Design of Natural-Product-Inspired Multitarget Ligands by Machine Learning

Area of Science:

• Computational chemistry
• Drug discovery
• Pharmacology

Background:

• (-)-Galantamine is a natural product used to treat Alzheimer's disease.
• Existing treatments often target single pathways, highlighting the need for multitarget approaches.

Purpose of the Study:

• To identify novel mimetics of (-)-galantamine using a machine learning-based virtual screening protocol.
• To discover compounds with polypharmacological profiles relevant to Alzheimer's disease treatment.

Main Methods:

• A fully automated virtual screening protocol employing machine learning models.
• Identification of compounds with bioactivity against macromolecular targets of (-)-galantamine.

Main Results:

• Eight compounds were identified with bioactivities on at least one target of (-)-galantamine.
• Two identified compounds demonstrated an expanded spectrum of bioactivity on targets relevant to Alzheimer's disease.
• These hits are suitable for further hit-to-lead expansion in drug development.

Conclusions:

• Advanced virtual screening protocols using chemically informed machine learning can effectively identify drug candidates.
• The findings support the strategy of multitarget drug design for complex diseases like Alzheimer's.
• The identified compounds represent promising starting points for developing new Alzheimer's therapeutics.