HISAT2 on Ivy Windows VM

* Please note that HISAT2 requires approval prior to installation on the VM

HISAT2 is a fast and sensitive tool for aligning short reads against the general human population
(as well as single reference genome). It indexes the genome using a Hierarchical Graph FM Index
(HGFM) strategy, i.e. a large set of small indexes that collectively cover the whole genome
(each index representing a genomic region of 56 Kbp).

HISAT2 Usage:

Alignment using HISAT2 is a 2-step process - indexing the reference genome, followed by aligning the sequence data.

  1. Create indexes of your reference genome of interest stored in reference.fasta file:
    hisat2-build [option(s)] <reference.fasta>
    This will create new files with the provided basename and extensions *.ht2. These files constitute the index.

  2. Align paired-end reads sampleR1.fq and sampleR2.fq to the reference genome indexed in the previous step, using N cores:
    hisat2 -x -1 <sampleR1.fq> -2 <sampleR2.fq> -p -S <output.sam>
    The alignment results in SAM format are written to the file output.sam

Note on using the --sra-acc option

Since Ivy VM’s do not allow outside connections, --sra-acc option will not work with HISAT2.
If users plan to use SRA data, they will have to download it and move into Ivy prior to alignment.
Please refer to our Globus user guide for help on how to do that.

More information

For detailed information, please refer to HISAT2 manual.

Citation

If you use HISAT2 for your work, please cite:
Kim D, Langmead B and Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nature Methods 2015