Hi All,
We have recently started a new transcriptomics experiment, where we analyzed five different plant tissues from the same organ on a HiSEQ machine. We used 4 individual plants to make 4 replicates, resulting in 4 x 5 = 20 samples. Denovo assembly was performed using Trinity. Now, RNA quality was ok, and also the inline controls and the number of (mapped) read suggest the run was ok.
The problem is exactly as the title states: high variabilty between replicates. Both, a cluster analysis and PCA suggest that there are not really "groups" of samples, neither replicates nor tissues cluster together. Maybe PC1 is associated with replicate and maybe PC2 is sligthly associated with tissues.
My questions are: What can i do to assess the problem ? Anyone had this before and could hint at a possible cause ?
On an unrelated issue: How can I handle the counts of different variants that are reported by Trinity ? Process them separetely, add them or means ?
Best Regards
Stefan
We have recently started a new transcriptomics experiment, where we analyzed five different plant tissues from the same organ on a HiSEQ machine. We used 4 individual plants to make 4 replicates, resulting in 4 x 5 = 20 samples. Denovo assembly was performed using Trinity. Now, RNA quality was ok, and also the inline controls and the number of (mapped) read suggest the run was ok.
The problem is exactly as the title states: high variabilty between replicates. Both, a cluster analysis and PCA suggest that there are not really "groups" of samples, neither replicates nor tissues cluster together. Maybe PC1 is associated with replicate and maybe PC2 is sligthly associated with tissues.
My questions are: What can i do to assess the problem ? Anyone had this before and could hint at a possible cause ?
On an unrelated issue: How can I handle the counts of different variants that are reported by Trinity ? Process them separetely, add them or means ?
Best Regards
Stefan