Prof. Dr. Oliver Koch

Heisenberg-Professor of Computational Drug Discovery
Institute of Pharmaceutical and Medicinal Chemistry
Corrensstr. 48, 48149 Münster

Tel.: +49 (0)251 - 83-33443
Fax: +49 (0)251 - 83-32144

E-Mail: Oliver.Koch@uni-muenster.de

https://orcid.org/0000-0001-9228-217X

 

Keywords: medicinal chemistry, computational molecular design, cheminformatics, structure-based design, fragment-based design, artifical intelligence, data-driven decision making, design-synthesize-test cycle 

The importance of computational methods in pharmaceutical drug research was recently highlighted in a special issue on computer-aided drug design (CADD) strategies in pharma (A CADD-alog of strategies in pharma). The identification and development of new drugs is nowadays hardly imaginable without the use of computational methods. These methods support the complete development workflow from initial target and hit identification up to the development of the final drug candidates.

My research interest lies in the development and application of computational methods in rational drug design with focus on structure-based design and ‘big data’ driven decisions in order to develop bioactive molecules and to understand selectivity, promiscuity and polypharmacology of protein-ligand interactions. This work is generally aimed at applying and improving the performance of computational methods in delivering novel and safe small molecule therapeutics. In an interdisciplinary way, the in-silico work is combined with biochemical evaluation and preparative organic synthesis for the identification and optimisation of promising molecules and the evaluation of newly developed methods (see figure). My development research projects are focused on (new) data-oriented methods and the application of artificial intelligence for the improvement of virtual screening approaches, and the analysis of protein-ligand interactions and the underlying framework of protein binding sites. The increase in the number of protein structures and the tremendous amounts of bioactivity data still need new tools and approaches or a better understanding of existing tools for efficient data mining and knowledge discovery. The aim is to use this knowledge, the available ‘big’ bioactivity data and protein structures for computational molecular design and identification of new bioactive compounds. A special focus is on the identification of promiscuous fragments that can generally be used for fragment-based design approaches.

On the following pages you can find more information about the group members and our research and scientific output. 

Vorlesung "Eine Einführung in die künstliche Intelligenz" auch für Studierende ausserhalb von Münster geöffnet.

Wie auch in den letzten Jahren bieten wir im Wintersemester wieder die Vorlesung "Eine Einführung in die künstliche Intelligenz" an. In diesem Wintersemester soll die Vorlesung auch für Studierende ausserhalb der Universität Münster angeboten werden: Die Vorlesung wird sowohl über Zoom als auch als Aufzeichnung zur Verfügung gestellt, um möglichst vielen die Teilnahme zu ermöglichen.

Neben den Vorlesung gibt es praktische Übungen, die mit Hilfe von Jupyter Notebooks bearbeitet werden. Diese Jupyter Notebooks sind eine spezielle Form des E-Learning, das es Ihnen erlaubt, sich mit der Thematik der künstlichen Intelligenz selbständig auseinander zu setzen. Die Notebooks werden dann jeweils nach Ablauf einer Woche in der praktischen Übung im Detail besprochen. An der Vorlesung kann aber auch teilgenommen werden, ohne Jupyter Notebooks zu bearbeiten.

Es gibt jeden Freitag eine Stunde Vorlesung, gefolgt von einer Stunde, in der die praktischen Übungen in Form der Jupyter Notebooks der vergangenen Woche besprochen weden. In der ersten praktischen Übung (11.10.24) wird nur die Benutzung von Jupyter Notebooks erklärt.

Weiter Informationen und die Möglichkeit sich anzumelden finden Sie unter folgendem Link:
https://indico.uni-muenster.de/event/2987/.

undefined

• 

• 

• 

• 

• 

• 

• 

• 

• 

This year's German Conference on Cheminformatics (GCC) was a remarkable event for our research group, as our full group participated in the event. We were thrilled to receive a poster prize for Samuel Hombergs work on explaining graph neural networks, as well as an honorable mention for Philipp Janssens poster on fragment libraries. Additionally, Felipe Victoria-Muñoz had the honor of delivering a talk, presenting his studies on Decoy selection for machine-learning based scoring functions to a wider audience. 

Our group hosted the 2st international summer school AI4MedChem which took place at our nice Pharmcampus. We used theoretical lectures combined with our available Juyper Notebooks to give 35 participants an introduction to artificial intelligence methods with regard to medicinal chemistry. The participants came from Austria, Cameroon, Costa Rica, France, Germany, Italy, Morocco, Netherlands, Poland, United Kingdom. Besides AI, we also had a nice social evening in a pub, a barbecue and a fantastic evening lecture by Nathan Brown about industrial application of AI and ML in Drug Design.
We will provide this summer school next year again. So, if you missed it, wait for the SUMMER SCHOOL AI4 MEDCHEM in September 2025.

The first international summerschool AI4MedChem took place at the PharmaCampus Münster from 19th to the 22nd of September. The 30 participants from France, Great Britain, Israel, Italy, Luxembourg, Spain and Germany learned about methods in artificial intelligence for medicinal chemistry and drug discovery.

05/2023 Joana @ FoChIn Symposium: Poster Award

At the FoChIn Symposium 2023, Joana received a prize for her poster about her phd project with the title: "KCa3.1 channel: Computational analysis of three known toxin inhibitors towards new extracellular inhibitors".

Joana conducts her thesis as a member of the RTG Chemibion. Congratulations!

