Discovery informatics: its evolving role in drug discovery.

Area of Science:

• Bioinformatics
• Computational Chemistry
• Genomics

Background:

• Drug discovery generates vast, complex datasets, posing significant informatics challenges.
• Current methods struggle with homogeneous or well-defined data types.
• The field is shifting towards new approaches for knowledge extraction from diverse information.

Purpose of the Study:

• To highlight the unmet informatics needs in drug discovery and development.
• To discuss the paradigm shift in creating knowledge from data.
• To outline the future requirements for effective drug discovery informatics.

Main Methods:

• Analysis of current data generation and integration limitations in drug discovery.
• Identification of the need for handling heterogeneous, distributed data.
• Exploration of integrating domain-specific information (chemical, genomic) and unstructured data (textual, graphical).

Main Results:

• Accessing, searching, and integrating large, diverse datasets is a critical bottleneck.
• There is a growing need to move crucial development decisions earlier in the discovery process.
• Effective drug discovery requires leveraging data from both current and past projects.

Conclusions:

• The future of drug discovery informatics hinges on integrating heterogeneous and distributed data.
• Advanced mining and integration of chemical and genomic data are essential.
• Managing and searching diverse data types, including textual and graphical, will be integral to building knowledge bases.