Hierarchical mapping of RNA-Seq data
1
0
Entering edit mode
3.9 years ago
Assa Yeroslaviz ★ 1.6k

Hi,

we're looking for a tool for Hierarchical mapping. Is there something like that out there? With Hierarchical I mean, i would like to start the mapping process with specific parameters and than iterate over the unmapped reads with increasing permissiveness (decreasing stringency). We mainly would like to better control the number of allowed mismatches and their allowed position in the reads. and the number of multi-mapped reads

The reason for this analysis is a data set of very shirt reads (~30-50nt), which have due to their biology high similarity. Due to the sequencing procedure we know that the used RT can cause mis-incorporations in specific positions, we would like to control for. Due to the high similarity of the reads we might get a lot of multi-mapped reads, which can differs only in one or two bases. This can cause problem in mapping those reads. We therefor would like to start very stringent and than re-run the mapping with different parameters, which can increase the results.

Does anyone know of a tool that can do this kind of analysis?

thanks in advance for any suggestions

Assa

mapping RNA-Seq Hierarchical • 781 views
ADD COMMENT
1
Entering edit mode
3.9 years ago
Rob 4.9k

Hi Assa,

I don't know of any tool that exactly does what you suggest. You might want to consider searching for "iterative alignment" or "iterative re-alignment", as searching the term "hierarchical mapping" in RNA-Seq will probably lead to a lot of hits for HISAT/HISAT2 (which is a great tool, but the hierarchy is in the index and doesn't refer to increasing permissiveness etc.). The closest thing I know of to what you're suggesting are the "2-pass" algorithms of HISAT and STAR. Here, the aligners will generally do novel splice site detection in the first pass, and will consider re-aligning certain reads in a second pass given the new "high-confidence" splice sites they have discovered. The permissiveness isn't exactly in terms of decreasing the matching requirements, but instead allowing spliced alignments in trickier situations where reads overlap splice sites in sub-optimal ways. These 2-pass algorithms do generally increase the mapping rate. Presumably, a similar type of strategy could be applied that would also iteratively increase permissiveness. However, I don't know of any off-the-shelf tools that do that.

ADD COMMENT
0
Entering edit mode

Thanks Rob. Yes I was thinking about different terms, but the term you offered gives me more of the regular alignment results, but not what I am looking for. I will continue looking for options.

ADD REPLY

Login before adding your answer.

Traffic: 2738 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6