Hi, I have a question about how to organize my data into the data 1, data 2, and replicates format required as input for FunPat.
My experimental design is a hybrid of scenario C (replicates at each time point) and scenario D (only one condition) of the vignette:
The goal of the study is to look at transcriptional changes in sciatic nerve in mice over time. We have four time points, comprised of 5 months, (baseline), 18 months, 21 months, and 24 months. Note that the time points are not even (i.e. the gap between 5 and 18 mos (months) is larger than between the other time points). I used DESeq2 to normalize counts and conduct DE analysis comparing two groups, but I'd like to use FunPat to look at patterns in gene expression over time which might identify genes relevant to aging for example.
Each time point has n = 6 biological replicates.
How should I construct my data 1, data 2, and replicates matrices?
I know that the number of rows will = number of genes. As for the number of columns...
data 1: 4 columns, each will be the same, i.e. the average expression at 5 mos (baseline) of the n = 6 replicates
data 2: 4 columns, the first = the expression (averaged across n = 6 replicates) at 5 mos.....the fourth column = the expression (averaged across n = 6 replicates) at 24 mos
replicates: again, we have n = 6 replicates for each of the four timepoints, including baseline. Would the replicates matrix be the exact same as my data 2 matrix, i.e. the rownames corresopnd to genes, and the columns correspond to the average expression of the 6 replicates for each time point including baseline? OR would this matrix only have one column, and equal the average expression (averaged over n = 6 replicates) just at baseline?
Or something else completely??
I was confused because for each example scenario in the vignette, the replicates matrix only has two columns.