Hl.agg.linreg() output question

Hail Query & hailctl
1

Hi there,

I know that this might be a stupid question, but when I used hl.agg.linreg() to run GWAS on alternative allele and a set of covariates, I’m not sure about which element in the output array corresponds to the alternative allele. Specifically, I used the following code to run GWAS:

mt = mt.annotate_rows(gwas=hl.agg.linreg(mt.quant_pheno,
[1, mt.covar.isMale, mt.covar.YEAR, mt.covar.isAxiom,
mt.covar.PC1, mt.covar.PC2, mt.covar.PC3, mt.covar.PC4,
mt.covar.PC5, mt.covar.PC6, mt.covar.PC7, mt.covar.PC8,
mt.covar.PC9, mt.covar.PC10, mt.GT.n_alt_alleles()]))

For each SNP in the summary statistics, the effect size, se, and p-value estimates are all stored in an array with 15 elements. In this case, I wonder which element in the array corresponds to the estimated coefficient for alternative allele? Thank you very much!

2

The betas (and every other array) are in the same order as the independent variables – in this case, the beta on GT.n_alt_alleles() is in the last position, which you can access with betas[-1].

You also might prefer to run hl.linear_regression_rows – it’s faster and makes the simple GWAS case a little easier.

4

Sounds good, thank you very much!