NSF-Simons National Institute for Theory and Mathematics in Biology

We're excited to kick off the Multi-scale Biological Adaptation to Environmental Variability workshop at NITMB! Throughout this week, mathematicians and biologists will convene at NITMB to discuss the effects of photoperiod, temperature, seasonality, and other types of environmental variability on biological systems across scales. Today we welcome keynote speaker Eve Marder, professor in the Department of Biology at Brandeis University. Explore this week's schedule at https://buff.ly/R4yJosZ

Can understanding the locomotion of living systems like jumping spiders help us build robots that save lives? Explore NITMB Fellow Alasdair Hastewell's research in a new NITMB video spotlight! Alasdair Hastewell’s research interests lie at the intersection of numerical applied mathematics and biology, combining techniques from spectral methods, optimization, and dynamical systems theory with experimental data. At NITMB, he works closely with experimental collaborators to develop data analysis and model inference frameworks that are broadly applicable across experimental datasets, with a particular focus on locomotion across scales from single-celled microorganisms to multi-limbed animals and datasets where only some dynamical variables are observed

🔥 🔥 Super excited to share our latest, led by Leon Schwartz: Most perturbation models answer “if I knock out gene X, what happens?” We asked the reverse: a patient’s cells are in a disease state, so what drove them there? 🔄 Meet ExPert (Extracting Perturbations).It’s a single model trained on 22 Perturb-seq screens, covering 19.5M cells and 12 cellular contexts. It learns one “prototype” per perturbation that holds up across cell types, so any cell state can be traced back to the most likely drivers. What it does: 📈 More than doubles held-out recall over the best existing method, using 1,000× less compute 🧬 Ranks ~20,000 candidate genes per query, more than half of them never seen in training 💊 Works on drugs, protein degradation and combinations without retraining Importantly, We then took it straight to patient samples, with no retraining. 🎯 🍷 Alcohol-associated liver disease: ExPert flagged defective mRNA export, a process nobody had linked to the disease. Differential expression missed it completely because none of those genes change expression. We went to the bench and confirmed it: transcripts trapped in the nucleus, and isoform switching across the export machinery. 🩸 Mitochondrial disease: from a routine blood sample, it pinpointed the genetically affected respiratory complex, traced the signal to monocytes, and caught a misdiagnosed control. This one truly took a village to see through to fruition: Leon Schwartz Jacob Schlamowitz, Ben Kuznets-Speck, Conor Johnson Dipta Das Emanuelle G. Tao Bai Hanxiao Sun Irish Senthilkumar Gillian Primavera Jaekwon Jung, Abdalla Abdrabou, Ph.D. Rogan Grant Carsten Marr Nav Chandel, Divakar Mithal Sam Weinberg Mazhar Adli Auinash Kalsotra and Ekta Prashnani and many others! Thanks for our funding bodies: The Pew Charitable Trusts The Alexander and Margaret Stewart Trust, Burroughs Wellcome Fund, Biohub, American Heart Association Schmidt Sciences, and many others Northwestern University Basic Sciences Northwestern University Department of Cell & Dev Biology Center for Synthetic Biology at Northwestern University NSF-Simons National Institute for Theory and Mathematics in Biology NVIDIA University of Illinois Urbana-Champaign Northwestern University Clinical and Translational Sciences (NUCATS) Institute Ann & Robert H. Lurie Children's Hospital of Chicago The Johns Hopkins University Duke University Link: https://lnkd.in/g57gZqRq

👻 What if the most interesting part of an ecological niche is the space a species doesn’t occupy? Holes, gaps and disconnected regions may carry ecological information. But when we reconstruct niches from occurrence data, our methods can smooth away the very structures we want to study. In our new preprint, Joseph Thomas and I introduce Q-niche: a quantum-mechanics-inspired approach to reconstructing ecological niche hypervolumes while preserving their internal structure. We use an inverse Schrödinger formulation to transform occurrence data into a potential landscape, whose low-potential regions define the estimated niche. We tested Q-niche against Gaussian kernel density estimation and one-class support vector machines using synthetic niches, virtual species and occurrence records from five Australian Acacia species. Key findings: 🔹 Structure matters. Across controlled synthetic benchmarks, Q-niche more consistently retained disconnected regions, loops and enclosed voids. 🔹 Preserving niche structure and predicting geographical ranges are distinct goals. Better preservation of environmental topology did not necessarily translate into better geographical overlap. 🔹 Your estimator shapes your ecological conclusions. The same Acacia occurrence data produced substantially different niche sizes and structures depending on the method used. For me, the exciting opportunity is to move beyond asking “How big is a species’ niche?” towards “How is that niche organised—and what can that organisation tell us?” A pleasure to develop this work with Joe! 📄 Read the preprint in bioRxiv & medRxiv: https://lnkd.in/duQx89Xc Faculdade de Filosofia, Ciências e Letras de Ribeirão preto (FFCLRP-USP) Universidade Federal do Paraná NSF-Simons National Institute for Theory and Mathematics in Biology FAPESP Simons Foundation National Science Foundation (NSF) BBSRC #Ecology #Biodiversity #EcologicalModelling #Topology #QuantitativeEcology

Thank you for joining us at Friday's NITMB Seminar Series 'Double Feature!' As part of the NITMB-SkAI New Faculty and Fellow Open House, Alan Hastings (University of California, Davis) and Joshua Weinstein (University of Chicago) shared insight into their work Alan Hastings, Distinguished Professor Emeritus of Environmental Science and Policy at UC Davis and NITMB Scientific Leader, explored species coexistence in his talk, "An Inverse Approach for Understanding Ecological Coexistence" Joshua Weinstein, Assistant Professor of Molecular Engineering at the University of Chicago, showed how DNA can be engineered to encode large physical networks of molecules in biological tissue in his talk "Programming DNA to Encode Molecular Organization in Tissues" 🎥 Recordings of both talks are available now! Alan Hastings Seminar: https://lnkd.in/gnVag3A8 Joshua Weinstein Seminar: https://lnkd.in/ggNbG4Vx Don't miss this Friday's entry in the NITMB Seminar Series! ANSHUL KUNDAJE, Assistant Professor of Genetics and Computer Science at Stanford University, will present "Decoding Regulatory DNA with Deep Learning Models" https://lnkd.in/gu6aUxJ2

Welcome to the last day of the NITMB Scientific Focus: Theory and Mathematics in Neuroscience – Geometry in Neuroscience Workshop! We're glad you could join us at NITMB this week. Find today's schedule of talks and discussions at https://lnkd.in/g-D78VbU

Welcome to day 2 of the NITMB Scientific Focus: Theory and Mathematics in Neuroscience – Geometry in Neuroscience Workshop! Join us today for exciting talks, posters, and discussions. Today's schedule is available at https://lnkd.in/g-D78VbU

We're excited to welcome all participants to the NITMB Scientific Focus: Theory and Mathematics in Neuroscience – Geometry in Neuroscience Workshop! Throughout the next three days, experimentalists, theorists, and mathematicians will convene at NITMB to identify geometric principles underlying neural computation, highlight open mathematical challenges, and foster collaborations at the interface of neuroscience and geometry Explore today's schedule at https://lnkd.in/g-D78VbU