Gerard van Westen appointed as full Professor of Artificial Intelligence and Medicinal Chemistry
Gerard van Westen has been appointed full Professor or Artificial Intelligence and Medicinal Chemistry. This chair has been jointly created by the Leiden Academic Centre for Drug Research (LACDR) and the Leiden Institute of Chemistry (LIC) as part of SAILS, the university stimulus program in Artificial Intelligence.
Gerard van Westen has been working on the application of various machine learning methods in pre-clinical and early-clinical drug discovery, and on the application of structure-based drug discovery methods. These research lines closely align with LACDR and LIC, but also overlap with other institutes at the faculty of science, which are gathered in the form of the Centre for Computational Life Sciences.
The AI revolution in Drug Discovery
Van Westen: “Drug Discovery has always been a long and expensive process with a high attrition rate. Computational methods, and in particular AI based methods, have the potential to reduce cost, development times, and attrition rates. However, these methods will likely be a synergistic addition to drug discovery rather than a revolutionary replacement of existing methods, similar to previous revolutionary techniques such as ‘Combinatorial Chemistry’ or ‘High throughput screening’.”
At the LACDR there has been a relatively long history of computational drug discovery going back as early as 2001. Key during this history is the close collaboration with the Leiden Institute of Advanced Computer Science (LIACS). The further exploration of new artificial intelligence algorithms opens up new possibilities in the design and synthesis (prediction) of new drug candidates for existing diseases. Both LACDR and LIC are currently actively involved in AI based research. At the same time the coming available of more and more public data, the recent inauguration of the ALICE supercomputer in Leiden, and the SAILS initiative make this the perfect time to explore AI and Medicinal Chemistry.
Diverse applications
Van Westen: “In my group we use artificial intelligence approaches in various areas. Examples include the use of machine learning (regressors) to predict affinity of drug candidates (molecules) for drug targets (proteins) in a polypharmacological context. Input for these models are properties of the molecule (chemical), properties of the protein (binding site), and properties of the interaction (interaction type). Furthermore, we use classifiers in a similar way to predict activity or toxicity of drug candidates, but we also use images as input (based on high-throughput microscopy) for these models.
Finally, we use generative neural networks to generate novel drug-candidates (using a sequential SMILES format) that meet one or more predefined criteria. In all these cases we use experimental validation to validate our models. For this we collaborate heavily with wet lab experimental groups (medicinal chemistry, chemical biology).”
Experience with machine learning
Van Westen obtained his PhD in proteochemometric modeling at the LACDR (Leiden University) on a grant sponsored by Janssen Pharmaceutica under the supervision of prof. A.P. IJzerman and prof. H.W.T. van Vlijmen. He explored the possibilities, caveats, and best practices to combine bioinformatics and cheminformatics for the creation of predictive polypharmacology models using machine learning.
Subsequently he was sponsored by a competitive Marie Curie / EMBL fellowship investigating the exploitation of diverse and complementary data for more accurate affinity models based on machine learning at the European Bioinformatics Institute supervised by prof. J.P. Overington (ChEMBL). Additionally, he branched out in novel fields of research including agrochemical research (collaboration with Syngenta), machine learning assisted Alzheimer’s disease diagnosis, text mining, and molecular dynamics analysis. Finally, his expertise in polypharmacology led to a guest scientist position at the University of Cambridge, working as an external expert for a number of projects.
Van Westen returned to Leiden as a postdoc in the Medicinal Chemistry group. His research plans were funded by NWO with a Veni personal grant, which he further developed into a tenure track at the LACDR. During that time, he also obtained a guest scientist position at Xi’an Jiaotong University in Xi’an (China). He currently collaborates with various commercial and non-commercial organizations and within the EU and is active in the EUToxRisk, eTRANSAFE, and ReSOLUTE IMI consortia.
Van Westen has published over 70 research papers, commentaries, reviews, book chapters and patents. For his independent research, he received several grants and awards including the ‘Discoverer of the year’ award by the faculty of science. He has established the ‘Computational Drug Discovery’ group at LACDR, part of the Division of ‘Drug Discovery and Safety’.