Hybrid Generative AI for De Novo Design of Co-Crystals with Enhanced Tabletability

TL;DR

This paper introduces GEMCODE, a hybrid generative and evolutionary AI pipeline for rapid de novo co-crystal design targeting tabletability, significantly advancing pharmaceutical development through validated predictions and exploration of language models.

Contribution

GEMCODE combines deep generative models with evolutionary optimization for automated co-crystal design, enabling broader chemical space exploration and faster pharmaceutical development.

Findings

GEMCODE effectively predicts novel co-crystals with desired tabletability.

Experimental validation confirms the accuracy of GEMCODE's predictions.

Language models can generate plausible co-crystal structures.

Abstract

Co-crystallization is an accessible way to control physicochemical characteristics of organic crystals, which finds many biomedical applications. In this work, we present Generative Method for Co-crystal Design (GEMCODE), a novel pipeline for automated co-crystal screening based on the hybridization of deep generative models and evolutionary optimization for broader exploration of the target chemical space. GEMCODE enables fast de novo co-crystal design with target tabletability profiles, which is crucial for the development of pharmaceuticals. With a series of experimental studies highlighting validation and discovery cases, we show that GEMCODE is effective even under realistic computational constraints. Furthermore, we explore the potential of language models in generating co-crystals. Finally, we present numerous previously unknown co-crystals predicted by GEMCODE and discuss its potential in accelerating drug development.