Development of synthetic chloride transporters using high-throughput screening and machine learning

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Molecular Biology

Background:

• Synthetic compounds for transmembrane chloride transport are crucial for treating diseases like cystic fibrosis.
• Identifying effective chloride transporters requires understanding key molecular features across diverse chemical structures.

Purpose of the Study:

• To identify novel scaffolds and molecular features for potent chloride transport using high-throughput screening and machine learning.
• To develop a generalizable model for predicting chloride transport ability across different chemical classes.

Main Methods:

• High-throughput screening of 1894 compounds to identify chloride transporters.
• Machine learning (ML) binary classification to determine critical molecular features for transport.
• Experimental validation of ML-predicted chloride transporters from the DrugBank database.

Main Results:

• 59 out of 1894 tested compounds demonstrated confirmed transmembrane chloride transport.
• MolLogP was identified as the most important molecular feature for predicting transport ability, though not sufficient alone.
• The best ML model successfully predicted and experimentally validated novel chloride transporters.

Conclusions:

• Machine learning combined with high-throughput screening is effective for discovering and optimizing synthetic chloride transporters.
• Understanding molecular features like MolLogP guides the design of potent compounds for therapeutic applications.
• This research provides a framework for developing advanced chloride transporters for various medical needs.