ExCAPE-DB

ExCAPE-DB aggregates and standardizes large-scale chemogenomics Structure-Activity Relationship (SAR) data to support development and validation of in silico target-prediction, polypharmacology, and off-target modeling.

Key Features:

• Extensive Dataset Compilation: Aggregates over 70 million Structure-Activity Relationship (SAR) data points from public databases such as PubChem and ChEMBL, including chemical structures, target annotations, and activity data.
• Industry-Scale Data Volume: Provides an industry-scale collection of chemogenomics data suitable for large-scale statistical and machine-learning analyses.
• Predictive Modeling Support: Supplies curated SAR data to build and train in silico models for predicting polypharmacology and off-target effects of chemical compounds.
• Benchmarking and Validation: Serves as a comprehensive reference dataset for validating and comparing cheminformatics algorithms and computational modeling approaches.

Scientific Applications:

• Target-prediction model development: Enables construction and training of computational models to predict chemical interactions with multiple protein targets.
• Off-target and polypharmacology analysis: Supports investigation of off-target effects and broad activity profiles relevant to drug safety and efficacy.
• Method and algorithm validation: Provides a large-scale benchmark for validating and improving cheminformatics tools and machine-learning algorithms.

Methodology:

Data curation and compilation of over 70 million SAR data points from public sources such as PubChem and ChEMBL.