Data-Driven Drug Discovery Network (D4N)
The Data-Driven Drug Discovery Network (D4N) is an initiative by researchers from Leiden University and collaborators to join efforts in applying and developing novel techniques from data science to drug discovery and related topics from bioinformatics.
- Gerard van Westen
Over the last decades a significant amount of data has been produced on the bioactivity of chemicals in pharmaceutical research. Recently more and more of this data is being made available publically through public private partnerships and open access work. Building on this, it has now become feasible for the first time to steer drug discovery projects starting from public data. To efficiently reach this aim we want to combine insights and expertise from both computational scientists and pharmaceutical scientists.
Image: The crystal structure of the adenosine A2A receptor (‘caffeine receptor’ , indicated in brown) with a designed ligand bound to it (ZM241385, shown in cyan). The A2A receptor is a relevant target in the treatment of, among others, Parkinson’s disease and cancer. (image by E.B. Lenselink)
Innovation through collaboration
The goal is to stimulate collaboration and with it innovation on the borderline between disciplines:
- Mathematics and Statistics
- Medicinal Chemistry
These collaborations can take many forms, for instance joint supervision of MSc or PhD Students, organizing joint symposia or interaction meetings, and in general to strengthen the impact of research in this interdisciplinary area.
Importance of Data-Driven Drug Discovery
Drug discovery has traditionally been a branch of science that embraced the use of computational methods. This is partly driven by an effort to reduce the costs of preclinical drug discovery but mostly it aims at understanding the complex multidimensional problem that is drug discovery. (Public) data has largely increased, but its effective use requires new computational and statistical methods. Besides there is a strong need to bridge the expertise in data science and drug discovery, in order to make proper use of novel techniques and identify relevant research questions.
Data-Driven Drug Discovery represents a different perspective on the use of large and variable data in drug discovery. The philosophy is that data that might not be of direct use in a project can prove very useful in related projects. Data that is generated in drug discovery and is commonly ignored when deemed not directly useful in a project should be stored and can be made available for sharing in other projects. Moreover, a thorough data science based exploratory effort should precede any project rather than a smaller scale literature study as is now customary. D4N was started to facilitate these two goals.
Leiden's strong expertise in life-sciences
Leiden University together with the Bio Science Park have a strong expertise in life-sciences. Moreover, the informatics and mathematics department in Leiden focus in their applications on these fields. To exploit this synergy for the important topic of drug discovery, the D4N provides a platform where researchers from Leiden, but also external collaborators, can benefit from this unique setting.
- Andreas Bender, Cambridge University
- Iryna Yevseyeva, De Montfort University, Leicester
- Richard Allmendinger, Manchester University
- Hugo Gutiérrez de Teran (Uppsala University)