Should PARs be hard-masked in human genome while using htseq-count for downstream analysis?

Question

[Gerde]

Hi all, I'm doing RNAseq analysis, and I have aligned reads to hg38 use HISAT2, as we know that htseq-count would discard multi-mapped reads, GENCODE gene moduls contain annotation in PARs, e.g. ENST00000399012.6 & ENST00000399012.6_PAR_Y, so should I hard mask PARs in chrY before alignment?

[WouterDeCoster]

As far as I know that is often the case, already. Sounds like a good idea if it isn't.

[Gerde]

Thanks @WouterDeCoster , I just wonder whether anybody has done that way before, and I found that reference used in Ilumina basespace APP is hard-masked in chrY PARs, maybe I should mask it before alignment, too. Thanks!