Tweet
Hi,
We have two different times of 6-sample RNA-seq(each time cells from different donors and library construction maybe also different). The first time had 12 samples (6-sample for technical replicates) and the second time had 6 samples. Let's say they are A1,B1, C1,D1,E1,F1, A1',B1', C1',D1',E1',F1' and A2,B2,C2,D2,E2,F2.
Every single time 6-sample relationship mapped to our experiments, and there was a good correlation(R2 around 0.95) between technical replicates (e.g., A1 vs. A1'). But there was a LOW correlation (ranging from 0.5 to 0.7) between the biological replicates(e.g., A1 vs. A2), again which were from different donors. My question is if I should combine the 24 samples to do differential expression, e.g., (A1,A1',A2) vs. (B1,B1', B2).
Any insight would be acknowledged!
Macch
We have two different times of 6-sample RNA-seq(each time cells from different donors and library construction maybe also different). The first time had 12 samples (6-sample for technical replicates) and the second time had 6 samples. Let's say they are A1,B1, C1,D1,E1,F1, A1',B1', C1',D1',E1',F1' and A2,B2,C2,D2,E2,F2.
Every single time 6-sample relationship mapped to our experiments, and there was a good correlation(R2 around 0.95) between technical replicates (e.g., A1 vs. A1'). But there was a LOW correlation (ranging from 0.5 to 0.7) between the biological replicates(e.g., A1 vs. A2), again which were from different donors. My question is if I should combine the 24 samples to do differential expression, e.g., (A1,A1',A2) vs. (B1,B1', B2).
Any insight would be acknowledged!
Macch