13

I have run Oxford Nanopore Technologies' MinION sequencing on the same DNA sample using three flowcells, each aligned against the same reference genome (E.coli K12 MG1655) using both BWA MEM and GraphMap and stored as BAM files.

How can I quantitatively and efficiently analyse the quality of alignment (percentage identity, insertion rate, deletion rate) of each of these files?

Scott Gigante
  • 2,133
  • 1
  • 13
  • 32

1 Answers1

6

Qualimap will do this for you.

  1. Go to qualimap.bioinfo.cipf.es
  2. Run qualimap (default params are fine) on each BAM file
  3. Open up the HTML output, and you can read off the %identity (they measure the opposite, i.e. mismatch rate, but 100% - mismatch rate is %identity of course), indel rate, etc.

One thing to watch out for (you don't mention it in your question, but just in case) is that you cannot directly compare Q scores - these are a bit of a mess and calculated very differently in each piece of software.

Unsolicited suggestion: you might also try NGM-LR for mapping MinION data. We've found it beats the others for our data (though we map to a distant reference).

Scott Gigante
  • 2,133
  • 1
  • 13
  • 32
roblanf
  • 962
  • 7
  • 15
  • Thanks for the suggestion. To clarify, is this the aligner you were referring to? https://github.com/philres/ngmlr – Scott Gigante May 17 '17 at 04:25
  • Yep. Poster is here: http://schatzlab.cshl.edu/publications/posters/2015/2015.GI.NextGenMap-LR.pdf

    You could check out Sniffles too - detects structural variants from 3rd gen data. https://github.com/fritzsedlazeck/Sniffles

     – roblanf May 17 '17 at 05:16