Entering edit mode
2.8 years ago
Herman Einstein
•
0
I am studying the similarity of two different cell types by their gene expression (RNA-seq). I am going to subset a group of genes (using all the genes may be unreasonable) and use spearman correlation (RNA-seq data is normalized by rlog) to see their similarity. May I know whether I am correct to do such an analysis?
The answer depends on the biological question you want to ask. What are you trying to achieve by running this correlation analysis? Are you looking for a global similarity measure, or are you trying to detect differentially-expressed genes? Please explain the purpose of the analysis.
I would like to know how close the two samples are. I can see that they are very similar by looking at the heatmap which is subjective. However, I would like to tell the exact similarity based on some analysis (i.e.correlation analysis) which is more objective, to illustrate that they share the same expression pattern no matter one's expression level is much higher than the others. p.s. I am not finding DEGs