Hi,
I've RNA-Seq data from 40 cancer samples. 4 of them possess a mutation in a specific gene.
How can I check for pathways associated with this specific mutated gene that are dysregulated.. or more generally genes that might be dysregulated due to the mutated state of this gene? Maybe check for co-expressed modules where this gene of interest is?
I checked a little bit the literature and found:
- PARADIGM-SHIFT to be interesting.
- DESnowball may be also an interesting start.
- SPARCLE
- CaSPIAN might be interesting.
I don't want to reinvent the wheel so maybe you can help me with this type of analysis..
FYI : I've also control (healthy) RNA-Seq samples
Anyone has other ideas.
Thanks
•
written
by
Nicolas Rosewick
1 answer
Hi NicoBxl,
You can use iPathwayGuide for pathway analysis based on SPIA as mentioned by Sam above. The tool is based on Impact Analysis approach that considers the role, function, interactions between the genes on each pathway. In couple of weeks, you will also be able to readily identify the SNP-gene relationships as described in dbSNP in addition to miRNAs. This is the most advanced systems biology approach for pathway analysis today.
Our objective to is create a comprehensive multi-omic analysis platform for the users. Currently iPathwayGuide application is free to use and you can register here.
Please feel free to reach out to me if I could be of further assistance.
Best,
Arjun Gupta
arjun (at) advaitacorporation (dot) com
Linkedin
www.AdvaitaBio.com
Log in to answer this question.
So of your 40 cancer samples, only 4 of them possess the mutation?
yes only 4/40 have a mutation in this specific gene
With your sample size, a co-expression module analysis might not give enough power. However, you can try pathway analysis using tools like SPIA which can gives you perturbation of KGG pathways. You can compare the expression pattern of the mutated and non-mutated cancer and provide the log2 fold change to SPIA as an input (preferably do filtering be-forehead e.g. require a mean count of at least 10 etc.)
You can look into MSigDB Collections but 4/40 could be because of personal omics isn't it?