This repository is under the NCATS organization. It documents computational workflows, datasets, and results from a study focused on discovering potential inhibitors for the SARS-CoV-2 PLpro protein. The study uses machine learning (ML) models and high-throughput screening data to identify promising hit compounds.
Project Overview
The primary goal of this research work is to identify and validate compounds with potential inhibitory activity against the SARS-CoV-2 PLpro protein, an essential viral protein involved in viral replication. This work was conducted as part of the Antiviral Program for Pandemics (APP) project at NCATS.
Key Objectives
- Develop ML models based on high-throughput screening data to predict activity against PLpro.
- Perform virtual screening using optimized ML models to identify promising hit compounds from the in-house Genesis molecular library.
- Perform the similarity-based screening of the hit compounds.
Repository Contents
This repository contains the following:
KNIME Workflows
-
Hyperparameter Optimization Workflows:
random_forest_hyperoptimization.knwf: A workflow for hyperparameter optimization of Random Forest model, using different combinations of Avalon, Morgan, and Atom-pair fingerprints along with RDKit descriptors.gradient_boost_hyperoptimization.knwf: A workflow for hyperparameter optimization of Gradient Boosting model, with the same variations of fingerprints and descriptors.xgboost_hyperoptimization.knwf: A workflow for hyperparameter optimization of an XGBoost model, using the fingerprint and descriptor combinations mentioned above.
-
Virtual Screening Workflow:
-
virtual_screening_workflow.knwf: Represents the virtual screening workflow against the Genesis molecular library, using the best, optimized ML model for hit discovery.
Screening Data
Genesis_library_HTS_data/: Contains high-throughput screening data for a small subset of the Genesis library.primary_assay_data.csv: Data from the initial primary assay screening.confirmed_active_compounds_follow-up_assay.csv: List of active compounds confirmed by follow-up assays.
NPACT_library_HTS_data/: Contains high-throughput screening data for a subset of the NPACT library.primary_assay_data.csv: Data from the initial primary assay screening.confirmed_active_compounds_follow-up_assay.csv: List of active compounds confirmed by follow-up assays.
Training and Test Sets
training_set.sdf: Structure data file (SDF) containing the training set compounds used for model building.test_set.sdf: Structure data file (SDF) containing the test set compounds used for model validation.
Selected Hits From Virtual Screening
selected_screening_hits.xlsx: A file listing the top hit compounds identified from the virtual screening workflow.
Analog Compounds Similarity
analog_compounds.xlsx: A file listing analog compounds screened in this study, along with their Tanimoto similarity values.