Optimizing Molecules using Efficient Queries from Property Evaluations
This work addresses molecule optimization for chemical discovery, offering incremental improvements over existing methods with demonstrated applications in drug and material design.
The paper tackled the problem of optimizing molecules for desirable properties by proposing QMO, a query-based framework that uses latent embeddings from an autoencoder, and showed it outperforms existing methods in benchmark tasks like improving drug-likeness and solubility, with significant gains in real-world tasks such as enhancing SARS-CoV-2 inhibitors and antimicrobial peptides.
Machine learning based methods have shown potential for optimizing existing molecules with more desirable properties, a critical step towards accelerating new chemical discovery. Here we propose QMO, a generic query-based molecule optimization framework that exploits latent embeddings from a molecule autoencoder. QMO improves the desired properties of an input molecule based on efficient queries, guided by a set of molecular property predictions and evaluation metrics. We show that QMO outperforms existing methods in the benchmark tasks of optimizing small organic molecules for drug-likeness and solubility under similarity constraints. We also demonstrate significant property improvement using QMO on two new and challenging tasks that are also important in real-world discovery problems: (i) optimizing existing potential SARS-CoV-2 Main Protease inhibitors toward higher binding affinity; and (ii) improving known antimicrobial peptides towards lower toxicity. Results from QMO show high consistency with external validations, suggesting effective means to facilitate material optimization problems with design constraints.