Publications

2026 The Massive Multimodal Biomedical Understanding Challenge A challenge benchmarking the visual perception of multimodal LLMs on open-ended biomedical VQA, with frontier, domain-adaptation, and efficient-model tracks.
2026 Lessons learned from a Kaggle challenge for particle picking in cryo-electron tomography Analysis of the CZII CryoET Object Identification Kaggle challenge: >1,000 participants, particle pickers that beat the prior state of the art, and a released benchmark of tomograms, ground truth, and winning annotations.
2026 A multimodal perturbation atlas defines the phenotypic resolution of cellular morphology
2026 copick: An open dataset interface and toolkit for collaborative annotation and analysis of cryo-electron tomography data An open dataset API and toolkit for collaborative cryoET annotation, with napari and ChimeraX plugins and a Model Context Protocol interface for AI-agent-driven pipeline generation.
2026 A generative AI framework for disease-specific lung microtissue bioengineering GLAM couples 3D tissue imaging with generative diffusion models and two-photon 3D bioprinting to design and fabricate disease-specific lung microtissues from healthy, fibrotic, and emphysematous mouse models.
2025 A realistic phantom dataset for benchmarking cryo-ET data annotation A Nature Methods publication establishing a benchmark phantom dataset for machine learning in cryoET annotation.
2025 Advancing Particle Identification in Cryo-Electron Tomograms with Deep Learning
2025 Polarity-JaM: An image analysis toolbox for cell polarity, junction and morphology quantification Polarity-JaM is an open-source Python package for image analysis of cell polarity, junction, and morphology, featuring deep learning-based segmentation and advanced statistical tools for comprehensive analysis.
2024 Open-source Tools for CryoET Particle Picking Machine Learning Competitions An ML competition and open-source tools for cryoET particle picking, enabling better data handling, labeling, and visualization for advancing structural biology.
2024 Surforama: interactive exploration of volumetric data by leveraging 3D surfaces Surforama is a Python package and napari plugin for interactive visualization, annotation, and analysis of proteins in cryo-ET data, with results exportable as RELION-formatted STAR files.
2024 A Connectome of the Male Drosophila Ventral Nerve Cord This study presents the first densely-reconstructed connectome for the ventral nerve cord (VNC) of a male Drosophila, providing detailed insights into the synaptic connectivity of neurons and premotor circuits that guide behavior.
2023 napari-imagej: ImageJ Ecosystem Access from napari This paper presents napari-imagej, a plugin that integrates the ImageJ ecosystem with napari, enabling access to ImageJ functions directly from napari without the need for extensive porting or maintenance.
2023 Single‐cell profiling and zebrafish avatars reveal LGALS1 as immunomodulating target in glioblastoma This study reveals the role of LGALS1 in glioblastoma through single-cell RNA sequencing and zebrafish avatars, highlighting the heterogeneity in GBM-induced GAM polarization and identifying LGALS1 as a key regulator of immunosuppression.
2023 State of the Art for Machine Learning in Bioimage Analysis
2023 Machine learning-based ozone and PM2.5 forecasting: Application to multiple AQS sites in the Pacific Northwest This study develops and evaluates a machine learning-based forecasting system for ozone and PM2.5 at multiple AQS sites in the Pacific Northwest, demonstrating improved accuracy and reliability over existing CTM-based forecasts.
2023 Image Informatics
2022 RS-FISH: precise, interactive, fast, and scalable FISH spot detection RS-FISH is a software tool for accurate and fast spot detection in microscopy images, supporting large datasets and offering interactive parameter tuning.
2022 PyZebrascope: An Open-Source Platform for Brain-Wide Neural Activity Imaging in Zebrafish PyZebrascope is an open-source Python platform for whole-brain neural activity imaging in zebrafish using light-sheet microscopy, designed to enhance dissemination and development of light-sheet microscopy techniques in neuroscience.
2022 Can fractal dimensions objectivize gastropod shell morphometrics? A case study from Lake Lugu (SW China) This study uses high-resolution 3D scanning technology and fractal dimension analyses to overcome the shortcomings of traditional morphometrics in describing gastropod shell shape, revealing more detailed evolutionary and ecological information.
