obtain RNA-seq read counts online

Question

[statfa]

Hi,

I'm trying to figure out how I can obtain read counts from SRA files online... I have read some online tutorials (like this one: http://www.bioconductor.org/help/workflows/rnaseqGene/) but they require the user to download files... Downloading all of those large files and then aligning them to a reference genome to obtain BAM files on my computer is not possible. (My laptop isn't powerful enough for such heavy process and I also find it difficult to obtain BAM files on my own)...

I'm looking for an online solution to calculate read counts from SRA files without the need to download the files... Is there a user friendly software or something to use in this case?

I'm very new at the analysis of RNA-seq data and I have no experience in it. I would like you to keep the explanation simple enough, then I can comprehend it...

Thanks

[GZ1995]

Take a look at recent updated Recount. Some studies also deposit processed data (e.g., counts) in GEO.

[statfa]

If I get it right, "Recount" offers read counts of some studies at gene- or exon- level... It means that I don't have to calculate read counts on my own and I just need to download the per-obtained read counts from "Recount"? That's great... But there is a problem... first, I wish to learn how to obtain read counts for further researches... second, the read counts of the study I'm looking for is not in "Recount" database... how can I request for it?

[GZ1995]

First question, yes. Second question, if counts are not available in any public repository, you can contact authors or align and get the counts on your own. You can take other's suggestions (galaxy, light-weighting algorithm, etc.), or if you don't have much local computational resources, try it on cloud usually won't cost much. This tutor may be useful to start.

[VHahaut]

You should have a look to the galaxy project:

https://galaxyproject.org/

You can find there most of the tool you would use to analyze data from SRA to readCount without having to run the program in local.

[statfa]

Thanks... That's a good option... I'll try to find some tutorials on how to use Galaxy for this purpose... I'd be grateful if you already know any tutorial to share with me...

[VHahaut]

Sorry I don't usually use it. But I'm sure you can find some great one. If not just ask directly the question on biostar, some of the users will certainly help you.

[harold.smith.tarheel]

For Galaxy tutorials, including RNA-Seq, look here.

Answer 1

[igor]

You can try Kallisto:

The pre-print for Kallisto is reporting run times of 5 minutes for 30 million PE reads for the pseudo-alignment and transcript abundance measurements using the same Expectation-Maximization machine learning algorithm as RSEM. All this on a standard laptop. Not having to rely on your University’s computer cluster of hundreds of compute nodes or having to wait for runtimes of hours/days for RNAseq analysis is an obvious cost/time advantage for Kallisto over existing pipelines.

Tutorial using SRA data: https://benchtobioinformatics.wordpress.com/2015/07/10/using-kallisto-for-gene-expression-analysis-of-published-rnaseq-data/