---

04/2023 A new publication online: Introduction to artificial intelligence and deep learning using interactive electronic programming notebooks

The article "Introduction to artificial intelligence and deep learning using interactive electronic programming notebooks" was just published in Archiv der Pharmazie and can be read here. The accompanying interactive notebooks are available here.

---

03/2023: Joana @ Molecular Modelling Workshop in Erlangen: Poster and Talk Award

Joana, a member of the RTG Chembion (www.chembion.de),  received  two awards at the Molecular Modelling Workshop (https://mmws2023.mgms-ds.de):

3rd Winner of the Talk Award for her talk about her master project with the title "Metadynamics Simulations of FPR2: Using an Enhanced Sampling Method to Elucidate The Mode of Action of a Diverse Set of Ligands"

Winner of the Poster Award for her phd project with the title "KCa3.1 channel: Computational analysis of three known toxin inhibitors towards new extracellular inhibitors"

Congratulations!

---

03/2023 Plenary Lecture @ Molecular Modelling Workshop

Prof. Koch gives an invited talk at the Molecular Modelling Workshop in Erlangen (https://mmws2023.mgms-ds.de) with the title:  "Neural Fingerprints: Structure- and activity-sensitive molecular representations based on neural networks for virtual screening approaches"

---

03/2023 Lecture in the context of the further education of the 'Apothekerkammer Westfalen Lippe'

Prof. Koch gives a talk "An introduction into artificial intelligence" on 22.03.2023 in the context of the further education of the 'Apothekerkammer Westfalen Lippe (AKWL)'. AKWL members are invited to join this lecture. 

---

10/2022: Teaching - An introduction into Artificial Intelligence (WS2022/2023, Teaching Language: German)

This course gives an introduction in into artifitical intelligence for pharmacists (and master chemistry and drug science). It combines a theoretical lecture (1 hour per semester) with practical excercise as jupyter notebooks (1 hour per semester). The development of this course was funded by the "Apothekerstiftung Westfalen Lippe)

The course will take place every friday, Seminarroom 1, Pharmacampus.
The theoretical lecture will take place from 10.00 c.t. to 11.00.
The practical excercise will take place from 11.00 ct. to 12.00. Please bring a laptop.

The link to the Learnweb course:  EEIDKI-2022_2

---

10/2022: Teaching - Drug Design and Development (WS2022/2023, Teaching Language: German)

This course gives an introduction into modern drug design and development methods. 
The link to the Learnweb course: VDDUE-2022_2

---

08/2022: Philipp Otten joins the group as a PhD student

After successfulling finishing his master thesis in our group, Philipp Otten will stay as a PhD student. He will focus on computational molecular design and synthesis of new thioredoxin reductase inhibitors.

---

11/2021:  Invited Talk at 'Discovery Chemistry US' conference

Dr. Koch gives a talk at the 'Discovery Chemistry US' confrence with the title "Neural Fingerprints: Generating Novel Domain-Specific Molecular Fingerprints Using Neural Networks"

---

08/2021 A new publication online:  Natural Product Scores and Fingerprints Extracted from Artificial Neural Networks

Menke, J., Massa, J., Koch, O.* Natural Product Scores and Fingerprints Extracted from Artificial Neural Networks. Comput. Struct. Biotechnol. J., 2021; online available. https://doi.org/10.1016/j.csbj.2021.07.032 

---

03/2021 Janosch @ Frontiers in Medicinal Chemistry: Poster presentation

Janosch Menke is presenting a Poster entitled "Kinase-specific Neural Fingerprints: How to teach a fingerprint
the key structural features of kinase inhibitors to enhance virtual screening" at the Frontiers in Medicinal Chemistry 2021.

---

02/2021 A new publication online: Neural network based fingerprints for virtual screening

Train a fingerprint the important structural features of a target class for enhancing the virtual screening performance. Great work by Janosch now available online. We also provide a kinase specific neural fingerprint for similarity search @ https://github.com/kochgroup/kinase_nnfp. Interesting fact, a graph convolutional network performs worse than an ECFP4 based.
https://doi.org/10.1021/acs.jcim.0c01208

---

11/2020 Member of the International Research Alliance for Antibiotic Discovery and Development (IRAAD-Network)

PD Dr. Oliver Koch has just joined the IRAAD (Member of the International Research Alliance for Antibiotic Discovery and Development) Network. The network aims to promote and accelerate translational science in the early stages of novel antibiotic discovery and lead candidate development. Oliver Koch will contribute based on his experience on computational drug design towards new antibiotic compounds.
https://www.iraadd.eu/en/

---

11/2020 Conference: Janosch @ German Conference on Cheminformatics 2020: Flash Talk

Janosch Menke contributes a Flash Talk entitled "Using Domain-specific Fingerprints Generated Through Neural Networks to Enhance Ligand-based Virtual Screening" at the German Conference on Cheminformatics 2020. www.gdch.de/gcc2020 

---

 

03/2020 - PD Dr. O. Koch receives the Innovation Award in Medical/Pharmaceutical Chemistry

The Innovation Award honours the outstanding, original and forward-looking research work of PD Dr. Oliver Koch in Medical/Pharmaceutical Chemistry. The prize is awarded jointly by the Division of Pharmaceutical/Medical Chemistry of the German Pharmaceutical Society (DPhG) and the Division of Medical Chemistry of the German Chemical Society (GDCh). Link to the GDCh announcement