This is a test version of Biostars. For the public version, visit https://www.biostars.org.
RNA-Seq:How to know RNA-seq data is Strand-Specific or not based on only fastq files available in public data repositories such as SRA?

How To Find RNA-seq data is Strand-Specific or not based on only fastq files from information available in public databases. I am looking for information because I might not know if such information is also deposited in public repositories. Right now I am interested in knowing this information about SRA/GEO dataset GSE72086?

rna-seq
• 4,097 views • link
updated by Shicheng Guo Shicheng Guo • written by ASid • ASid

5 answers

If you simply quantify the data with salmon using library type -l A, it will infer the most likely library format (stranded vs. unstranded) for you.

• 0 views • link
updated by ATpoint ATpoint • written by Rob Rob

If the library prep kit information is included, you can know from that. But it's probably easier to align and gauge strandedness empirically.

• 0 views • link
written by swbarnes2 swbarnes2

You will have to align the data (if the information about specificity can't be located). Then use infer_experiment.py included in RSeQC to find strandedness.

• 0 views • link
written by GenoMax GenoMax

This is probably the fastest method you'll find. Adding on this, once you quantified your data, you'll find in the specified output directory in the subfolder logs a log file in the specified output directory, which contains a line like:

[2019-01-04 11:18:39.780] [jointLog] [info] Automatically detected most likely library type as ISF

which tells you the inferred library type.

• 0 views • link
written by ATpoint ATpoint

the best way should be infer_experiment.py in RSeQC. 

pip install RSeQC 
infer_experiment.py -r knowngene.hg19.bed12 -i Aligned.sortedByCoord.out.bam

knowngene.hg19.bed12 is downloaded from table browse in UCSC and an example like this and knowngene.hg19.bed12 can be downloaded here

• 0 views • link
written by Shicheng Guo Shicheng Guo

Log in to answer this question.

More posts like this