Fighting high molecular weight in bioactive molecules with sub-pharmacophore-based virtual screening

Area of Science:

• Computational chemistry
• Medicinal chemistry
• Drug discovery

Background:

• Traditional virtual screening methods face challenges in identifying novel bioactive molecules.
• Medicinal chemistry programs require efficient strategies to discover starting points for drug development.

Purpose of the Study:

• To introduce and validate a new subpharmacophore-based virtual screening method.
• To assess the method's ability to identify small bioactive molecules from larger ones.
• To compare the subpharmacophore approach against chemical fingerprint descriptors.

Main Methods:

• Assembled a dataset from the ChEMBL database for 133 targets (450-850 Da).
• Derived subpharmacophore descriptors (5-7 points) from query pharmacophores.
• Screened 1079 test sets (decoys and spikes, <400 Da) using subpharmacophore descriptors.
• Compared performance against three chemical fingerprint descriptors.

Main Results:

• The subpharmacophore approach successfully identified active molecules for 85 out of 133 targets.
• This method outperformed traditional chemical fingerprint descriptors in detecting active molecules.
• Applied to a medicinal chemistry program, it yielded new molecules <400 Da from an 800 Da lead.

Conclusions:

• Subpharmacophore-based virtual screening is a powerful technique for discovering small bioactive molecules.
• This method offers a significant advantage over chemical fingerprints for lead identification in drug discovery.
• The approach effectively addresses the need for lower molecular weight drug candidates in medicinal chemistry.