Hey. I am working on smallRNA-seq data with the final goal of investigating differential expression analysis.

I used bowtie1 for mapping with the following specification: bowtie -p 10 -n 2 -m 1 -k 1 --strata --best -x index -q *fq

The summary statistics I got are as follows:

#reads processed: 2,00,1187
# reads with at least one alignment: 1,559,437 (77.93%)
# reads that failed to align: 441,750 (22.07%)
# reads with alignments suppressed due to -m: 1,368,859 (68.40%)

Given 68% of the reads were suppressed, I feel a huge chunk of the reads were non-unique and I'm not sure if retaining only the best 1 (-m 1 -k 1 --strata --best) is most appropriate. Please any suggestion from experienced smallRNA-seq persons will be appreciated.

Thanks