2022 Development of a Machine Learning Approach for Local-Scale Ozone Forecasting: Application to Kennewick, WA This study developed machine learning models to improve O3 forecasts for Kennewick, WA, demonstrating improved accuracy and reduced computational resources compared to traditional chemical transport models.
2021 Album: executable building blocks for scientific imaging routines, from sharing to LLM-assisted orchestration
2021 Complex population dynamics in a spatial microbial ecosystem with Physarum polycephalum This study explores the interactions between Physarum polycephalum and a red yeast in a spatial ecosystem, revealing an inverse growth relationship and successional dynamics, using advanced image analysis for semantic segmentation.
2021 SNT: a unifying toolbox for quantification of neuronal anatomy SNT is a comprehensive framework for neuronal morphometry and connectomics, providing tools for tracing, proof-editing, visualization, quantification, and modeling of neuroanatomy, available via Fiji's ImageJ distribution.
2021 Review of *The Digital Cell: Cell Biology as a Data Science*. By Stephen J. Royle
2021 Active perception during angiogenesis: filopodia speed up Notch selection of tip cells in silico and in vivo This study demonstrates that filopodia speed up collective endothelial decisions during angiogenesis by conferring a bistable switch-like property to Notch lateral inhibition, ensuring rapid and robust tip cell selection.
2020 Bionic Tracking: Using Eye Tracking to Track Biological Cells in Virtual Reality We introduce Bionic Tracking, a method for solving cell tracking problems using eye tracking in virtual reality. By following cells with their eyes in a VR headset, users can efficiently track cells in 3D space, leveraging smooth pursuit eye movements and a graph-based algorithm to connect time points.
2020 Tales from the Trenches: Developing sciview, a new 3D viewer for the ImageJ community This paper describes the development of sciview, a modern 3D viewer for the ImageJ community, highlighting sustainable development, community building, and software engineering best practices.
2020 Mouse retinal cell behaviour in space and time using light sheet fluorescence microscopy This study uses light-sheet fluorescence microscopy (LSFM) to demonstrate improved imaging techniques for mouse models of eye diseases, revealing new pathological features and offering advancements in understanding neurovascular and degenerative processes.
2019 scenery: Flexible Virtual Reality Visualization on the Java VM We introduce scenery, a flexible VR/AR visualization framework for the Java VM that can handle mesh and large volumetric data. scenery supports multiple views, timepoints, and color channels, and is designed to facilitate quick prototyping and full application development. It is free, open-source, cross-platform, and uses Vulkan or OpenGL rendering APIs.
2019 A large pool of actively cycling progenitors orchestrates self-renewal and injury repair of an ectodermal appendage Actively cycling epithelial progenitors in the mouse incisor generate enamel-producing ameloblasts and non-ameloblast cells, coordinating tissue renewal and repair through progenitor-cell proliferation and Notch1-mediated cell conversion.
2019 Escalation of Memory Length in Finite Populations This study explores the escalation of strategic complexity in coevolutionary systems, using an expanded model of memory length evolution. It demonstrates that larger population sizes support greater complexity escalation and that such escalation is sensitive to perturbations during transitions.
2019 Multi-sample SPIM image acquisition, processing and analysis of vascular growth in zebrafish We present a workflow for multi-sample SPIM imaging, processing, and analysis to study vascular growth in zebrafish, allowing simultaneous imaging and automated data processing, yielding precise quantification of vascular development.
2018 Neuromodulated Learning in Deep Neural Networks This paper proposes a method of deep artificial neuromodulation for stochastic gradient descent, applying biological neuromodulation concepts to modify learning parameters at each layer in a deep neural network during training. The method shows adaptability to different models and new problems, demonstrating dynamic, location-specific learning strategies.
2018 Artificial Gene Regulatory Networks—A Review A review of computational models of gene regulatory networks, covering modelling techniques and their application to control dynamics, evolutionary algorithms, morphogenesis, and neuromodulation.
2018 Acquisition and reconstruction of 4D surfaces of axolotl embryos with the flipping stage robotic microscope
2018 Evolving Differentiable Gene Regulatory Networks A GPU-based implementation of differentiable gene regulatory networks, enabling local optimization of GRN architectures by stochastic gradient descent and evaluating how evolution and SGD can be combined.
