Learning analysis of RNA-Seq data from assembly to final publishable

Question

[drragill]

Hi, I am new in the field of RNA-Sequencing. I wanted to learn the RNA-Seq data analysis starts from sequencing to final publishable form. What combinations of softwares I should start learning? Please tell me from simple to complex. Thanks!

[steve]

one of the first results on Goolge: http://biocluster.ucr.edu/~rkaundal/workshops/R_mar2016/RNAseq.html

[drragill]

Many thanks for your kind response!

Answer 1

[ATpoint]

My favourit workflow uses salmon/tximport/DESeq2 for standard differential analysis, see this workflow. If you are new to RNA-seq, also make sure to read this guide written by some of the experts in the RNA-seq analysis field.

I personally recommend against the "new tuxedo" pipeline suggested by Buffo as it involves a transcript assembly step. For well-annotated organisms like mouse or human I simply do not see the need to do that. Well-maintained transcriptomes are already present. It only creates unnecessary complexity given that potential new transcripts are not reliable as they (at least in my opinion) would require experimental confirmation which is beyond the scope of a typical RNA-seq experiment. The pipeline linked about (salmon) is computationally inexpensive, therefore fast and has an elaborate way to deal with multimapping reads plus corrects for potential GC bias (see the alpine package at Bioconductor for more invormation on that).

[drragill]

Many thanks for your kind suggestions!

Answer 2

[Buffo]

The new tuxedo protocol is probably the most popular pipeline for RNAseq analysis.

[drragill]

Many thanks for your kind response!