lognull"], USE.NAMES=FALSE)
pvals <- pchisq(as.numeric(D), 1, low=F)

For tritetra allelic SNPs

LH1 <- as.numeric(sapply(lmm, getLH1, USE.NAMES=FALSE))
D <- sapply(LH1, get_deviance, lognull=lognull, USE.NAMES=FALSE)
pvals <- apply(data.frame("D" = D, "num.alleles" = num.alleles), 1, get_pvals)
############################################

Thanks