PyBioMed

PyBioMed computes molecular and interaction descriptors for chemicals, proteins, and nucleic acids to support molecular characterization and predictive modeling.

Key Features:

• Extensive Molecular Descriptor Calculation: Calculates 775 chemical descriptors, 19 types of chemical fingerprints, over 9,920 protein descriptors derived from sequences, and more than 6,000 DNA descriptors based on nucleotide sequences.
• Interaction Descriptor Computation: Computes interaction descriptors for pairwise samples using three distinct combining strategies.
• Integrated Data Analysis Pipeline: Provides APIs to acquire molecular data, perform pretreatment, represent molecules, and construct machine learning models.
• Online Molecular Object Download: Supports online downloading of molecular objects using different identifiers.

Scientific Applications:

• Molecular characterization and interaction studies: Enables calculation of descriptors and interaction analyses for chemicals, proteins, and DNA to support molecular characterization and interaction studies.
• Drug discovery: Facilitates descriptor-based analyses relevant to drug discovery workflows.
• Genomics: Supports genomic studies by providing DNA and nucleotide-sequence-derived descriptors.
• Proteomics: Supports proteomic analyses by providing protein sequence–derived descriptors.
• Systems biology: Enables integrative analyses by linking chemical and biological descriptor spaces.

Methodology:

Integrates molecular representation techniques with data mining algorithms and machine learning to handle large datasets and link chemical and biological spaces.