We are thrilled to announce our next Meetup on May 14th at Oerlikon.
Format:
- 2 talks (each ca. 40 min incl. discussion)
- Time for networking + food + drinks before, in between and after the presentations
- Talks are held in English
- We will be taking photos and/or film footage at the event. These will be used to share news about our meetups and to publicize upcoming events.
The lineup:
First talk:
Dr. Pit Vanhoefer - GenAI & Chatbots: Transforming Knowledge Management
Abstract:
In his upcoming talk, Pit will take a closer look at the R(etrieval) and A(ugmented) part of a RAG architecture and discuss the groundbreaking impact of GenAI and chatbots on smart knowledge management. How can organizations create added value with these technologies, what challenges are Data Scientists facing and what new innovations will the future offer?
Bio:
Dr. Pit Vanhoefer is Data Science Team Lead at the Oerlikon Digital Hub and joined the company in 2020. After studying matter-antimatter differences at the high energy physics experiment Belle, Pit worked as a Data Science consultant in various industries. At Oerlikon, he is involved in a number of digital initiatives and is currently working on data solutions for connected equipment, among other topics. When he’s not working with data, Pit enjoys traveling, good food and good company.
Second talk:
Dr. Till Siebenmorgen - MISATO: A dataset and community project for AI-based drug discovery
Abstract:
Large language models (LLMs) have greatly enhanced our ability to understand biology and chemistry. Yet, relatively few robust methods have been reported for structure-based drug discovery. Highly precise biomolecule-ligand interaction datasets are urgently needed in
particular for LLMs, that require extensive training data. We (drug discovery group at Helmholtz Munich) present MISATO, the first dataset that combines quantum mechanics properties of small molecules and associated molecular dynamics simulations of about 20000 experimental protein-ligand complexes. The manuscript is accepted now at Nature Computational Science. Starting from the PDBbind dataset, semi-empirical quantum mechanics was used to systematically refine these structures. The largest collection to date of molecular dynamics traces of protein-ligand complexes in explicit water are included, accumulating to 170 μs. We give AI baseline models and simple Python data loaders, and aim to foster a thriving community around MISATO (https://github.com/t7morgen/misato-dataset). MISATO is actively being developed by the community and there are cooperations with the AI-supported pharmaceutical industry (Khumbu GmbH, Helmholtz AI, Deltawave). The community recently received the DATIpilot funding from the BMWF. Our goal with MISATO is to bring about a paradigm shift in drug research. In the long term, we want to
develop new, cheaper and improved therapeutic approaches through our efforts. The project is an opportunity to bring together "Open Science" with the potential for major social impact.
Bio:
Dr. Till Siebenmorgen is a scientist in the drug discovery group of Helmholtz Munich since 2021. His research focuses on AI driven approaches for structure-based drug discovery, in particular GenAI and GNN applications. Till did his PhD at the Technical University of
Munich in the field of biomolecular simulations. Till’s second biggest passion apart from drug discovery is contributing to music subculture. In 2022 he co-founded the blechsonne music festival and the non-profit association blechsonne e.V..