search

vahuynh / dynGENIE3

Semi-parametric approach for the inference of gene regulatory networks from time series of expression data
21 stars 17 forks source link

Advice on how to input different conditions and replicates #2

Closed rbatorsky closed 2 years ago

rbatorsky commented 4 years ago

Hello,

I was hoping that you would have time to give some advice on how to input different conditions and replicates to dynGENIE3. I'm successfully running the R implementation on my time series RNAseq data. However, I'm not quite sure the recommended way to input the different replicates and conditions.

My data is: 6 time points 8 conditions plus control 2-3 replicates per condition.

The way I am doing it now, I am inputting each condition and replicate as a separate time series and treating control as a separate condition and running them all at once. Do you recommend this, or to separate conditions into different runs and perhaps treat the control as steady state data?

Thank you for the helpful tool! Best, Rebecca

vahuynh commented 4 years ago

Hello Rebecca,

Apologies for the very late response.

I think that the best solution would be to have one time series per condition (using indeed the control as a separate condition) and use the mean (or median) expression over the replicates at each time point.

Best, Vân Anh