Exploiting open data: a new era in pharmacoinformatics.

Area of Science:

• Drug discovery and development
• Bioinformatics and computational chemistry
• Open science and data sharing

Background:

• Open data concepts have gained significant traction in drug discovery over the last decade.
• Public databases like ChEMBL and PubChem provide extensive bioactivity data.
• Platforms such as the Open PHACTS Discovery Platform integrate diverse life science data beyond simple ligand-target interactions.

Purpose of the Study:

• To evaluate the utility of large public bioactivity datasets for computational modeling.
• To highlight open-source initiatives in drug discovery data.
• To outline the possibilities and limitations of exploiting public bioactivity data.

Main Methods:

• Review of major public bioactivity databases (e.g., ChEMBL, PubChem).
• Analysis of data integration platforms (e.g., Open PHACTS).
• Assessment of data heterogeneity and assay description uniformity.

Main Results:

• Public databases offer a wealth of bioactivity data, increasing accessibility.
• Data heterogeneity and non-uniform assay descriptions are significant challenges.
• The utility of these datasets for computational modeling is questioned due to data quality issues.

Conclusions:

• While open data initiatives have democratized access to bioactivity information, data quality remains a critical concern.
• Further efforts are needed to standardize assay descriptions and improve data homogeneity.
• Careful consideration of data limitations is essential when using public datasets for computational drug discovery models.