Machine learning to learn metabolite-histone-drug interactions in cancer.
There are 5 notebooks relevant to this analysis:
- PCC: Generates pearson correlation coefficient profiles between metabolite/PTM/drug datasets.
- TrainLASSO: Trains LASSO models using TensorFlow.
- EMT: Applies the trained LASSO models to EMT datasets.
- AggregateCoefficients: Aggregates coefficients from LASSO, stepwise regression, and k-TSP.
- SynergyAntagonism: Visualizes cell count percent changes using metabolite/drug profiles.