Sorry if this is anyone's peeve question, but my search has largely turned up mixed responses.
I have a single-celled eukaryotic system with a plastic phenotype. I want to induce the phenotype to two different degrees and then have a non-induced control and do a DE. So that would be:
High treatment Low treatment Control
I know that gene expression for the response varies over time, but I'm less concerned with WHEN things are happening in a time series and more concerned with WHAT is happening in that time. I'm also a bit strapped for cash.
My solution to the latter point was to take several samples from different time points and pool those samples across time into samples by treatment to submit for sequencing. I would be replicating each of those pooled treatments three or four times, depending on if I can get the extra money for a fourth replicate.
I'm seeing conflicting opinions on pooling samples, but much of the concern seems to be pooling that negates replication through limitation of statistical robustness, and since I'm only concerned in comparing the treatment types (which I am replicating) I was unsure if I has an issue.
Any advice appreciated.
