Experimental Design Given One Cell Line

[dbroh11]

We are interested in examining the effects of four different drug compounds (with 1 being a "control" in the form of a growth factor) on differential gene expression in a motor neuron cell line.

We are planning to split the same motor neuron cell culture into 3 separate dishes for all 4 drug exposure conditions (12 total dishes per experiment). We would then isolate RNA from all three dishes per drug, combine it and add a single unique barcode during the Total Stranded RNA-Seq prep for Illumina sequencing.

After reading many Anders posts, I see the inherent problem if we were to only test 4 total barcoded samples against each other (even though each is being comprised of RNA from different treated cell cultures) as ultimately it is one replicate per condition.

While biological replicates aren't possible in this case since it is one cell line being tested across conditions, would the solution then be to repeat this experimental design at three different time points, and then run the 12 barcoded samples (each representing aggregate RNA from three split cell cultures to a given drug response at one of the three time points)? Can we run the RNA-Seq analyses (either Tuxedo Suite or DESeq workflow) treating these as if they were biological replicates even though they are not?

i.e - Drug 1 - 3 barcoded samples (made from aggregate RNA from 9 total separate plates) from time point 1, 2, and 3... and same for Drug 2, 3, and 4?

Thanks!

Dave Brohawn

[bruce01]

I am having to design multiple cell line experiments having never done so before and have similar issues. I don't know much about cell culture but wanted to answer you with our strategy, maybe someone else has more to add.

I am not sure about your strategy of combining RNA from 3 replicates, it never seems like a good idea to combine individual extractions. My approach would be to use these as biological replicates, despite being cell lines I still view them as being distinct entities, and there will be some difference in expression between them, even in controls, possibly based on sequence error. However I expect the treated cells to cluster very closely in PCA plot or such.

The problem then becomes "how many replicates". Recent paper answers this reasonably well, and in the expected manner: "how many replicates can you afford?".

[mbblack]

Have a look at this statisticians discussion of replication with cell lines (since yes, conventional "biological" replication is not truly possible) - http://labstats.net/articles/cell_culture_n.html (see the "design 3 setup).

One thing I would suggest is a set of strictly vehicle controls, as well as your growth factor. It is always good to have a measure of "untreated" background expression.