Hi everyone, I'm pioneering an RNA-seq experiment in my lab for a non-model bacterium, and my experience with RNA-seq analysis is limited. I've generated a data set of bulk time course RNA-seq data (7 time points and 6 replicates each) and gotten gene counts.
One of my goals is to cluster genes based on their expression changes over time. Looking at literature, ensemble clustering seems like a good way to do robust clustering analysis. However, all of the packages I've found are designed to cluster samples or cells in scRNA-seq experiments, rather than clustering the genes. I've found the tools to do individual clustering experiments, like K-means or hierarchical clustering, but with those you have to make some choices in cutoffs and numbers of clusters that I hesitate to make. Tools like the clusterExperiment/RSEC packages from BioConductor seem like a good way to go, since they can output a consensus cluster from multiple experiments. Those tools are geared towards clustering samples/cells, rather than genes, though.
Is there a way to modify those existing tools to cluster genes rather than samples (transposing the rows and columns feels really wrong but is it actually that simple)? Are there other tools I could look into? Or am I overthinking this and doing a few K-means clustering runs while manually changing the cluster number sufficient for this kind of analysis.
I appreciate any and all advice! Thank you
