Hi, I have downloaded the raw_readcounts files (.txt files) from an RNA-Seq experiment using GSE id, NCBI. I have two patients' count data A13, A38. I want to perform a differential expression comparing these two patients' count data using galaxy. I can do it in R, but the task I've been given is to perform this using GALAXY. I appreciate your time for reading this.
Thanks, Sri
Hello,
The tools in the group NGS: RNA Analysis at Galaxy Main (http://usegalaxy.org) include a few software packages for differential expression analysis. I am fairly certain that none of these work with this particular input file format directly. Processing the counts with the tools provided in Galaxy from sorted BAM alignment data files (and often reference data) is how to prepare and process the data. You may want to try a few tool suites to see which is the best fit for your experiment, goals, and desired outputs.
Several tutorials covering RNA-seq analysis in Galaxy are here: https://galaxyproject.org/learn/
General help for inputs and common usage questions can be found here: https://galaxyproject.org/support/
Not all tools are covered in tutorials but the general usage in Galaxy is the same as it would be line-command. Meaning, set the same options on the tool form as you would set options in the command string. The original tool manuals plus the help on the Galaxy tool forms will guide. But please let us know if you have trouble using a specific wrapped tool and we can help.
Thanks! Jen, Galaxy team
Hi Jen,
Thanks for your response. I have the SRA file links for all the samples, just wanted to know if this works, so if I retrieve BAM files for SRA files by NCBI SRA tools do you think I will be able to get it done? One of my concerns is the time it's going to take, each SRA file is around 5GB.
Thanks, Sri
Hi Sri, Data of this size should not be a problem - much larger datasets process fine all the time. As for total run time, that depends on how busy the server is. If the data is relatively the same in content between the samples, running the tools on one group through the prep steps can give you an idea of how long the rest will take. You won't know the final job's runtime (the actual differential expression tests) until everything is merged, organized, and processed by that tool.
Give the public server a try first. If you decide later that you want to speed things up, the analysis could be moved to a local or cloud Galaxy where you control the resources (and are the only one competing for them!). https://galaxyproject.org/choices/
Please note that there are some job issues today (in most part localized to the Jetstream cluster) but these should be resolved soon or you can just avoid that cluster for the tools that allow "job resources" to be specified (most mapping and RNA-seq tools have this option as the last option on the tool forms). See the other recent posts here about job delays for examples of what is happening. The correction is not finalized but when it is we'll be posting back in those threads. And the whole system will be undergoing a reorganization in the next week or so to reduce/eliminate those issues going forward.
Jen
