HISAT2 V.S. STAR

Question

[langya]

Hey I used to use tophat for alignment and realized it is a bad tool to use now. i am wondering what are the advantages and disadvantages of these two methods (human genomes). Is there huge difference between the output of these two? Thanks!

Answer 1

[Devon Ryan]

The main benefit of hisat2 is that is uses fewer resources than STAR and that it can better handler known SNPs if you make the aligner aware of them. Aside from that, I essentially always get better results from STAR, which is why we use it in our standard pipelines instead of hisat2 (this also bears out in published comparisons).

[langya]

thanks! I will then stick to STAR and DESEP2 method. One more question is if i use DESEQ2, but i actually want to know the abundance of my transcript, in this case would you still recommend using RPKM from Cuffdiff? or other methods?

[Devon Ryan]

You can get normalized counts from DESeq2, so either use them directly or convert them to FPKMs or just divide by transcript length if you want some sort of length-normalized value.

[takeshi]

For reference, here is a published comparison: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5792058/

[Anand Rao]

• Based on their latest respective releases, do you still consider STAR better than HISAT2?
• What are the qualifications behind this assertion, are they different from before?

[Devon Ryan]

• yes
• At least STAR has added a plethora of new features

Answer 2

[Michael]

I have made the same principal experience as Devon. In addition, STAR seems to be much better suited to draft genomes and low quality genomes. The genome we are working with, for example, is of moderate size but consists of 33000 scaffolds, and contains ambiguity symbols. With STAR we on average achieve (unique) mapping rates > (90) 95% (<1% mismatches) and mapping speed of up to 400Mreads/h. With other aligners, which is mostly used on desktop computers, mapping rates are reported as low as 50%. Therefore, we have given up on Hisat, tophat & co.

[Anand Rao]

• Based on their latest respective releases, do you still consider STAR better than HISAT2?
• What are the qualifications behind this assertion, are they different from before?

[langya]

i see. because i have heard a lot of people saying that HISAT2 is better. i guess then I will just use STAR.

[Michael]

Better is depending on your data and requirements. Star can be a bit resource demanding, earlier versions crashed sometimes running out of memory. Also, you have to build the desired read length into the reference genome.