Following the success of the Unil+EFPL Health Days on the Ecublens campus, the EPFL Valais Wallis campus will host its first Health Day in Sion, featuring health check-ups and tests, information booths, workshops, and an expert talk.
\r\n
\r\n11:00-14:30 – Health Check-ups & Tests, Innovation & Sports\r\n
Weekly BioEngineering Talks ( Lunch Provided)
\r\n
\r\nINVITED SPEAKER:
\r\n
\r\nA full-stack viral toolkit for systems neuroscience
\r\nProf. Lief Fenno, University of Texas at Austin
\r\nHost: Prof. Yoon Seok Kim
\r\n
\r\nAbstract:
\r\n Understanding and manipulating neural circuits requires technologies enabling access to specific cells, manipulation of genome structure, and delivery of increasingly complex genetic programs. I will describe viral technologies developed by my group that address each of these challenges. These include INTRSECT, a Boolean viral targeting strategy for cell-type- and circuit-specific access; VESPA, a novel split-AAV platform that enables efficient reassembly and expression of genetic payloads too large for a single AAV; and CRISPR/Cas9-based strategies for precise, cell-type- and circuit-specific gene knockdown. Together, these tools form an integrated viral toolkit for programmable access to and manipulation of neural circuits across multiple levels of biological complexity.
\r\n
\r\nBio:
\r\nDr. Lief Fenno is a physician-scientist at The University of Texas at Austin and Dell Medical School. His research integrates systems neuroscience, synthetic biology, and bioengineering to develop genetic, viral, and neuromodulation technologies for programmable control of neural circuits in vivo. His laboratory develops molecular and viral platforms for selectively accessing and manipulating defined neural populations, alongside approaches for neural stimulation and emerging therapeutic applications. His work spans fundamental circuit neuroscience, neurotechnology, and the translation of biological tools to treatment of brain disorders. Dr. Fenno is also a board-certified addiction psychiatrist, Chair of the American Psychiatric Association Council on Addiction Psychiatry, and is active in commercial scientific translation and biotechnology innovation.
\r\n
\r\n
\r\n-------------------------------------------------------------------------
\r\n
\r\nBioEngineering Talks mandatory EDBB Seminar Attendance (1st-Year PhD Students)\r\n
Weekly BioEngineering Talks (Lunch Provided)
\r\n
\r\nINVITED SPEAKER:
\r\nFrom single-molecule biophysics with Graphene Energy Transfer to Brownian DNA computing
\r\nProf. Dr. Philip Tinnefeld, LMU Munich
\r\nHost: Prof. Aleksandra Radenovic
\r\n
\r\nAbstract:
\r\nGraphene-on-glass coverslips are an emerging microscopy platform as resonant energy transfer to graphene offers an axial ruler in the 5-40 nm range. I present single molecule biophysics experiments of protein/DNA interactions using Graphene Energy Transfer (GET) in combination with the discovery that dsDNA adopts a vertical orientation on graphene. Axial information of DNA structure and protein positions is deduced from GET-efficiency and xy-information is obtained e.g. by photon-efficient pMINFLUX revealing simultaneous bending and torsion of enzyme induced DNA conformations. Finally, I discuss how multi-color smFRET with dark quenchers probes states of DNA nanorobots and Brownian DNA computers that work close to the thermodynamic optimum.
\r\n
\r\nBio:
\r\nPhilip Tinnefeld is a Professor of Physical Chemistry at Ludwig-Maximilians-Universität München, leading a highly interdisciplinary NanoBioSciences research group that brings together molecular biology, biophysics, nanophotonics, and physical chemistry. He specialises in single-molecule detection and super-resolution microscopy, and he uses DNA origami to build modular nanoscale devices—such as force sensors, fluorescence amplifiers, and molecular robots—that can sense, compute, and respond autonomously. Under his leadership, the lab is also developing advanced biosensing platforms and imaging tools by combining DNA nanotechnology with 2D materials like graphene, pushing the frontiers of how we observe and control biological processes at the molecular level.
\r\n
\r\nSTUDENT SPEAKER:
\r\n
\r\nUnderstanding oxidative stress response in animals - a case study of the essential enzyme adenosylhomocysteinase, AHCY
\r\nYong-Qi Gao, Zenk Lab
\r\n
\r\n
\r\n------------------------------------------------------------------------------
\r\n
\r\nBioEngineering Talks mandatory EDBB Seminar Attendance (1st-Year PhD Students)\r\n
Biological systems have already paved several routes for bioinspired materials, e.g., for structural colours, wetting, adhesion, and mechanical properties. They have largely involved equilibrium or kinetically trapped properties. By contrast, biological systems allowing multifunctional properties are characteristically dissipatively and dynamically adaptive and incorporate homeostatic feedback mechanisms to regulate the dynamic states. They also typically use signal transduction to mediate the signal from the sensory elements to the responsive sites, unlike the classic man-made stimuli responsive materials. These concept suggest avenues towards more complex life-like functions in soft matter. Herein, optically driven hydrogel systems are presented to drive homeostatically controlled thermal oscillations and to mediate signal transduction via thermal intermediate signalling (1). On the other hand, biological magnetoreception in elasmobranch organisms involves electromagnetic induction (EMI) and suggests avenues for bioinspired electrical signal transductions. We show magnetic profile recognition of translocated magnetically encoded soft matter objects using electrical intermediate signaling by EMI and decoding the magnetic information with machine learning (2). Combination of high stiffness and promoted dynamics for self-healing has turned challenging to combine. Self-healing hydrogels, still combining high tensile stiffness close to that of skin were shown by using polymers upon nanoconfinement in 2D-nanosheet layered confinements (3). Finally, we show trainable actuations in bilayer gel actuators (4). Ever more complex responses can be created, such as mimicking the Pavlovian classical conditioning (5), all paving ways for growingly life-like functions for soft robotics and interfacing with biology.
