Calling indels against a very small reference sequence?

[sjuzhet]

Hi y'all,

So I'm trying to detect a CRISPR-induced indel in a specific region of the genome, so I only sequenced that particular region and I was planning on using that specific region as a reference sequence, rather than the whole genome. However, most workflows I've researched involved calling indels against large genomes - which is probably gross overkill for my application.

Does anyone have any thoughts on what would be the best way to detect what percentage of my reads contain an indel within a 500-bp region?

[HESmith]

It would be easier to provide suggestions if you described 1) how you selected the region of interest (amplicon? oligo capture?) and 2) the sequencing platform and type of data you generated (Sanger? Illumina PE-250?).

[sjuzhet]

They're PCR amplicons, yes, and sequenced on a MiSeq. Several different cutting conditions were multiplexed, and I was hoping to compare them all against the reference sequence.

[GenoMax]

Try BBMap and align to the region of interest (you will need to make BBMap index). Using appropriate maxindel= (and any additional parameters that make sense).

[HESmith]

Or, assuming the paired-end reads overlap, you could use BBMerge (part of BBMap) and filter by sequence length. Any merged reads that are shorter than the predicted amplicon length contain deletions.

But BBMap alignment is fairly quick and easy to use, so that's a valid option.