a beginner question in RNA-seq workflow - what is the next step?

Question

[galkimhi123]

Hi there! I have no formal training in bioinformatics, other than a course that i'm now taking, and I wanted to get some practice done. i'm a true beginner in RNA-seq, but a friend of mine who's working in the lab gave me his data from and RNA-seq experiment that he's working on. they are trying to find whether a certain transcription factor is responsible for certain cellular changes regarding leukemia, and the data was received from 6 different cells that had different mutations in the gene coding for that transcription factor.

now here's what i don't understand - i now have the gene counts after normalization in an excel file. what is the next step? how do we actually get a meaningful result from this data?

I understand - that must sound like a really stupid question but i'm just trying to learn... thanks...

[Buffo]

I think this protocol is one of the most useful for begginers (like me). Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks

[GenoMax]

At this stage in the game if you are going to recommend a protocol then let it be HISAT/StringTie one. TopHat is deprecated and even the developers of TopHat steer people away from that suite.

[Buffo]

The protocol, not the software. That paper describes what are you doing and why, something that most bioinformatic`s papers doesn´t have. If the software was a joke I don´t know.

[GenoMax]

Glad you clarified that point.

Figure 1 in the HISAT paper provides an orientation to the RNAseq data analysis protocol as well.

TopHat was the first (and for a while) the only "suite" capable of doing end-to-end analysis of RNAseq data. One could still use it but there are much better options available at this time.

[Buffo]

Once again thanks genomax2 I didn`t know that paper, and looks pretty good.

[galkimhi123]

thanks! i will try that

[GenoMax]

Please use ADD COMMENT/ADD REPLY when responding to existing posts to keep the threads logically organized.

[galkimhi123]

Thanks again for all the help! again - a question - is there a "standart" for RNA-seq analysis software? i mean, there are SO MANY software suites out there and it seems like it's really hard to keep up with the pace that new software are emerging.... i would guess that 20 different people would suggest 20 different suites... I am going to try and learn how to use HISAT (actaully HISAT2 is newer as i understand...) but by the time that i will be done there will probably be a newer software out there.... suggestions??

[GenoMax]

Outside of a commercial software package HISAT2/StringTie/Ballgown is about as close to a "suite" you are going to get for RNAseq data analysis. As long as your aligner is splice-aware (if that is a consideration in your organism) then you can use any (STAR/BBMap/bwa/bowtie2 are all great choices). featureCounts (and HTSeq-count) is the way to go for counting. Finally DESeq2/edgeR for the DE analysis.

Answer 1

[mastal511]

Use something like the Bioconductor package DESeq2 to find differential expression between mutant cell lines and wild type. If you have a look at the Bioconductor website you will find workflows for RNA-Seq analysis.