Could HISAT2-build build Hierarchical Graph FM index (HGFM) for a reference genome plus transcripts for Rat (for example, Rn6)

Question

[Peng Huang]

Hello everyone! I am doing rat RNA-seq analysis, previously using tophat2, as HISAT2 is a successor for tophat2, I plan to switch to it. And no pre-built indexs exist for Rat, hence I am going to build manually but I am wondering is it necessary and available to build Hierarchical Graph FM index (HGFM) for a reference genome plus transcripts for rat? For example, with reference genome plus transcriptome sequence of rat as input for HISAT2-build?

Answer 1

[Devon Ryan]

You would only include the reference genome, not the transcriptome sequence. You can also provide the splice junctions and there's a python script that comes with hisat2 to generate that file for you.

[Peng Huang]

Yes, I know what you mean, I just saw the description for the option ----known-splicesite-infile before I wrote this post, because it said:

"Note that it is better to use indexes built using annotated transcripts (such as_genome_tran or genome_snp_tran), which works better than using this option

So, is it still invalid to build such indexes like that?

or should I just use -ss --exon options to provide a list of exons and splice sites exacted from GTF file to build indexes?

[Devon Ryan]

That's referring to building the index with the --ss and such options. That's better than building the index without them and then giving hisat2 the same options.