2018 FunImageJ: a Lisp framework for scientific image processing FunImageJ is a Lisp framework for scientific image processing built upon the ImageJ software ecosystem, offering a functional programming style and performance suitable for big data biological image analysis.
2017 Dynamic Habitat Construction Through Evolving 3D Swarms
2017 Competitive Dynamics in Eco-evolutionary Genetically-regulated Swarms This paper introduces a model of evolving genetically-regulated swarms within an eco-evolutionary system, highlighting the benefits of competitive selection pressure for collision avoidance, which leads to improved navigation abilities and overall agent behavior.
2016 Asymmetric division coordinates collective cell migration in angiogenesis Asymmetric division in endothelial tip cells generates daughters of distinct sizes and Vegfr signalling levels, driving coordinated collective cell migration in angiogenesis. This Dll4/Notch-independent mechanism integrates cell proliferation with migration, potentially aiding tissue growth during development, regeneration, and cancer invasion.
2016 Review of Wolfgang Banzhaf and Lidia Yamamoto: *Artificial Chemistries*, MIT Press, 2015
2016 A circuit basis for morphogenesis This paper explores the consequences of a circuit basis for reaction–diffusion systems on morphogenesis, including the reachability of patterns and the logical basis of pattern stability.
2016 Configurable NOR gate arrays from Belousov-Zhabotinsky micro-droplets We investigate the BZ reaction to establish a basis for computation using chemical oscillators coupled via inhibition, producing reliable NOR gates in a microfluidic experiment.
2016 Temporal modulation of collective cell behavior controls vascular network topology This study identifies temporal regulation of the VEGF/Notch feedback loop as a key mechanism in determining vascular network density, using computational modeling and in vivo live imaging to show how the rate of tip-cell selection influences network topology.
2015 Genetically-regulated Neuromodulation Facilitates Multi-Task Reinforcement Learning We use gene regulatory networks (GRNs) to modulate the parameters of the SARSA reinforcement learning algorithm, demonstrating that GRNs can be optimized to enhance learning and generalize across multiple tasks without specific problem knowledge.
2015 Gene Regulatory Network Evolution Through Augmenting Topologies This paper presents a new algorithm for the evolution of artificial gene regulatory networks (GRNs), using a network distance metric for speciation and genetic similarity. The algorithm is inspired by the neuroevolution of augmenting topologies (NEAT) algorithm and demonstrates superior performance compared to standard GA and evolutionary programming across various experiments.
2015 Functional Image Processing with ImageJ/FIJI
2014 Can active perception generate bistability? Heterogeneous collective dynamics and vascular patterning This study explores the role of active perception in generating bistability during angiogenesis. By examining endothelial cells (ECs) in a simulated environment, the study demonstrates that sensorimotor feedback can lead to bistability and influence collective decision timing.
2014 Feedback Control of Evolving Swarms This paper explores the impact of proportional-integral control on the fitness and genetics of evolving swarms, revealing that while energy accumulation increases, so does the rate of expiration. The study highlights the importance of information quality over control force in swarm dynamics.
2014 The Resilience of a Swarm Ecosystem Under Environmental Variation This study explores the impact of climate-change-related environmental variation on evolving swarms of birds, highlighting how desertification and tropicalification affect swarm dynamics, population stability, and genetic traits.
2014 Coevolution in Hide and Seek: Camouflage and Vision This paper models the coevolution of prey appearance and predator vision, demonstrating that predator-prey coevolution can maintain engagement and that background complexity supports this interaction.
2014 Modelling nonlinear chemical oscillators with inhibitory coupling
2014 Eco-evolutionary dynamics of complex social strategies in microbial communities This mini-review discusses the eco-evolutionary dynamics of complex social strategies in microbial communities, focusing on cooperation, competition, and cheating behaviors.
2014 Escalation of Coevolutionary Arms Races
2013 Robot Coverage Control by Evolved Neuromodulation This study uses neuromodulatory gene regulatory networks to improve reinforcement learning agents in solving the robot coverage control problem, demonstrating better generalization and learning capabilities compared to fixed parameter settings.
2012 Computational Neuroecology of Communicated Somatic Markers This paper investigates the somatic marker hypothesis in predator-prey simulations, exploring how communicated somatic markers like fear, happiness, and surprise can increase survival rates and be favored by natural selection.
