HISAT2: Extracting splice sites and exons for index

Question

[psm]

Hello all, RNAseq newbie here, sorry if the question doesn't make sense. I want to build an index using the hg38 reference genome and the .gtf annotation file. I see some people build the index directly with just these two files, whereas other sources suggest to first extract splice sites and exons and then index the genome.

When would I use either approach? My goal is gene level (not transcript level) differential expression analysis using a HISAT2 --> FeatureCount --> edgeR approach. Any insight would be much appreciated.