From intuition to AI: evolution of small molecule representations in drug discovery

Abstract

Within drug discovery, the goal of AI scientists and cheminformaticians is to help identify molecular starting points that will develop into safe and efficacious drugs while reducing costs, time and failure rates. To achieve this goal, it is crucial to represent molecules in a digital format that makes them machine-readable and facilitates the accurate prediction of properties that drive decision-making. Over the years, molecular representations have evolved from intuitive and human-readable formats to bespoke numerical descriptors and fingerprints, and now to learned representations that capture patterns and salient features across vast chemical spaces. Among these, sequence-based and graph-based representations of small molecules have become highly popular. However, each approach has strengths and weaknesses across dimensions such as generality, computational cost, inversibility for generative applications and interpretability, which can be critical in informing practitioners’ decisions. As the drug discovery landscape evolves, opportunities for innovation continue to emerge. These include the creation of molecular representations for high-value, low-data regimes, the distillation of broader biological and chemical knowledge into novel learned representations and the modeling of up-and-coming therapeutic modalities.

Protocol for creating representations of molecular structures using a polymer-specific decoder

“…These machine learning techniques for quantitative structure-property relationship are of ever-increasing interest and mostly just the starting point of every state-of-the-art lab, whether for efficiency enhancing design-of-experiment approaches or autonomous experimental platforms. 3 , 4 Therefore, the ability to utilize such foundational techniques should be of great interest to any scientist planning to optimize or explore new compound in the chemical space of molecules.…”

PaCh (Packed Chemicals): Computationally Effective Binary Format for Chemical Structure Encoding

“…In modern chemistry and data analytics, the proper encoding of molecular and reactional data holds paramount significance for efficient storage, retrieval, and processing. Accurate representation of chemical structures is essential for maintaining the integrity and reliability of vast databases used in various fields such as drug discovery, material science, and computational chemistry. − Proper molecular data encoding ensures a seamless exchange of information between different software tools and facilitates complex analyses and simulations.…”

Integrating Prior Chemical Knowledge into the Graph Transformer Network to Predict the Stability Constants of Chelating Agents and Metal Ions

“…20,21 Concurrently, due to the intrinsic link between molecular structure and its corresponding function, traditional molecular descriptors have primarily focused on capturing structural properties. 22 The combination of two-dimensional representations of molecular structures (such as various molecular fingerprints) with GNNs can further enhance the model's performance. 23 The introduction of the transformer in 2017 has revolutionized deep learning, particularly in natural language processing, computer vision, and biological sequence modeling.…”

“…Graph neural networks (GNNs) are a type of deep learning model specifically designed for processing various types of graph-structured data . Due to their excellent performance and ability to learn natural and expressive features from molecular graph representations using node and edge relationships, GNNs have been widely applied in different areas of drug design. , Concurrently, due to the intrinsic link between molecular structure and its corresponding function, traditional molecular descriptors have primarily focused on capturing structural properties . The combination of two-dimensional representations of molecular structures (such as various molecular fingerprints) with GNNs can further enhance the model’s performance …”