Hello. I want to align my 21bp RNA-seq data to the grape mRNA database. I want to use Bowtie to do the align but I never used it before. Does any one here has experiences on this software?
It all depends on your data and goals, and you may elaborate a bit so we know what you would like to do. You'll have to adjust the parameters to accommodate your needs. For example, if you want to do expression profiling, mapping accuracy may be less important than doing variant calling. Also, the parameters are also dependent on your data, like single-end or paired-end, base calling qualities (do you need to trim bad bases?), etc.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Thanks. Vitis. Our RNA-seq data are from cDNA laibray. Each read is single and only 21bo. THe main propose of this work is to do the expression profiling analysis. We already done the trim.
I did something very similar in another plant organism for expression profiling. We used -n 2 -e 70 -l 25 and -m 1 to suppress non-uniquely mapped results (multiple mapping turned out to be low at around 4%). I think if you already trimmed low quality bases, you may use the -v mode instead.
Thanks for your response. Actually we trid -n and -v mode with the similar paramters you suggested. But the percentage of aligned reads always be 20-25%. There is different type of grape we used compare with reference, but we can't do anything with this low aligned results.
It's a bit surprising that the mapping rate is that low. We've got close to 80% mapping rate for the similar experiment. Do you have any idea how distant the one you sequenced is related to the reference? We also mapped reads from closely related species before, but we estimated the divergence before mapping. Basically, you'd have to do a de novo assembly to assemble the reads, and blast/blat back to the reference to see how divergent they are, then determine your bowtie parameters.