\r\n
\r\n1. H. Zhang, H. Zeng, A. Eklund, H. Guo, A. Priimagi, O. Ikkala, Feedback-controlled hydrogels with homeostatic oscillations and dissipative signal transduction, Nat. Nano, 17, 1303 (2022).
\r\n2. Z. Miao, X. Hu, K. Liu, S. Hu, G. Yan, H. Tan, O. Ikkala, Z. P. Lv, B. Peng, Bioinspired sensory transduction for magnetic profile recognition and encryption, Adv. Mat. 38 e15994 (2026).
\r\n3. C. Liang, V. Dudko, O. Khoruzhenko, X. Hong, Z.-P. Lv, J. Breu, O. Ikkala, H. Zhang, Stiff and self-healing hydrogels by polymer entanglements in co-planar nanoconfinement, Nat. Mater. 24, 599 (2025).
\r\n4. S. Hu, Y. Fang, C. Liang, M. Turunen, O. Ikkala, H. Zhang, Thermally trainable dual network hydrogels, Nat. Commun., 14, 3717 (2023).
\r\n5. H. Zhang, H. Zeng, A. Priimagi, O. Ikkala, Programmable responsive hydrogels with classical conditioning algorithm, Nat. Commun., 10, 3267 (2019).
Since 2006, the Erna Hamburger Prize is awarded by the EPFL-WISH Foundation (Women In Science and Humanities) and distinguishes influential, leading women scientists who are transforming their field and executing change.
\r\n
\r\nThe Erna Hamburger Prize 2026 will be awarded to Prof. Margo Seltzer. The award ceremony will be held on November 11th 2026 at 6:30 pm at SG 1138. On this occasion, the awardee will give a conference on the EPFL Campus. Don’t miss this exceptional opportunity to hear from one of the most influential figures in contemporary computer science.
\r\n
\r\nConference abstract
\r\nIt's 2026, and the answer to every problem today seems to be a multiple choice exercise of selecting answers from: AI, Agents, LLMs or machine learning. Is our future really just all about prompt engineering?
\r\n
\r\nAt a time when artificial intelligence, autonomous agents and large language models seem to be presented as the answer to every challenge, Margo Seltzer offers a thought-provoking perspective on the value of simplicity in computer science and machine learning. She will demonstrate how simpler and more transparent approaches can deliver performance comparable to that of the most sophisticated models, while providing greater interpretability, more efficient use of resources and a more natural way of addressing issues of fairness and privacy. The lecture will also explore a theme that lies at the heart of her career: collaboration. Drawing on her own experience of collaborative research, Margo Seltzer will explain how working together can broaden one’s impact, help overcome imposter syndrome, and enable individuals to discover and develop their unique strengths.
\r\n
\r\nAbout Prof. Margo Seltzer
\r\nProf. Margo Seltzer is Co-Head of the Department of Computer Science at the University of British Columbia (Vancouver), where she moved after two decades as a Full Professor at Harvard University. She earned her Ph.D. in Computer Science in 1992 from the University of California, Berkeley. A pioneering computer systems researcher, she is the recipient of numerous prestigious honors, including the 2023 ACM Athena Lecturer Award and the USENIX Lifetime Achievement Award. She is a Member of the National Academy of Engineering. Alongside her academic career, Margo Seltzer is an entrepreneur: she co-founded Sleepycat Software, maker of Berkeley DB, a widely used embedded database engine later acquired by Oracle. A recognized leader in her field, she is also a devoted mentor to young women in computer science and founded the \"Systers Lunch,\" the original diversity gathering at SOSP, the flagship conference of the systems research community.
What if robots didn’t replace humans, but collaborated with us?
\r\n
\r\nOn the occasion of the exhibition Do Robots Dream of Construction Sites?, on view at Archizoom from 9 September to 5 December 2026, we invite you to explore a selection of scientific and popular science books, as well as science fiction novels dedicated to human-robot interactions, a theme at the heart of the exhibition.
\r\n
\r\nThis selection offers different perspectives on robotics, automation, work, and the future forms of collaboration between humans and machines.
\r\n
\r\nAll documents can be borrowed according to the Library’s standard borrowing conditions.
\r\n
\r\nDiscover the selection at the Espace Découverte from 5 October to 7 December 2026, as well as online.
\r\n
\r\nBrowse the Library's online reading selection: go.epfl.ch/library-robots
\r\n
\r\nLearn more about the Archizoom exhibition: memento.epfl.ch/event/do-robots-dream-of-construction-sites-archizoom
\r\n
Thesis Director: Prof. M. Jaggi,
\r\nComputer and Communication Sciences doctoral program
\r\nThesis Nr. 10949
\r\n
\r\nTo take part in the public defense, please contact directly the speaker
Thesis Director: Prof. S. Kashyap,
\r\nComputer and Communication Sciences doctoral program
\r\nThesis Nr. 11376
\r\n
\r\nTo take part in the public defense, please contact directly the speaker