EPFL BioE Talks SERIES "Programmable Protein Therapeutics via Generative Language Models"
Event details
| Date | 20.05.2025 |
| Hour | 14:15 › 15:15 |
| Speaker | Prof. Pranam Chatterjee, Duke University, Durham, NC (USA) |
| Location | Online |
| Category | Conferences - Seminars |
| Event Language | English |
WEEKLY EPFL BIOE TALKS SERIES
--> PLEASE MIND THE UNUSUAL WEEKDAY and TIME FOR THIS SEMINAR, namely TUESDAY, 14:15 (not Monday, 12:15)!
A flight cancellation made us reschedule the talk as the speaker could not make it to Lausanne before Tuesday.
We apologize for any inconvenience and thank you for your understanding.
Abstract:
We develop sequence-based algorithms to design functional proteins, from CRISPR enzymes for precise genome editing to transcription factors that direct differentiation of ovarian cells for fertility applications. The primary focus of this talk will be on our new language models (PepPrCLIP and PepMLM) that design de novo peptides capable of binding and modulating undruggable, disordered disease targets, such as oncogenic fusion proteins and neurodegenerative protein aggregates. To enhance design applications, we will further introduce new models that extend the representation capacity of language models to post-translational modifications (PTM-Mamba) and oncogenic fusion proteins (FusOn-pLM). Additionally, we will highlight new frameworks for discrete generation of highly-specific, therapeutic peptides that are simultaneously optimized for both target binding and drug-like properties (moPPIt and PepTune). Overall, we believe that language models, trained with biologically meaningful priors and augmented with application-aware sampling, can unlock the next generation of therapeutics for currently intractable diseases.
Bio:
Pranam Chatterjee is an Assistant Professor of Biomedical Engineering and Computer Science at Duke University. His Programmable Biology Group develops generative algorithms for de novo protein and peptide design, integrating the newest advances in language modeling with translational in vitro and in vivo experimental platforms. Specifically, the lab uses these tools for diverse protein modulation, genome editing, and cell engineering applications. Having received his SB, SM, and PhD from MIT, Professor Chatterjee has received the Hartwell Individual Biomedical Research Award and multiple NIH and foundation grants for his work. He has also co-founded Gameto, Inc. and UbiquiTx, Inc. which translates his research into fertility-related solutions and cancer therapeutics, respectively.
Zoom link (with one-time registration for the whole series) for attending remotely: https://go.epfl.ch/EPFLBioETalks
Instructions for 1st-year Ph.D. students planning to attend this talk, who are under EDBB’s mandatory seminar attendance rule:
IN CASE you cannot attend in-person in the room, please make sure to
--> PLEASE MIND THE UNUSUAL WEEKDAY and TIME FOR THIS SEMINAR, namely TUESDAY, 14:15 (not Monday, 12:15)!
A flight cancellation made us reschedule the talk as the speaker could not make it to Lausanne before Tuesday.
We apologize for any inconvenience and thank you for your understanding.
Abstract:
We develop sequence-based algorithms to design functional proteins, from CRISPR enzymes for precise genome editing to transcription factors that direct differentiation of ovarian cells for fertility applications. The primary focus of this talk will be on our new language models (PepPrCLIP and PepMLM) that design de novo peptides capable of binding and modulating undruggable, disordered disease targets, such as oncogenic fusion proteins and neurodegenerative protein aggregates. To enhance design applications, we will further introduce new models that extend the representation capacity of language models to post-translational modifications (PTM-Mamba) and oncogenic fusion proteins (FusOn-pLM). Additionally, we will highlight new frameworks for discrete generation of highly-specific, therapeutic peptides that are simultaneously optimized for both target binding and drug-like properties (moPPIt and PepTune). Overall, we believe that language models, trained with biologically meaningful priors and augmented with application-aware sampling, can unlock the next generation of therapeutics for currently intractable diseases.
Bio:
Pranam Chatterjee is an Assistant Professor of Biomedical Engineering and Computer Science at Duke University. His Programmable Biology Group develops generative algorithms for de novo protein and peptide design, integrating the newest advances in language modeling with translational in vitro and in vivo experimental platforms. Specifically, the lab uses these tools for diverse protein modulation, genome editing, and cell engineering applications. Having received his SB, SM, and PhD from MIT, Professor Chatterjee has received the Hartwell Individual Biomedical Research Award and multiple NIH and foundation grants for his work. He has also co-founded Gameto, Inc. and UbiquiTx, Inc. which translates his research into fertility-related solutions and cancer therapeutics, respectively.
Zoom link (with one-time registration for the whole series) for attending remotely: https://go.epfl.ch/EPFLBioETalks
Instructions for 1st-year Ph.D. students planning to attend this talk, who are under EDBB’s mandatory seminar attendance rule:
IN CASE you cannot attend in-person in the room, please make sure to
- send D. Reinhard a note well ahead of time (ideally before seminar day), informing that you plan to attend the talk online, and, during seminar:
- be signed in on Zoom with a recognizable user name (not any alias making it difficult or impossible to identify you).
Practical information
- Informed public
- Registration required
Organizer
- Prof. Maria Brbić, Institute of Computer & Communication Sciences and Institute of Bioengineering
Contact
- Institute of Bioengineering (IBI), Dietrich REINHARD