DMinDD_​Project

Evaluate model with ROC curves and Scorer

Forest model with shortlisted Molecular Properties

This workflow adapts the KNIME workflow ....

Active, Inactive, Substrates, and Designer Drugs 2C9 Datasets

Step 2
Generate fingerprints and prepare data for ML

Step 1

Data Cleaning and prepare data for calculation of Molecular Properties

Step 4
Evaluate model with all descriptors

Step 5
Train Decision Tree with all the descriptors to determine the important Molecular Properties

Step 6
Evaluate DT model

Step 7
Add the missing Columns Actvitiy and the Fingerprints Morgan and MACCs

Step 8
Train ML model with only important Molecular Properties

Step 9
Evaluate model with shortlisted Molecular Properties

Step 10
Train ML model with only important Molecular Properties for the Designer Drugs Dataset

Step 11
Prepare data for collection

Ligand-based screening: Machine learning

With the continuously increasing amount of available data, machine learning (ML) gained momentum in drug discovery and especially in ligand-based virtual screening (VS) to predict the activity of novel compounds against a target of interest. In the following, different ML models are trained to discriminate between active and inactive compounds with respect to the CYP 2C9 isoform.

Fingerprint Attributes

Descriptor Attributes

Step 3
Train ML model with all Molecular Properties