search

alexdobin / STAR

RNA-seq aligner
MIT License
1.82k stars 502 forks source link

Use library information when outputting XS tag #24

Closed jdidion closed 9 years ago

jdidion commented 9 years ago

Currently, STAR only uses intron motif information to determine the strand output in the XS tag. Other aligners (TopHat, HISAT) allow library information to be explicitly specified. When aligning reads from a stranded library (fr-firststrand), the XS tags are highly discordant between STAR and HISAT. This is not an issue when using Cufflinks for assembly (since it allows the library type to be specified), but the new StringTie assembler does not have a way to specify the library type, and relies only on the strand reported in the XS tag.

alexdobin commented 9 years ago

This feature is high on my TODO list, hopefully will be added in the next month or so.

dolittle007 commented 8 years ago

I have exactly the same problem. I am planning to make a RNA-seq pipeline using STAR-Stringtie-edgeR. STAR is the best choice also due to GATK RNAseq pipeline is choosing STAR.

alexdobin commented 7 years ago

Hi @dolittle007

please do not comment on a closed issue, since I may not see it quickly enough. Is your data unstranded or stranded? For the unstranded data, you can use --outSAMstrandField intronMotif option to add the XS flag to all spliced alignments.

Cheers Alex

dolittle007 commented 7 years ago

I am handling stranded library. Stringtie doesn't allow the library type to be specified, and require "XS" tag no matter the strandness of reads is. So it forces me to use Cufflinks instead. Thank you Ting-You Wang