Computational approach for decoding malaria drug targets from single-cell transcriptomics and finding potential drug...

Area of Science:

• Computational biology
• Parasitology
• Drug discovery

Background:

• Malaria, caused by Plasmodium parasites, is a major global health threat.
• Plasmodium falciparum has developed resistance to existing antimalarial drugs, driving the need for new therapeutic agents.
• Machine learning and single-cell transcriptomics offer novel approaches for identifying drug targets and predicting new medicines.

Purpose of the Study:

• To develop a computational pipeline for identifying essential Plasmodium falciparum proteins from transcriptomic data.
• To predict novel drug molecules targeting identified key proteins.
• To discover lead drug candidates with desirable ADMET and drug-likeness properties.

Main Methods:

• Utilized a mutual-information-based feature reduction algorithm and classification to select important proteins from Plasmodium falciparum transcriptomic datasets (sexual and asexual stages).
• Constructed and analyzed Plasmodium falciparum protein-protein interaction (PPI) networks to identify vital survival proteins.
• Employed deep learning techniques for computational drug prediction based on identified protein targets and binding sites.
• Filtered predicted drug molecules for ADMET and drug-likeness properties.

Main Results:

• Identified key proteins crucial for Plasmodium falciparum survival through PPI network analysis.
• Computationally predicted a set of potential drug molecules targeting these key proteins.
• Reported lead drug molecules demonstrating favorable ADMET and drug-likeness profiles.

Conclusions:

• The study presents a generalizable computational pipeline for identifying drug targets from single-cell RNA sequencing (scRNA-seq) data.
• This approach facilitates the discovery of novel drug candidates to combat drug-resistant malaria.
• The identified lead compounds represent promising starting points for further antimalarial drug development.