Question: Bowtie2: trying to implement position specific scoring options?
gravatar for brismiller
6 months ago by
Bellingham, WA, USA
brismiller10 wrote:

Hello all,

Question: Is there a way to have bowtie2 give an alignment a score boost if specific regions of the read mapped perfectly? For example, say I have a 24nt read and I am using bowtie2 in –very-sensitive-local mode, if there is a match from positions 2-8 (from the 5' end of my read) I want the match score to be +4, and any match outside of those positions gets the default match score of +2. Is this possible?

Context of my question: I have a sRNA-Seq dataset for Tetrahymena thermophila and I am trying to predict the targeted genes from a class size of siRNAs. Given that Tetrahymena does not have a scRNA database such as miRBase I plan to align the reads to genome looking for alignments to a given gene or its UTRs. As supported by this paper bowtie2 with the --very-sensitive-local parameter is the best option to use. Findings have shown that specific regions of sRNAs (miRNAs, siRNAs, etc.) dictate a sRNA-mRNA interaction more than others (see this paper). Thus, I would like to fine-tune the alignment scoring options if possible.

ADD COMMENTlink written 6 months ago by brismiller10

I think not easily. Of course, you can have a go at editing bowtie2 source, but that wouldn't be easy (at least for me). What you can implement more or less easily is filtering post-alignment using the CIGAR string.

ADD REPLYlink written 6 months ago by h.mon24k

Thanks h.mon this seems like the best option (CIGAR parsing)! I'll do a little more digging on how to do this and then post what I have found. Or if anyone else has done this before I would be happy to hear what they did.

ADD REPLYlink modified 6 months ago • written 6 months ago by brismiller10
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 812 users visited in the last hour