Cheminformatics infrastructure for peptides, oligonucleotides, and ADCs requires urgent standardization

The pharmaceutical industry is rapidly expanding into new drug modalities such as therapeutic peptides, modified oligonucleotides, and antibody-drug conjugates (ADCs). Unlike traditional small molecules, which benefit from mature representation standards and robust data exchange, these novel therapeutics face significant challenges. Current cheminformatics infrastructure struggles to accurately capture their intricate structural and chemical complexity, leading to inefficiencies in drug discovery and development. This gap hinders interoperability and data governance, slowing the translation of promising research into clinical applications.

Drawing on their extensive experience at AstraZeneca, researchers examined the fundamental gaps in cheminformatics infrastructure for new drug modalities. They analyzed current approaches, specifically highlighting the limitations of systems like HELM (Hierarchical Editing Language for Macromolecules). Their methodology involved a critical review of existing representation standards and data exchange protocols, identifying where these systems fall short in handling the unique characteristics of peptides, oligonucleotides, and ADCs. The study did not involve experimental protocols or specific compound doses, but rather a comprehensive analysis of industry practices and technical constraints.

The analysis revealed that current cheminformatics systems exhibit significant limitations in representing new drug modalities. These limitations stem from both technical constraints and organizational gaps. Technically, new modalities exceed the scope of purely atomistic or sequence-based representations, which are adequate for small molecules or simple biopolymers but fail to capture the diverse modifications and conjugations common in peptides, oligonucleotides, and ADCs. Organizationally, there is a critical lack of unresolved standardization and governance across the industry. The authors argue that:

Local, proprietary solutions exacerbate data fragmentation and hinder interoperability, creating significant bottlenecks in drug development workflows. They found that existing systems struggle with the chemical complexity of these molecules, leading to unreliable data exchange and registration. This fragmentation impedes efficient research and development, making it difficult to share and integrate data across different platforms and organizations.