ATOM Modeling Pipeline (AMPL) for Drug Discovery, a Hands-on Machine Learning Tutorial

About this Class

The workshop focuses on using the Atom Modeling PipeLine (AMPL), an open-source conda-based software that automates key drug discovery steps. AMPL is designed to take molecular binding data (IC50, ki) and carry out the ML steps with minimal user intervention. The workshop will demonstrate data ingestion, cleaning and curation on AMPL. This 90-minute workshop will use Google COLAB notebooks. 1. Notebook 1: Ingestion, Cleaning and Exploratory Data Analysis of Binding Assay Data 2. Notebook 2: Standardization of SMILES, Featurization and Compound Overlap Diversity 3. Notebook 3: Curate and Merge Datasets to Create the Final ML-ready Dataset Instructor: Sarangan Ravichandran, Ph.D., PMP, Senior Data Scientist, Frederick National Lab for Cancer Research