Hello,
Maybe someone has an answer regarding the issue of multi-reads/multi-maps with Tophat.
With Bowtie, parameters can be set such that it only reports alignments in the specified stratum range using '-m'. Along with --best-strata I can force Bowtie to report only the best alignment and only if there is only one "best" alignment. (-m1 --best--strata)
My question is, how can I force Tophat to perform similar behavior? As far as I can tell there is no way to force Tophat to invoke Bowtie with --best-strata options...
I can get around this by running Bowtie alone before then using Tophat to map the remaining unmapped reads i suppose..
Tophat would then identify junction spanning reads, but as far as I can see it is only possible to prevent reporting of alignments when a read has more than x valid alignments- via the '-g' param.
Essentially, when a read maps to more than one location given a certain mismatch threshold, I wish to report the best alignment but only if the best alignment score is unique. This is a way to allocate reads given multi-maps.
Any thoughts?
Maybe someone has an answer regarding the issue of multi-reads/multi-maps with Tophat.
With Bowtie, parameters can be set such that it only reports alignments in the specified stratum range using '-m'. Along with --best-strata I can force Bowtie to report only the best alignment and only if there is only one "best" alignment. (-m1 --best--strata)
My question is, how can I force Tophat to perform similar behavior? As far as I can tell there is no way to force Tophat to invoke Bowtie with --best-strata options...
I can get around this by running Bowtie alone before then using Tophat to map the remaining unmapped reads i suppose..
Tophat would then identify junction spanning reads, but as far as I can see it is only possible to prevent reporting of alignments when a read has more than x valid alignments- via the '-g' param.
Essentially, when a read maps to more than one location given a certain mismatch threshold, I wish to report the best alignment but only if the best alignment score is unique. This is a way to allocate reads given multi-maps.
Any thoughts?
Comment