This is a test version of Biostars. For the public version, visit https://www.biostars.org.
how to analyse gene expression using RNA-seq data

I want know how do we calculate gene expression using RNA-seq data available on NCBI. For example I want analyze gene expression of a family of transcription factors, I've the sequences of all these transcription factors and then i download the RNA-seq data from NCBI's SRA database. What do i do next in order to get the FPKM value for creating a heatmap. I am very new to Bioinformatics and learning on my own and also do not know any language except some Biopython and R. Thank you

rna-seq gene expression python

Check out RNA-seqlopedia, lots of information on understanding the process, not just the technical bioinformatics http://rnaseq.uoregon.edu

2 answers

You could visit a few RNA seq tutorials. They are by far the easiest to find in bioinformatics. Alternatively you can read through the manuals for the cufflinks suite.

http://cole-trapnell-lab.github.io/cufflinks/manual/

The basic workflow is Reads -> Mapping to genome (STAR/Tophat etc are good splice aware aligners for this job) -> Perform a guided/unguided transcriptome assembly (Cufflinks) -> Create a merged assembly from your samples (Cuffmerge) -> Quantify against a known reference (Cuffcompare)

If you found the sequences from SRA may be you could also find the associated BAM files and reduce your analysis time.

Thanks Vivek. That is exactly what I was looking for. Can you also explain what is guided/unguided transcriptome assembly?

In guided transcriptome assembly you give the assembler a GTF file of known transcripts/isoforms for the organism you are interested in to aid as a guiding reference. In unguided assembly, you don't which can lead the assembler into assembling more de novo isoforms, if that is your objective.

Alright I got it. That means I'll be doing a guided assembly.

Take a look at DESeq2. Here's a very nice tutorial from their authors : https://f1000research.com/articles/4-1070/v2

Log in to answer this question.