Exploring Scoring Function Space: Developing Computational Models for Drug Discovery.

Area of Science:

• Computational chemistry
• Drug discovery
• Machine learning

Background:

• Scoring function space provides a systems-level framework for predicting drug molecule affinity.
• This approach is crucial for advancing drug discovery methodologies.

Purpose of the Study:

• To review the concept of scoring function space.
• To explore its application in developing machine learning models for protein-ligand binding affinity prediction.

Main Methods:

• Literature search of PubMed for "scoring function space" related articles.
• Utilized protein crystallographic structures from the Protein Data Bank (PDB) to represent protein space.

Main Results:

• Systems-level approaches offer a holistic view of receptor-drug interactions in drug discovery.
• Scoring function space introduces flexibility by framing drug discovery within mathematical spaces.

Conclusions:

• The scoring function space concept integrates various drug discovery methods.
• It facilitates a computational search within this space to identify optimal models for predicting receptor-drug binding affinity.