Hello everyone!
I would like to compare DEGs across transcriptomic datasets and compare them with DEGs across proteomic datasets available for the same disease condition. Are there any standard protocols to follow for this specific purpose?
I have transcriptomics datasets from different tissues but for the same condition. Should I start with raw data and re-analyze the DEGs and then compare, or is there any other method I can follow?
Should I compare data from the same tissues after DEG together or can I compare from all tissues?
For proteomics data, the papers have given the DEG protein list. Again, should I start from raw data or use the list? What are the best parameters to consider? Like eg. MIST, fdr-p value, fold change?
How can I compare the DEGs from different transcriptomics datasets? Should I use a statistical test?
After meta-analyzing transcriptomics data and arriving at one single list of DGE genes, and similarly one meta-analyzed list of DGE genes from proteomics datasets, how can I correlate the data? Is there any statistical analysis or test I can follow?
If you can suggest me a paper or any material, even if it's a book, that'll be more than helpful!
Thank you so much, and if this post made you roll you eyes on how dumb it is, I'm sorry, its my very first time doing this kind of analysis and I simply can't find any proper review papers!
So I figured out the answer!
To compare any two gene lists you can do a hypergeometric test or a Fisher's exact test. I am using the R package GeneOverlap.
Hello! I’m currently delving into the realm of transcriptome meta-analysis and have encountered some issues that have been causing me quite a bit of distress, much like the ones you mentioned in your post.
I was wondering if the following approach is appropriate: first, using edgeR to normalize the counts for each individual dataset separately, and then integrating data from multiple studies for differential analysis.
Is it necessary for me to conduct separate differential analyses for each dataset? Since my research focuses on tumors, each dataset includes both tumor and non-tumor samples. As a beginner in meta-analysis, I’m feeling a bit overwhelmed.
Any guidance you could offer would be immensely appreciated. Thank you so much! Best wishes:)