HISAT2 command help

Question

[dina.hesham139]

Hey,

It's my first time to use HISAT2 for alignment and the manual is full of parameters that got me confused.

I need to write down a command that would include the following:

map against hg19, my samples are paired-end, I need to have xs attributes to the output, and I need the output to be compatible with Stringtie for downstream analysis.

Have anyone worked on this pipeline before and can help me?

Thank you!

[dina.hesham139]

For the --known-splicesite-infile, it was stated in the manual:

You can create such a list using python extract_splice_sites.py genes.gtf > splicesites.txt, where extract_splice_sites.py is included in the HISAT2 package, genes.gtf is a gene annotation file, and splicesites.txt is a list of splice sites with which you provide HISAT2 in this mode. 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.

Is using genome_tran index would substitute the genome+ the annotation gtf file parameters in Tophat?

[Devon Ryan]

Yeah, if you want to download the prebuilt indices then just get genome_tran and call it done.

[dina.hesham139]

Thank you!!!

[dina.hesham139]

One more question, for alignment, what would you generally recommend STAR or HISAT2?

[Devon Ryan]

I generally prefer STAR, though it requires significantly more RAM.

Answer 1

[Devon Ryan]

hisat2 -x /path/to/hg19/indices -1 sample_1.fq.gz -2 sample_2.fq.gz | samtools view -Sbo sample.bam -

The resulting BAM file should work with stringTie or cufflinks. You probably want the --known-splicesite-infile option though.

Edit: You probably need to sort and index the BAM file. At least cufflinks would require that.

[DareDevil]

Why can't use samtools sort to generate sorted bam file

hisat2 -x /path/to/hg19/indices -1 sample_1.fq.gz -2 sample_2.fq.gz | samtools sort -o sample.bam

[CandiceChuDVM]

Hi Devon, I work with dog genome. If I don't have premade indexes such as genome_tran or genome_snp_tran, should I go ahead to put my gtf file in the --known-splicesite-infile option or should I no using any gtf at all?

Thanks!

[Devon Ryan]

I think the GTF needs to be preprocessed first (there's a script for that that comes with hisat2), but aside from that yes.

[CandiceChuDVM]

Thanks for your suggestion! I will process it first with extract_splice_sites.py.

[fshimizu]

Hi Devon, could you tell me what is the meaning of the '-' symbol at the end of your samtools view command above (after 'sample.bam')? I am running a similar code without the '-', and sometimes it works fine but other times the program ends with a parsing error.. thanks!

[Devon Ryan]

- means "read from the pipe (|)" in this case. This is particular to samtools. If you're running samtools on a file then you would never use -.

[epigene]

hi Devon, what's the benefit of including --known-splicesite-infile option in the command line? if i'm only doing differential gene expression analysis, would using or omitting --known-splicesite-infile option make much difference?

[Devon Ryan]

You'll likely get slightly better alignments by using that and therefore slightly higher counts for DE testing.

[epigene]

thanks for the info! could you also comment on -k option, would you recommend using the default (5)? If there is other options I should use, I'd love to hear your thoughts on them. Thanks.

[Devon Ryan]

The defaults should be fine. The only other thing to change is the number of threads used.

[1769mkc]

How do I align paired end files , would it be the same command "-1 sample_1.fq.gz -2 sample_2.fq.gz" I have to give -1 and -2 ?