Entering edit mode
8.5 years ago
statfa
▴
790
Hi, I'm going tow work on the analysis of RNA-seq data. I have been struggling a lot in these 2 days to find out how to analyze them.
I feel like "Detection of Differentially Expressed Genes" could be a good subject to work on. The problem is I don't know how to input the data in R packages.
Does anyone have the experience in using statistical R packages such as "GPseq", "DEGseq", "edgeR", "DESeq", "NBPSeq", etc.?
If I only have the fastq files, how should I input the data? I also appreciate it If you introduce me any other software which is more user friendly and is also based on statistical approaches.
Thanks
Would you happen to have a bioinformatician in the vicinity? You have to follow a pipeline to assemble/align and obtain counts before you can run statistical analyses. Given that you're focused on the statistics end of this, I'd recommend getting the alignment done by a bioinformatics professional. You can then work with them on obtaining the counts, and then on differential expression analysis.
Yes, Counts! they are exactly what I read in input commands... Do you think I can do it on my own? if there are any codes or softwares that are meant for this... Also, what more information/data do we need to align and obtain counts? Sorry for my simple questions, I'm an amateur...
Your best bet is to obtain a pre-built RNA-seq processing pipeline from another lab or from your bioinformatics core (if you have one). Something like this would process your fastq files into mapped bam files that can then be input into featureCounts to obtain the count information you will need to conduct differential gene expression analysis.
and what is "count" exactly? count raw reads?
Generally yes. Packages like DESeq2 expect you to provide raw read counts for genes/features.
You can do the counting on your own. featureCounts and HTseq-count are two popular counting programs, if you have aligned (BAM) data files. STAR aligner has options to do counts during alignment directly.
Counts are the number of mapped reads that appear in a specific genomic region.
@natalia: Is there a reason you are "bumping" your past posts up to main page?
Oh, no I didn't know I'm bumping them... How can I stop it? I was editing some parts of them and didn't mean to post them again on main page.
No problem. Editing an original post "bumps" it up to main page automatically. That is a feature of biostars. I was not sure if you were doing that intentionally.