I am working on a RNA-seq project involving 93 individual plants of the same species, with 31 genotypes and 3 clonal replicates/genotype. I already made libraries for each plant and am working with read counts. I am running a WGCNA with my data, specifically trying to link gene modules to certain plant traits. My issue is the following: my plant trait data was collected from a different set of plants and not from the plants that I collected RNA from. They are, however, the same 31 genotypes (just different individual clones). This means that I cannot assign a particular trait data point to a particular plant in my read counts table. I would have to rely on trait genotype averages and come up with some sort of merged read count table (where clonal replicates of the same genotype are merged). Would this be an acceptable practice in any way? I have not found any example of someone doing this. I don't even know how I would come up with a merged counts table (during preprocessing or after getting your raw counts?).
I apologize if the question seems convoluted. I've been trying to find an answer unsuccesfully for days, and don't know what else to do.
Ideally, you should do RNA-seq and trait on the same plants, but if you are doing averages, the data is kinda usable. The RNA-seq counts merging needs to be after preprocessing and normalization imho.