Doubt about -k and --best options in bowtie 1

[diego diaz]

Hi all

I have a doubt about -k option in bowtie 1. According to the bowtie manual, this option is used to report up to n alignments, where n is the number given to -k.

I pose this scenario,

Bowtie configuration : -k 1 -- best

If I had, for example, a read with 5 valid alignments, and two of these aligments have the same score (best score).

Bowtie has to report best of these 5 alignments, but two of them are the best.

Does bowtie report these two alignments or choose one at random?


Thanks in advance!

[dpryan]

It chooses one at random (well, it probably just returns which ever one it found first).

[diego diaz]

Thanks for your reply.

Other question regarding this subject,

I have to quantify tRFs (tRNA-derived RNA fragment) levels in different conditions, but the problem is that tRNA genes are redundant, hence, several reads have multiple mapping.

Do you know some way to analyze this data? or some advice at least.

I have seen in some papers that the authors map the reads and report only one of the multiple alignments (-k 1 --best), but I'm not really sure that this is enough to quantify