This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Can I convert HTSeq count into RPKM or TPM value or standard unit of RNA-Seq

Now, I'm comparing RNA expressions that have RNA-Seq and HTSeq count How can I interpret it together with different unit or Can I convert HTSeq count equivalent RNA-Seq? or if you have other suggestions, please feel free to suggest me.

expression rna htseq
• 4,822 views • link
updated by Marco Pannone • Marco Pannone • written by wathunya.h • wathunya.h

Already answered here

• 0 views • link
written by Shred Shred

It's not very clear what you are trying to achieve. Generally, if you are comparing expression values (in an RNA-seq dataset) across different sample groups you want to normalize by using for example median of ratios (default DESeq2 normalization method for DE analysis). RPKM/FPKM and TPM are more suitable for gene to gene comparison within the same sample.

• 0 views • link
written by Marco Pannone • Marco Pannone

Thanks for the information. Now, I have datasets from different databases such as cBioportal, GEO database. Dataset in cBioportal give mRNA expression as RNA-Seq v2 but some datasets in GEO database give RNA-Seq-htseq count because of analysis with high-throughput methods. For my questions and doubts, I want to compare different datasets in DE analysis. Can I convert RNA count from htseq-count equivalence to same RNA-seq dataset or should not convert and compare direct (head-to-head)?

• 0 views • link
written by wathunya.h • wathunya.h

As far as I understood, I would suggest that in all datasets where you have retrieved raw data you process them until generating a counts matrix with htseq count. This way, at the end, you have counts matrices generated by htseq count for all datasets. However, keep in mind 2 important factors:

  1. If you want to be as accurate as possible, you should make sure all datasets at the end are processed the sam way, starting from raw data through alignment etc.;
  2. htseq count does not normalize the read counts. Normalization happens in downstream analysis, for example while performing DE analysis in DESeq2 which normalizes the counts by median of ratios as default.

Hope I managed to help somehow!

• 0 views • link
written by Marco Pannone • Marco Pannone

Thanks a lot for the useful suggestion. You are so helpful.

• 0 views • link
written by wathunya.h • wathunya.h

0 answers

No answers yet.

Log in to answer this question.

More posts like this