So I've run bowtie2 against my Trinity assembly and get a high proportion of paired end reads that align concordantly at more than one place. I am suspicious that they're aligning to multiple isoform variants all from the same "gene" group.
Fortunately, I've got a Trinity assembly that also includes a gene_trans_map that identifies which isoforms belong with which "gene" grouping.
How can I get bowtie2 to ignore isoform variants with a given "gene" group (defined by gene_trans_map) when it calculates the percent of reads that concordantly align exactly once or more than one time? I haven't been able to find anything like this in the help page but I would suspect this may be a common thing people want to do... I would prefer not to have to use the -a
parameter given how much this slows things down.
Thanks! This is just what I was looking for and does shift about 20% of my reads from the "aligned concordantly >1 times" category to the "exactly one time" category.