Ewa Lis
Chief Executive Officer
Koliber Biosciences
Talk Information
Computational Empowerment in Peptide Science
18 June 2025, 03:15pm - 03:30pm, in the Pacific Jewel Ballroom
– AI-Driven Peptide Discovery: Unlocking the Potential of Peptide Arrays for Therapeutic Development
Dr. Ewa Lis is the Founder and Chief Executive Officer of Koliber Biosciences, a San Diego-based biotechnology company specializing in artificial intelligence-driven solutions for peptide drug discovery and microbiome engineering. With a unique blend of expertise in computational science and molecular biology, Dr. Lis leads efforts to accelerate therapeutic development through innovative machine learning platforms.
Academic Background
Dr. Lis earned her B.A. in Chemistry from Cornell University and completed her Ph.D. in Biological Sciences at The Scripps Research Institute. Her academic training provided a strong foundation in both the chemical and biological sciences, enabling her to bridge disciplines in her subsequent professional endeavors.
Professional Experience
Prior to founding Koliber Biosciences in 2014, Dr. Lis held leadership positions at Life Technologies, Genomatica, and Reveal Biosciences. In these roles, she developed innovative technologies ranging from algorithms for pathology tissue classification to genome engineering research tools and microbially derived renewable chemicals. Her diverse experience has been instrumental in shaping Koliber's multidisciplinary approach to biotechnology.
Research Focus
At Koliber Biosciences, Dr. Lis focuses on developing machine learning and artificial intelligence systems to solve complex biological data problems. Her work includes creating proprietary peptide encodings and deep learning architectures to enhance the discovery and optimization of therapeutic peptides. Additionally, she leads projects in microbiome and probiotics engineering, aiming to develop genome engineering technologies for novel microbiome therapeutics.
Notable Contributions
Dr. Lis has pioneered the development of AI-assisted peptide drug discovery platforms capable of analyzing complex datasets derived from peptide variants. These platforms have been validated on several peptide datasets and have received funding from the National Science Foundation. Her work has significantly advanced the application of machine learning in peptide science, enabling more efficient identification and optimization of therapeutic candidates.
Professional Engagements
Beyond her role at Koliber Biosciences, Dr. Lis is an active participant in the scientific community. She has presented her research at various conferences, including the American Peptide Society Symposium, where she discussed peptide hit identification and lead optimization using artificial intelligence approaches. Her commitment to advancing the field extends to mentoring and collaborating with other professionals in biotechnology and computational biology.
Through her interdisciplinary expertise and innovative leadership, Dr. Ewa Lis continues to drive advancements in peptide science and biotechnology, contributing to the development of next-generation therapeutics.
AI-Driven Peptide Discovery: Unlocking the Potential of Peptide Arrays for Therapeutic Development
Koliber Biosciences Inc, USA; Robust Diagnostics, USA
Peptide hit discovery methods such as phage display and mRNA display are widely used tools for identifying bioactive peptides. However, these approaches face significant limitations. Phage display often suffers from poor reproducibility and high false-positive rates, necessitating extensive validation. While mRNA display enables the incorporation of unnatural amino acids and can yield high-potency hits, it comes with substantial licensing costs and is prone to biases from non-uniform clone amplification. Additionally, hits identified through these methods are frequently hyperoptimized, making it challenging to introduce modifications that enhance developability without compromising potency. These constraints hinder the discovery of novel peptide therapeutics.
Peptide arrays, traditionally underutilized due to constraints in library size and accessibility, are emerging as a viable alternative. In this presentation, we will demonstrate how Koliber’s machine learning technology, in collaboration with the peptide array platform developed by Robust Diagnostics, can overcome these challenges. We will show that large libraries are not essential for hit discovery and that initial hits can be rapidly optimized to nanomolar binding affinities using machine learning driven approaches.
Additionally, we will introduce visualization techniques to analyze binding modes, providing insights into peptide–target interactions. This approach paves the way for machine learning driven peptide array technologies to play a more prominent role in the discovery and development of novel peptide therapeutics.