TAIGET

TAIGET identifies and annotates small-molecule protein targets to support target prioritization and interpretation in cancer-focused drug discovery.

Key Features:

• Docking Module: Predicts binding poses using AutoDock Vina and LeDock with inputs of UniProt IDs linked to PDB structures (up to five proteins) and ligands in SMILES format (up to three ligands).
• Target Screening Module: Applies a Bayesian-Gaussian Mixture Model (BGMM) to screen and prioritize potential targets from docking outputs.
• Target Annotation Module: Provides literature-derived annotations curated from over 14,000 cancer-related publications by three medical experts and ten medical students, associating drugs with genes, gene regulation, cancer outcomes, phenotypes, 2,170 cancer cell types, and 73 cancer types.
• PPI Network Visualization: Generates visualizations of protein-protein interaction (PPI) networks related to identified targets.

Scientific Applications:

• Oncology target identification: Enables identification, prioritization, and annotation of small-molecule targets relevant to cancer research.
• Hypothesis generation and validation: Combines docking predictions, BGMM-based screening, and literature annotations to support hypothesis generation and validation in drug discovery.
• Drug repurposing and target exploration: Facilitates exploration of novel therapeutic targets and repurposing of existing drugs using integrated computational and literature-derived evidence.

Methodology:

Computational docking with AutoDock Vina and LeDock, statistical target screening using a Bayesian-Gaussian Mixture Model (BGMM), and enrichment with curated annotations derived from over 14,000 cancer-related literature sources, with outputs including predicted binding poses, prioritized targets, and PPI network visualizations.