ichespa

ichespa evaluates chemical space for small molecules by predicting and analyzing molecular descriptors—including predicted chemical properties, molecular substructures, AI-based chemical space definitions, and chemical class ontology—to assess mass spectrometry (MS)-based observability and ionizability.

Key Features:

• Customizable Chemical Space Definitions: Allows definition of chemical space parameters tailored to specific compounds and research goals.
• Clustering and Visualization: Applies clustering algorithms to organize compounds within defined chemical spaces and visualizes trends and patterns.
• Molecular Descriptor Prediction: Predicts molecular descriptors including predicted chemical properties, molecular substructures, AI-based chemical space definitions, and chemical class ontology.
• MS-Based Observability Analysis: Evaluates how chemical space definitions correlate with mass spectrometry (MS) observability and ionizability to inform detectability.

Scientific Applications:

• Improved Experimental Design: Supports a priori selection of experimental conditions by predicting sample composition and observability characteristics.
• Candidate Structure Reduction: Filters candidate structures by predicted likelihood of ionization to reduce search space during compound identification.
• Enhanced Identification Confidence: Decreases false discovery rates and increases confidence in small-molecule identifications through observability-informed analysis.

Methodology:

Integrates computational prediction of molecular descriptors, uses clustering algorithms and visualization tools to organize and interpret chemical spaces, and incorporates empirical data such as the U.S. EPA nontargeted analysis collaborative trial for validation.