arXiv subjects

Leonardo Morsut

Publications and source records attributed to Leonardo Morsut.

2 recordsLinked to original sources

arXiv2026

BioEVAL: A global, multi-institutional benchmark of large language and multimodal models for bioengineering

Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science. However, existing LLM benchmarking emphasizes factual recall, offering limited insight into model performance on frontier and multimodal tasks. We assembled BioEVAL (BioEngineering Validation of AI and LLMs), a global, multi-institutional initiative designed to assess experimental reasoning capability across bioengineering (BE) subfields. BioEVAL spans 11 major BE subfields plus a set of uncategorized items, bringing together 22 research groups to create a PhD-level benchmark comprising 608 evaluation items: 1) 380 multiple-choice questions (MCQs, 359 retained after audit), 2) 218 literature synthesis tasks, and 3) 10 multimodal problems with experimental image interpretation. Benchmark items underwent authoring-group expert review and centralized quality control before evaluation. Following evaluation, a blinded cross-group consensus audit of the highest- and lowest-accuracy MCQ items flagged 21 questions for revision or removal; these were withheld, and all reported MCQ results are computed on the 359 retained items. We evaluated diverse cloud-scale foundation/multimodal models (e.g., ChatGPT, Gemini, and Grok) and locally deployable models suitable for inference on consumer-grade GPUs. Models achieved the highest accuracy of up to 90% on MCQs, similarity score of 0.72 on literature synthesis, and accuracy of 80% on a small sample of multimodal reasoning questions, with substantial performance variation across subfields. Leaderboard rankings characterize current capabilities, limitations, and development priorities across the evaluated BE task categories. BioEVAL is maintained as an extensible benchmark with standardized protocols for continuing expert item contribution and model evaluation.

cs.AI↗

arXiv2021

Cell density controls signal propagation waves in a multicellular synthetic gene circuit

During organismal development, biochemical reaction networks sense and respond to mechanical forces to coordinate embryonic patterning with embryo morphogenesis. Factors such as cortical tension, cell density, and matrix mechanical properties influence differentiation and cell fate decisions by modulating gene regulatory signaling networks. A major goal in synthetic development is to construct gene regulatory circuits that program the patterning and morphogenesis of synthetic multicellular structures. However, in the synthetic context, little is known regarding how the physical properties of the growth environment impact the behavior of synthetic gene circuits. Here, we exploit physical-chemical coupling observed in a synthetic patterning circuit in order to control the size and spatial distribution of patterned synthetic cell sheets. We show that cell density attenuates the propagation of signal between neighboring cells in a multicellular sheet containing a contact-dependent patterning circuit based on the synNotch signaling system. Density-dependent attenuation leads to a signal propagation wave that exhibits distinct qualitative phases of persistent propagation, transient propagation, and no propagation. Through computational modeling, we demonstrate that cell growth parameters determine the phase of propagation observed within a growing cell sheet. Using growth-modulating drugs and spatial density gradients, we control the size of synNotch-activated cell populations and generate tissue-scale activation gradients and kinematic waves. Our study reveals that density-dependent synNotch activity can be exploited to control a synthetic multicellular patterning circuit. More broadly, we show that synthetic gene circuits can be critically impacted by their physical context, providing an alternate means for programming circuit behavior.

q-bio.CB↗