How do research on RNA-seq analysis with simple and crude machinery

[ips]

I’m a graduate student study on induced pluripotent stem cell. I want to elucidate the molecular event happened after I added formaldehyde in the culture system of stem cell .I notice some research have done similar research based on RNA-seq(Do RNA-seq for each sample of stem cell before and after adding formaldehyde.And their raw data of RNA-Seq have been deposited in the high-throughput sequencing database of GEO).I wander what kind bioinformatics research I can do if I download their data?Plus, I only have a laptop with 2G RAM?

[marcora]

I’m a graduate student study on induced pluripotent stem cell. I want to elucidate the molecular event happened after I added formaldehyde in the culture system of stem cell .I notice some research have done similar research based on RNA-seq(Do RNA-seq for each sample of stem cell before and after adding formaldehyde.And their raw data of RNA-Seq have been deposited in the high-throughput sequencing database of GEO).I wander what kind bioinformatics research I can do if I download their data?Plus, I only have a laptop with 2G RAM?

I would recommend at least 4 Gb of RAM and a multiprocessor CPU (like Intel i7). I have been doing all of my work with such a machine (using Ubuntu Linux 64 bit) w/o any major problem.

Hope this helps!

[ips]

thank you very much

[simonandrews]

If you can get hold of the mapped data from the study you're interested in then you could try the software we develop to do your downstream analysis. It's specifically designed to work on normal desktop PCs. We're just finishing the analysis of a study containing 25 lanes of Illumina data, and that project will run on a 32-bit machine with 2GB RAM (just!).

You can have a look at the video we made which shows a simple RNA-Seq analysis which will give you an idea of what you might be able to do with it.

[Simon Anders]

For RNA-Seq data analysis, you don't need much RAM. My HTSeq tools, for example, have negligible memory footprint, because they work in a streaming fashion, i.e., loading reads one at a time (buffered, of course).

Most aligners process reads as a stream, too, but they need some RAM to keep the reference index (i.e., the genome) in memory. Even a vertebrate genome fits easily into the memory of a standard PC, but your 2 GB laptop might get to its limits. And, of course, you'll need a bit of patience if your processor is not the newest model.

Hence, getting your boss to buy you a newer PC might be a reasonable approach. You don't need anything fancy but a 2 GB laptop is stretching it.

[drio]

Consider uploading your data to the cloud (Amazon and others) and instantiate as much computer power as you need.