This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Downsampling long-read BAM files

Im looking for the best way to downsample nanopore long-reads. Ideally I would want to downsample to a specific average read depth e.g. 5x

I have found a few methods e.g. using samtools but this seems to work for downsampling to a specific number or fraction of reads. Rasusa downsamples to a specific coverage /depth but requires fastq input and I would prefer to downsample my BAM files so I don't have to re-run my alignment on many samples again.

nanopore bam qc ont downsampling
• 3,035 views • link
written by eebloom • eebloom

You can use reformat.sh from BBMap suite. But this may not satisfy the "specific avg depth".

reformat.sh -Xmx10g in=your.bam out=sampled.bam parameter_from_below

Sampling parameters:

reads=-1                Set to a positive number to only process this many INPUT reads (or pairs), then quit.
skipreads=-1            Skip (discard) this many INPUT reads before processing the rest.
samplerate=1            Randomly output only this fraction of reads; 1 means sampling is disabled.
sampleseed=-1           Set to a positive number to use that prng seed for sampling (allowing deterministic sampling).
samplereadstarget=0     (srt) Exact number of OUTPUT reads (or pairs) desired.
samplebasestarget=0     (sbt) Exact number of OUTPUT bases desired.
• 0 views • link
written by GenoMax GenoMax

Okay, thanks. Just double-checking this will work with long-reads?

• 0 views • link
written by eebloom • eebloom

Yes it should work with long read BAM's.

• 0 views • link
written by GenoMax GenoMax

2 answers

This is not what I needed for this particular use case, as capping the coverage would lose the information on regions of copy number amplification / the variation in coverage across the genome

• 0 views • link
written by eebloom • eebloom

random downsampling of bam files is now possible using rasusa v1.0.0 with the aln subcommand

• 0 views • link
written by eebloom • eebloom

Is this what you used ultimately? Did this tool satisfy the "specific average depth" requirement?

You could accept (green check mark) this answer to provide closure to this thread. If @Pierre's answer was also useful then it can also be accepted.

• 0 views • link
written by GenoMax GenoMax

rasusa aln subsamples all sites to a specified coverage but it does not randomly subsample to a specified random coverage which is important to note if you are interested in preserving copy number alterations

• 0 views • link
written by eebloom • eebloom

Log in to answer this question.

More posts like this