How To Use Tophat With Strand Specifc

Question

[camelbbs]

Hi,

While I use tophat to align rnaseq, if my data is strand-specific, do I need to add the option --library-type?

What is the default settings in tophat for that.

thanks,

Ch

[KCC]

What do you mean by strand specific?

[camelbbs]

The library was prepared by strand-specific assay

Answer 1

[Ido Tamir]

yes library-type. Then you need to know the way the library was prepared. Currently the dUTP protocol is mostly used and for this you need to specify --library-type fr-firststrand. Default is fr-unstranded.

[camelbbs]

Thanks a lot !! I found some public data don't have the library preparation info, such as ERP000546. If I don't know that info, how to set the parameter. Is there big affection to mapping result?

[Ido Tamir]

The reads will be always map to the same locations irrespective of correctness of library-type (not tested). Downstream analysis will be affected: cufflinks and htseq count when the data is in the wrong orientation or no orientation is given.

Map 1. with unstranded and then you see the result (i.e. if the data is really stranded in a genome browser).

[camelbbs]

Hi Ido, Thanks very much for your help. I just want to ask how to identify if the data are stranded through genome browser. I've mapped 1. with unstranded and got the bam file. I am using the IGV.