Ligand-receptor pairing via tree comparison

Area of Science:

• Computational Biology
• Bioinformatics
• Drug Discovery

Background:

• Drug discovery involves mapping cell receptors to ligands, a complex and costly process.
• Existing tree comparison methods assume predefined node mappings, which is unsuitable for receptor-ligand interactions.

Purpose of the Study:

• To introduce a novel class of tree comparison problems for scenarios where node correspondence is the primary challenge.
• To formulate combinatorial optimization problems for mapping coevolving biological classes, such as receptors and ligands.

Main Methods:

• Formulation of novel combinatorial optimization problems based on biological mapping needs.
• Analysis of problem hardness and development of efficient algorithms for specific cases.

Main Results:

• Established hardness results for the newly defined tree comparison problems.
• Developed and demonstrated an efficient algorithm for a restricted version of the problem.
• Showcased the applicability of the developed methods in the context of drug discovery.

Conclusions:

• The novel tree comparison framework effectively addresses the challenge of mapping coevolving biological entities.
• The developed algorithms offer practical solutions for receptor-ligand mapping in drug discovery pipelines.
• This work opens new avenues for computational approaches in understanding molecular interactions.