Industrializing AI-powered drug discovery: lessons learned from the Patrimony computing platform

Area of Science:

• Computational drug discovery
• Pharmaceutical research and development
• Precision medicine

Background:

• A mid-size international pharmaceutical company launched a high-throughput computing platform, 'Patrimony', four years ago.
• The platform leverages proprietary and public data sources to support drug discovery initiatives.
• It aims to foster a Computational Precision Medicine approach powered by artificial intelligence.

Purpose of the Study:

• To identify novel therapeutic target candidates using the Patrimony platform.
• To document the transformational impact of this industrial computational platform on R&D.
• To demonstrate how the platform enhances competitiveness, saves time, and reduces costs in drug discovery.

Main Methods:

• Development and implementation of a high-throughput computing platform ('Patrimony').
• Integration of proprietary and public data sources.
• Application of artificial intelligence and model-based selection for therapeutic targets and drug candidates.

Main Results:

• Successful identification of novel therapeutic targets in immuno-inflammatory diseases.
• Ongoing expansion of applications to oncology and neurology.
• Demonstrated improvements in R&D competitiveness, time, and cost-effectiveness.

Conclusions:

• The Patrimony platform has had a significant impact on the company's R&D processes.
• Model-based selection of targets and drug candidates leads to more efficient drug discovery.
• Challenges in data governance, infrastructure, and user adoption were encountered but addressed.