How to efficiently normalized imputation probabilities of the genotype matrix to between zero and one?

privefl / bigstatsr

R package for statistical tools with big matrices stored on disk.

https://privefl.github.io/bigstatsr/

179 stars 30 forks source link

How to efficiently normalized imputation probabilities of the genotype matrix to between zero and one? #148

Closed garyzhubc closed 2 years ago

garyzhubc commented 2 years ago

How to efficiently normalized imputation probabilities of the genotype matrix to between zero and one by dividing every number over two? big_scale doesn't seem to be able to perform this operation.

privefl commented 2 years ago

What is the point of dividing the whole matrix by 2? What do you want to use this for?