Hi,
I have seen that there are two sets of differential analysis on the reprocessed data. One with GC content and gene length variable considered and one without. I am aware that the final version too the two variables into consideration. I just wanted to know what difference did it make to the analysis with or without these two variables .
Also in the differential expression file there is a B value. What does this value mean for the analysis.
Thanks
Devika
Created by Devika Agarwal dpag0891 Hi @dpag0891
The comparison was done as Control-AD. Since these are logFC simplest way to look at this as subtraction between median logFC, so if the number is +ve then up in control comparing to AD, negative vice versa.
Hi Thanneer @th_vairam
Thank you for your reply. I had another query,
if the DEG of comparison is done for Control-AD and then the direction of the differential expression is up . Then is it up in Control or in AD samples? As I assume the log2 fold change is done for AD vs the control. But I wanted to be sure that I am interpreting the differential expression the right way in the files.
Hi @dpag0891
CQN based GC content and gene length adjustments control for false positives when performing differential expression in both the ends of the spectrum (i.e., long-short gene length and high-low gc content). For more detailed explanation I recommend you to look at this article https://academic.oup.com/biostatistics/article-lookup/doi/10.1093/biostatistics/kxr054
And the B-statistic is the log-posterior odds of differential expression.
@th_vairam
Drop files to upload
syn8468023.10 question regarding the differential expression analysis page is loading…