Genome Modeling, Protein Synthesis using AI and Safety in Drug Discovery

Hello Everyone,
We would like to invite you to our AI in Pharma MeetUp event that would focus on generative AI in Genome, Protein and Safety in Drug Discovery.
Save the date (23 September 2025) for this AI in Biochemistry MeetUp event in Munich (Address: Luise-Ullrich-Straße 14, 80636 Munich).

Thanks to our co-sponsor: Reply Concept

Agenda Overview:
6:30 PM - Door Open, Drinks & Networking

7:00 PM - Welcome Introduction

7:05 PM - Talk 1: Genome modeling and design across all domains of life with Evo 2
Garyk Brixi
Genetics PhD Student, Machine Learning for Biology
Stanford University / Arc Institute / Evo 2 Team

7:35 PM - Talk 2: From Discovery to Safety: BSL BIOSERVICE Munich - Integrated Approach & Novel Anaphylactic Shock Model
Dr. Sebastian Stiller
Scientific Director,
BSL BIOSERVICE Scientific Laboratories Munich GmbH

Dr. Robert Torka
Head of Business Development,
BSL BIOSERVICE Scientific Laboratories Munich

8:05 PM - Drinks, Snacks & Networking

Talks Details:
Talk 1
Title: Genome modeling and design across all domains of life with Evo 2 (Remote Presentation)

Abstract:
All of life encodes information with DNA. While tools for sequencing, synthesis, and editing of genomic code have transformed biological research, intelligently composing new biological systems would also require a deep understanding of the immense complexity encoded by genomes. We introduce Evo 2, a biological foundation model trained on 9.3 trillion DNA base pairs from a highly curated genomic atlas spanning all domains of life. We train Evo 2 with 7B and 40B parameters to have an unprecedented 1 million token context window with single-nucleotide resolution. Evo 2 learns from DNA sequence alone to accurately predict the functional impacts of genetic variation—from noncoding pathogenic mutations to clinically significant BRCA1 variants—without task-specific finetuning. Applying mechanistic interpretability analyses, we reveal that Evo 2 autonomously learns a breadth of biological features, including exon–intron boundaries, transcription factor binding sites, protein structural elements, and prophage genomic regions. Beyond its predictive capabilities, Evo 2 generates mitochondrial, prokaryotic, and eukaryotic sequences at genome scale with greater naturalness and coherence than previous methods. Guiding Evo 2 via inference-time search enables controllable generation of epigenomic structure, for which we demonstrate the first inference-time scaling results in biology. We make Evo 2 fully open, including model parameters, training code, inference code, and the OpenGenome2 dataset, to accelerate the exploration and design of biological complexity.

Speaker:
Garyk Brixi, Genetics PhD Student
Stanford University / Arc Institute / Evo 2 Team

Biography:
Garyk Brixi is a PhD student in the Department of Genetics at Stanford University, mentored by Professor Brian Hie. Garyk works at the intersection of machine learning and biology, developing novel computational methods to analyze and learn from genomic sequences.

Prior to Stanford, Garyk completed his undergraduate studies at Harvard University with a degree in Applied Mathematics. Previously Garyk worked on human disease genetics, machine learning for protein design, and connecting statistical and deep learning models of protein evolution.

Practical Application:
Designing Protein from generative AI DNA sequences using Evo 2 Foundational Model
https://build.nvidia.com/nvidia/evo2-protein-design

Additional Information:
In addition to Stanford University and Arc Institute, the research was jointly done with researchers from the following institutions:
NVIDIA, OpenAI, Liquid AI, University of California, Berkeley, Goodfire, Columbia University, University of California, San Francisco.

Talk 2
Title: From Discovery to Safety: BSL BIOSERVICE Munich - Integrated Approach & Novel Anaphylactic Shock Model

Abstract:
This presentation introduces a comprehensive approach to overcoming challenges in early drug development, with a focus on integrating scientific consultancy and project management to enhance success rates. After a brief introduction to the Eurofins company, we explore the complexities of preclinical development starting from AI-driven virtual screening approaches, highlighting the importance of in-vitro toxicology assessment with the intention of hazard identification and hazard elimination with the goal of 3R Principles application to animal experiments (Replacement, Reduction and Refinement). The synergistic collaboration between EUROFINS Discovery and BSL BIOSERVICE ensures streamlined workflows and scientific excellence with the value of tailored consultancy and coordinated project execution.
Despite the availability of advanced in vitro and computational methods, certain aspects of drug development and toxicity assessment still require evaluation in animal models. One such example is the study of Type I hypersensitivity reactions, which are IgE-mediated and can result in acute, life-threatening symptoms. These reactions often involve cardiovascular effects that may alter the pharmacokinetics of therapeutic agents, potentially compromising their efficacy in humans. In this presentation, we introduce anaphylaxis as a multifaceted model for toxicity testing, highlighting parameters that cannot be adequately assessed using in vitro systems—such as routes of administration, systemic exposure, and stress-related physiological responses. We will further discuss the relevance and application of anaphylaxis models in both CD34⁺ fully humanized mice and rat models, emphasizing their value in translational safety assessment and drug development.
Finally, we summarize how our in-silico, in-vitro and in-vivo drug safety pipeline contributes to risk minimization in drug safety. The presentation aims to foster collaboration and innovation in preclinical safety strategies, reinforcing the importance of integrated solutions in state of the art drug development.

Speakers:
Dr. Sebastian Stiller
Scientific Director,
BSL BIOSERVICE Scientific Laboratories Munich GmbH

Dr. Robert Torka
Head of Business Development,
BSL BIOSERVICE Scientific Laboratories Munich GmbH

Biography:
Dr. Sebastian Stiller studied Biochemistry at the Eberhard-Karls University Tübingen Germany with main emphasis on immunology. For his PhD thesis Sebastian was working on hematology at department of Molecular Medicine, Max-Planck-Institute of Biochemistry, Germany. He continued his career at applied sciences at Aurigon Life Science working as study director and group leader in Preclinical Pharmacology in vitro and in vivo.
2013 Sebastian moved to BSL Munich working as study director and group leader in Pharmacology/Toxicology with main emphasis on preclinical development of drug candidates including both efficacy and in vivo toxicology. Having meanwhile more than 20 years of experience Sebastian changed his position to scientific director at BSL Munich. He is focusing on tailor design of full IND study packages based on intended human use with various substances such as small molecules, biologics. vaccines and ATMPs. Besides he is designing studies for new pharmacological and efficacy models.

Dr. Robert Torka is experienced in translational drug development and preclinical pharmacology, currently serving as Head of Business Development at BSL BIOSERVICE Scientific Laboratories Munich GmbH. With over two decades of experience across academia, biotech, and CROs. Prior to joining BSL, Robert held senior leadership roles at Reaction Biology and Vyant Bio, where he spearheaded the Director of Scientific Office and European business development. His academic tenure includes lecturer and group leadership at the University of Halle/Wittenberg, where he directed research on synthetically lethal drug combinations and drug resistance mechanisms. He was group leader at Prof. Dr Axel Ullrich´s department at Max Planck Institute of Biochemistry and contributed to small molecule RTK inhibitor development and held roles in breast cancer research at Olympus and Bayer HealthCare. He graduated at the University of Bonn. His scientific contributions span target validation, phenotypic high content screening, tumor microenvironment modeling, and companion diagnostics, with a strong emphasis on translational relevance and innovation.