@jaybee84 suggested looking at the correlation of the viper protein-lv correlations. I think this makes sense as it might reveal why we are getting similar drugs from LV to LV.
Turns out, there are about 5 main clusters of viper proteins, which is kind of helpful, because it allows us to bin the LV-Viper correlation information into these five clusters.
I then did the same drug set enrichment analysis we previously did but instead used the average correlations across all LVs within the cluster, instead of looking LV by LV. Then just to tie it off I plotted the average expression of the lv's within each cluster by tumor type. So, for example, I think that this analysis shows that cluster 5 is expressed most highly in cNFs, so we could consider cluster 5 enriched drugs (which turn out to be all HDAC inhibitors) for those tumors. I think cluster two is another nice example - neurofibroma/mpnst seem to express this more than cNF, so perhaps they'd be sensitve to the cluster 2 enriched drugs (a mess of kinase inhibitors....let's ignore staurospaurine :) )
@jaybee84 suggested looking at the correlation of the viper protein-lv correlations. I think this makes sense as it might reveal why we are getting similar drugs from LV to LV.
Turns out, there are about 5 main clusters of viper proteins, which is kind of helpful, because it allows us to bin the LV-Viper correlation information into these five clusters.
I then did the same drug set enrichment analysis we previously did but instead used the average correlations across all LVs within the cluster, instead of looking LV by LV. Then just to tie it off I plotted the average expression of the lv's within each cluster by tumor type. So, for example, I think that this analysis shows that cluster 5 is expressed most highly in cNFs, so we could consider cluster 5 enriched drugs (which turn out to be all HDAC inhibitors) for those tumors. I think cluster two is another nice example - neurofibroma/mpnst seem to express this more than cNF, so perhaps they'd be sensitve to the cluster 2 enriched drugs (a mess of kinase inhibitors....let's ignore staurospaurine :) )