2012 Tag-Based Modularity in Tree-Based Genetic Programming This paper presents a technique for incorporating tag-based modularity into tree-based genetic programming systems, demonstrating its benefits on the lawnmower problem but noting challenges on the Boolean even-4-parity problem.
2012 Autoconstructive Evolution for Structural Problems This paper contrasts autoconstructive evolution, where evolving programs determine their own variations, with the PushGP genetic programming system. A problem size scaling analysis is performed on structural problems to study the dynamics of this hybrid approach.
2012 The Effects of Finite Populations and Selection on the Emergence of Signaling This study examines the emergence of signaling using the Sir Philip Sidney Game, comparing infinite and finite population models. It finds that honest signaling equilibria are difficult to reach from non-communicative origins and highlights the limitations of finite models in capturing the dynamics of deceptive signaling.
2012 The Effects of Tags on the Evolution of Honest Signaling This study examines the role of social tags in the emergence and maintenance of signaling, using the Sir Philip Sidney Game. The findings suggest that while inclusive fitness theory alone may not explain the evolution of cooperation, tag-based relatedness can achieve cooperativity in some cases.
2011 Communicated Somatic Markers Benefit Both the Individual and the Species This paper explores the use of emotional communication in a predator-prey game to evaluate the tradeoff between socio-emotional behavior at individual and species scales, suggesting that such dynamics may support the emergence of altruistic behavior.
2011 Tag-Based Modules in Genetic Programming We present a technique for evolving modular programs with genetic programming using tags to label and refer to code fragments. The tag-based module technique allows for incremental co-evolution of labels and code, scaling well with problem size and performing effectively in complex environments.
2011 Fecundity and selectivity in evolutionary computation
2011 What’s in an Evolved Name? The Evolution of Modularity via Tag-Based Reference This chapter describes a new approach to names in genetic programming based on Holland’s concept of tags, demonstrating their use in extending the power of genetic programming systems and solving a robot navigation problem.
2011 Imaging analysis of clock neurons reveals light buffers the wake-promoting effect of dopamine Environmental light suppresses the wake-promoting effects of dopamine in fly brains. A subset of clock neurons, the 10 large lateral-ventral neurons (l-LNvs), are wake-promoting and respond to dopamine, octopamine as well as light. Behavioral and imaging analyses suggested that dopamine is a stronger arousal signal than octopamine. Light exposure not only suppressed the l-LNv responses but also synchronized responses of neighboring l-LNvs. The regulatory mechanisms described here may contribute to the control of sleep stability while still allowing behavioral flexibility.
2011 Computational Emotions Encourage Collective Behavior in Population Dynamics
2011 Compositional Autoconstructive Dynamics
2010 En Route to Signal Inversion in Chemical Computing This paper explores the use of the Belousov-Zhabotinsky (BZ) reaction as a substrate for computation, demonstrating a technique to invert input signals and compute logical operators like NOT and NOR, with experimental progress in implementing these operators using soft lithography.
2010 Robot phylogenetics
2010 Conspecific Emotional Cooperation Biases Population Dynamics: A Cellular Automata Approach
2008 Emotions for Strategic Real-Time Systems This paper proposes a paradigm for enhancing strategic real-time systems by introducing emotions into simulated agents, enabling better decision-making and cooperation that mimics human behavior through internal and external emotional control.
2007 Adaptive Multi-modal Sensors This paper presents a system that uses local diffusion and reinforcement learning to predict an entire input space based on partial observation, addressing bandwidth constraints in real-time input. It optimizes for multi-dimensional input spaces and maintains the ability to react to rare events.
2007 Predicting reactions from amino acid sequences in S. cerevisiae: an evolutionary computation approach
2007 Predicting reactions from amino acid sequences in S. cerevisiae: an evolutionary computation approach
2005 Teaching the Evolution of Behavior with SuperDuperWalker SuperDuperWalker is a framework for experiments on the evolution of locomotion, using a 3D physical simulation and a genetic algorithm to control agents. It supports an inquiry cycle for educational purposes.
2005 Selection Songs: Evolutionary Music Computation This article introduces evolutionary computation in the arts, focusing on the evolution of music-making systems using genetic programming. It discusses challenges in fitness assessment and experiments with evolving music-producing virtual organisms.