This is a test version of Biostars. For the public version, visit https://www.biostars.org.
HISAT2 - soft-masked genome from UCSC

Hi,

I recently started using genomes from UCSC, but it seems like they only have soft-masked and hard-masked. Obviously I do not want to use masking for aligning RNA-seq and just wanted to check whether HISAT2 treats the lower-case sequences like the upper case ones to allow mapping to the entire genome regardless of repetitive sequences.

I could not find this information in the documentation, sorry if I missed!

Thank you in advance.

rna-seq
• 4,907 views • link
updated by Matt Shirley Matt Shirley • written by lshepard lshepard

1 answer

If you take a look at the HISAT2 source code (which nobody should expect you to do) it appears that all FASTA characters are converted to their uppercase representation when reading the reference file:

# ref_read.cpp
while(c != -1 && c != '>') {
    if(rparms.nsToAs && asc2dnacat[c] >= 2) c = 'A';
    uint8_t cat = asc2dnacat[c];
->  int cc = toupper(c);
    ...
• 0 views • link
written by Matt Shirley Matt Shirley

Thanks Matt! It would have been nice to have this clear on the manual, but I have actually just noticed that some of their pre-built indexes are from UCSC, so that is re-assuring.

• 0 views • link
written by lshepard lshepard

Log in to answer this question.

More posts like this