I've created a category called "genomics", which contains all the posts related to genomics.
Genomics
Computational biology and genomics
Computational biology and genomics
I've created a category called "genomics", which contains all the posts related to genomics.
Nice site with nice tips.
I'd like to ask and advice.
After having performed the alignment, what is the best way to evaluate the quality of mapping? Any particular tools one can use?
Hi Nosa,
Thanks for the compliment. It depends on what type of alignment you're talking about and what type of problem.
If you're talking about multiple sequence alignment, there's BAliBASE, which can be used as a benchmark for alignment tools. But that's to evaluate an alignment program. It's probably not what you're asking for but I thought I'll throw it in for good measure.
If you're talking about local alignment tools like BLAST, there's the E value.
If you're talking about short read alignments, have a look at the mapping quality. I wrote about mapping qualities here. There are tools such as SAMStat that can summarise mapping qualities from BAM files.
Hope that helps,
Dave
I was looking for a tool on how to assess the quality of the mapping reads. So SAMsat seems useful.
Thanks a lot for the response.
Hi nosa,
To get a sense of your read mapping quality scores (in phred scale, I'm assuming), check out FastQC. Once you get a sense of what your reads look like as far as quality is concerned, check out Btrim.
Btrim lets you set up search windows that average out your quality score and if a minimum threshold is reached for quality they get filtered out. The nice thing about Btrim is you can also adjust the minimum length of the reads you'd like. So it will only filter reads if it can also maintain a minimum read length you don't want to go under.
Good luck,
Alex P
P.S.
Awesome advice, Dave!!! Just found your blog and have benefitted nicely from it. Keep up the good work.
Hi Alex,
Thanks for the advice and the kind compliment! I'm glad it's been useful.
Cheers,
Dave
there are two that I know. I had a hard time working with RNAseqc though RSeqc works fine.
http://www.broadinstitute.org/cancer/cga/rna-seqc
https://code.google.com/p/rseqc/