Populating Chemical Space with Peptides Using a Genetic Algorithm.

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Organizing molecules by structure and properties is crucial in drug discovery using the concept of chemical space.
• Generating novel molecules within specific regions of chemical space, near existing molecules of interest, is a key challenge.

Purpose of the Study:

• To introduce the peptide design genetic algorithm (PDGA), a computational tool for generating peptide sequences.
• To enable the creation of diverse peptide topologies (linear, cyclic, dendritic) and nonpeptide molecules in proximity to a reference molecule within a defined chemical space.

Main Methods:

• Development of the peptide design genetic algorithm (PDGA).
• Utilizing the macromolecule extended atom-pair fingerprint (MXFP) to define chemical space based on molecular shape and pharmacophores.
• Generating peptide and nonpeptide molecules in proximity to a molecule of interest.

Main Results:

• The PDGA successfully generates high-similarity analogues of bioactive peptides with varied topologies.
• The tool can also produce nonpeptide target molecules.
• An interactive 3D map visualizes the chemical space accessible by the PDGA using MXFP properties.

Conclusions:

• The PDGA is a versatile computational tool for generating novel peptide sequences and related molecules.
• The PDGA facilitates exploration of chemical space, offering broad utility in drug discovery and molecular design.
• The generated molecules can be diverse in topology and chemical nature, expanding the accessible chemical space.