This symposium treated the impact of Artificial intelligence (AI) and Machine Learning (ML) on chemistry and biology, including drug design. The underlying theme of the discussions was what we can now do with AI and ML that we couldn’t do before. For instance, is AI simply a new interpolation method in the old traditional computer-aided drug design (CADD)? Are the methods applied nowadays unreliable in contrast to the those applied in the early days? How can we combine the development and application of new materials with AI-generated knowledgebases? How much can AI help in the prediction and retro-synthesis of new chemical compounds, and how useful are these compounds? Still, the quality of published data matters as meaningful AI applications depend on the availability of high-quality test datasets.
AI in Chemistry and Biology:
Evolution or Revolution?
Beilstein Bozen Symposium
2024
June 4–6, 2024
Hotel Jagdschloss Niederwald
Rüdesheim, Germany
Scientific Committee:
Tim Clark / Friedrich-Alexander-University Erlangen-Nürnberg
Antonella Di Pizio / Leibniz Institute for Food Systems Biology and Technical University Munich
Oliver Koch / University of Münster
Carsten Kettner / Beilstein-Institut
Overview
These and more questions have been addressed:
- chemical sensing
- chemical aided drug design
- protein structure prediction
- AI models from physical models
- synthesis planning
- human brain project
- medical image recognition, and other interdisciplinary topics
In the tradition of the Bozen Symposia participants from widely differing areas of specialization presented provoking talks for a general but qualified scientific audience and researchers from the different disciplines came into contact and led to fruitful “out of the box” discussions.
Scientific Program
Tuesday, June 4
9:00
Welcome and Opening
Carsten Kettner
Session Chair: Oliver Koch
9:20
When Did Cheminformatics Become ML? [Slides]
Tim Clark, Friedrich-Alexander University Erlangen-Nürnberg
10:00
Evolution of the Chemical Space and the Role of AI in its Further Expansion [Slides]
Guillermo Restrepo, MPI for Mathematics in the Sciences, Leipzig
10:40
Lightning Talks
11:10
Coffee break and poster session
11:40
AI (R)evolution in Small Molecule Drug Design
Christian Kramer, Hoffmann-LaRoche, Basel
12:20
Also in Chemistry, Deep Learning Models Love Really Big Data [Slides]
Christoph Steinbeck, University of Jena
13:00
Lunch
Session Chair: Christoph Steinbeck
14:15
Software Lightning Talk
14:25
AI in Medicinal Chemistry: The Revolution yet to Unfold
Christian Tyrchan, AstraZeneca, Gothenburg
15:05
Extrapolation with Chemical Machine Learning
Johannes T. Margraf, University of Bayreuth
15:45
Coffee break and poster session
16:15
Should We Expect a Revolution from "Artificial" Intelligence?
Benjamin Risse, University of Münster
16:55
Machine Learning for Molecular Sensing [Slides]
Giovanni Cuniberti, Technical University Dresden
17:35
Opportunities of Shared AI Infrastructure in Chemistry Research
Beth A. Plale, Indiana University, Bloomington
18:15
Close
19:30
Dinner
Wednesday, June 5
9:00
Opening
Session chair: Tim Clark
9:05
AI in Drug Design - Let's Talk about Data
Oliver Koch, University of Münster
9:45
Drug Discovery and Repurposing in the LLM Era
Tudor Oprea, Expert Systems Inc., San Diego
10:25
Coffee break and poster session
11:00
Artificial Intelligence in Drug Discovery - Towards Virtual Drug Discovery
Gerard J.P. van Westen, Universiteit Leiden
11:40
Artificial Intelligence in Drug Discovery - From Model to Process, From Academic Publication to Decision Making [Slides]
Andreas Bender, Cambridge University/Pangea Bio
12:20
Hybrid AI and Open Source for Kinase-focused Drug Design
Andrea Volkamer, Saarland University, Saarbrücken
13:00
Lunch
14:00 - 17:30
Excursion
19:30
Dinner
Thursday, June 6
9:00
Opening
Session chair: Antonella Di Pizio
9:05
Ensuring Neural Networks Robustness: Problems and Opportunities [Slides]
Ekaterina Komendantskaya, University of Southampton
9:45
Conditional Protein Design with Unsupervised Language Models
Noelia Ferruz, Barcelona Institute of Molecular Biology (online)
10:25
Coffee break and poster session
10:55
Protein Design 2.0
Birte Höcker, University of Bayreuth
11:35
New AI Toolkits for Structural Biology: Opportunities and Limitations [Slides]
Stefan Arold, KAUST, Thuwal
12:15
Application of Artificial Intelligence (AI) and Computer Simulations in GPCR Research
Jana Selent, Hospital del Mar Medical Research Institute (IMIM) & Pompeu Fabra University, Barcelona
13:00
Lunch
Session chair: Beth A. Plale
14:15
Cheminformatics and Modelling of Flavour Molecular Systems: Applications and Limitations of AI
Antonella Di Pizio, Technical University Munich
14.55
Siimulation-based Inferecne for Biophysics
Pilar Cossio, Simons Foundation, New York (Online)
15:35
Coffee break
16:05
Assisting Sampling of Physical Systmes with Generative Models
Marylou Gabrié, École Polytechnique Paris
16:45
Summary - Quo Vadis AI in Chemistry and Biology?
Karl-Heinz Baringhaus, Sanofi-Aventis GmbH, Frankfurt
17:25
Closing Remarks
Carsten Kettner
19:30
Dinner
Posters
No. 1
Providing Machine Learning Models for Property Predicition for Research Partners
Conrad Stork, BASF, Ludwigshafen
No. 2
Thyroid Ultrasound: Diagnostic Criteria and Artificial Intelligence Techniques
Karima Bahmane, ENSA, Agadir
No. 3
Recurrent Neural Chemical Reaction Networks that Approximate Arbitrary Dynamics
Alexander Dack, Imperial College London
No. 4
You Can't Improve what You Don't Measure - Measuring ML/AI Impact in Drug Discovery Projects
Christian Herhaus, Merck KGaA, Darmstadt
No. 5
Machine-learning Assisted ReaxFF Simulations for Accurate Computational pKa Estimations of Organic Molecules
Filip Sagan, Jagiellonian University, Krakow
No. 6
3D-QSAR Meets ML for Binding Affinity Prediction
Katarina Stanciaková, OpenEye, Cologne
No. 7
Efficient Torsion Scans of Small Molecules with TRIP SE(3) Transformer and the ZONTAL Platform
Dennis Della Corte, ZONTAL Inc., Provo