Machine learning interpretable models of cell mechanics from protein images

Tissue Active Matter: Integrating Mechanics and Signaling into Dynamical Models

The importance of physical forces in the morphogenesis, homeostatic function, and pathological dysfunction of multicellular tissues is being increasingly characterized, both theoretically and experimentally. Analogies between biological systems and inert materials such as foams, gels, and liquid crystals have provided striking insights into the core design principles underlying multicellular organization. However, these connections can seem surprising given that a key feature of multicellular systems is their ability to constantly consume energy, providing an active origin for the forces that they produce. Key emerging questions are, therefore, to understand whether and how this activity grants tissues novel properties that do not have counterparts in classical materials, as well as their consequences for biological function. Here, we review recent discoveries at the intersection of active matter and tissue biology, with an emphasis on how modeling and experiments can be combined to understand the dynamics of multicellular systems. These approaches suggest that a number of key biological tissue-scale phenomena, such as morphogenetic shape changes, collective migration, or fate decisions, share unifying design principles that can be described by physical models of tissue active matter.

Synergistic Anticancer Strategy Targeting ECM Stiffness: Integration of Matrix Softening and Mechanical Signal Transduction Blockade in Primary Liver Cancers

The development of primary liver cancer (hepatocellular carcinoma [HCC] and intrahepatic cholangiocarcinoma [ICC]) is linked to its physical microenvironment, particularly extracellular matrix (ECM) stiffness. Potential anticancer strategies targeting ECM stiffness include prevention/reversal of the stiffening process and disruption of the response of cancer cells to mechanical signals from ECM. However, each strategy has limitations. Therefore, the authors propose integrating them to maximize their strengths. Compared with HCC, ICC has a stiffer ECM and a worse prognosis. Therefore, ICC is selected to investigate mechanisms underlying the influence of ECM stiffness on cancer progression and application of the integrated anticancer strategy targeting ECM stiffness. In summary, immunofluorescence results for 181 primary liver cancer tissue chips (ICC, n = 91; HCC, n = 90) and analysis of TCGA mRNA-sequencing demonstrate that ECM stiffness can affect phenotypes of primary liver cancers. The YAP1/ABHD11-AS1/STAU2/ZYX/p-YAP1 pathway is a useful entry point for exploration of specific mechanisms of mechanical signal conduction from the ECM in ICC cells and their impact on cancer progression. Moreover, a synergistic anticancer strategy targeting ECM stiffness (ICCM@NPs + siABHD11-AS1@BAPN) is constructed by integrating ECM softening and blocking intracellular mechanical signal transduction in ICC and can provide insights for the treatment of cancers characterized by stiff ECM.

Biomechanics in the tumor microenvironment: from biological functions to potential clinical applications

Immune checkpoint therapies have spearheaded drug innovation over the last decade, propelling cancer treatments toward a new era of precision therapies. Nonetheless, the challenges of low response rates and prevalent drug resistance underscore the imperative for a deeper understanding of the tumor microenvironment (TME) and the pursuit of novel targets. Recent findings have revealed the profound impacts of biomechanical forces within the tumor microenvironment on immune surveillance and tumor progression in both murine models and clinical settings. Furthermore, the pharmacological or genetic manipulation of mechanical checkpoints, such as PIEZO1, DDR1, YAP/TAZ, and TRPV4, has shown remarkable potential in immune activation and eradication of tumors. In this review, we delved into the underlying biomechanical mechanisms and the resulting intricate biological meaning in the TME, focusing mainly on the extracellular matrix, the stiffness of cancer cells, and immune synapses. We also summarized the methodologies employed for biomechanical research and the potential clinical translation derived from current evidence. This comprehensive review of biomechanics will enhance the understanding of the functional role of biomechanical forces and provide basic knowledge for the discovery of novel therapeutic targets.

Data-driven discovery and parameter estimation of mathematical models in biological pattern formation

Mathematical modeling has been utilized to explain biological pattern formation, but the selections of models and parameters have been made empirically. In the present study, we propose a data-driven approach to validate the applicability of mathematical models. Specifically, we developed methods to automatically select the appropriate mathematical models based on the patterns of interest and to estimate the model parameters. For model selection, we employed Contrastive Language-Image Pre-training (CLIP) for zero-shot feature extraction, mapping the given pattern images to latent space and specifying the appropriate model. For parameter estimation, we developed a novel technique that rapidly performs approximate Bayesian inference based on Natural Gradient Boosting (NGBoost). This method allows for parameter estimation under minimal constraints; i.e., it does not require time-series data or initial conditions and is applicable to various types of mathematical models. We tested the method with Turing patterns and demonstrated its high accuracy and correspondence to analytical features. Our strategy enables efficient validation of mathematical models using spatial patterns.

Peeking into the future: inferring mechanics in dynamical tissues

Cells exert forces on each other and their environment, shaping the tissue. The resulting mechanical stresses can be determined experimentally or estimated computationally using stress inference methods. Over the years, mechanical stress inference has become a non-invasive, low-cost computational method for estimating the relative intercellular stresses and intracellular pressures of tissues. This mini-review introduces and compares the static and dynamic modalities of stress inference, considering their advantages and limitations. To date, most software has focused on static inference, which requires only a single microscopy image as input. Although applicable in quasi-equilibrium states, this approach neglects the influence that cell rearrangements might have on the inference. In contrast, dynamic stress inference relies on a time series of microscopy images to estimate stresses and pressures. Here, we discuss both static and dynamic mechanical stress inference in terms of their physical, mathematical, and computational foundations and then outline what we believe are promising avenues for in silico inference of the mechanical states of tissues.

Tumor Biomechanics-Inspired Future Medicine

Malignant tumors pose a significant global health challenge, severely threatening human health. Statistics from the World Health Organization indicate that, in 2022, there were nearly 20 million new cancer cases and 9.7 million cancer-related deaths. Therefore, it is urgently necessary to study the pathogenesis of cancer and explore effective diagnostic and treatment strategies. In recent years, research has highlighted the importance of mechanical cues in tumors, which have become a new hallmark of cancer and a key factor in regulating tumor behavior. This suggests that studying the mechanical properties of tumors may open potential new avenues for understanding the pathogenesis, diagnosis, and therapeutic intervention of cancer. This review summarizes the mechanical characteristics of tumors and the development of tumor diagnostics and treatments targeting specific mechanical factors. Finally, we propose new ideas and insights for the application of mechanomedicine in cancer diagnosis and treatment in the future.

Invited Review for 20th Anniversary Special Issue of PLRev "AI for Mechanomedicine"

Mechanomedicine is an interdisciplinary field that combines different areas including biomechanics, mechanobiology, and clinical applications like mechanodiagnosis and mechanotherapy. The emergence of artificial intelligence (AI) has revolutionized mechanomedicine, providing advanced tools to analyze the complex interactions between mechanics and biology. This review explores how AI impacts mechanomedicine across four key aspects, i.e., biomechanics, mechanobiology, mechanodiagnosis, and mechanotherapy. AI improves the accuracy of biomechanical characterizations and models, deepens the understanding of cellular mechanotransduction pathways, and enables early disease detection through mechanodiagnosis. In addition, AI optimizes mechanotherapy that targets biomechanical features and mechanobiological markers by personalizing treatment strategies based on real-time patient data. Even with these advancements, challenges still exist, particularly in data quality and the ethical integration into AI in clinical practice. The integration of AI with mechanomedicine offers transformative potential, enabling more accurate diagnostics and personalized treatments, and discovering novel mechanobiological pathways.

T cells use focal adhesions to pull themselves through confined environments

Immune cells are highly dynamic and able to migrate through environments with diverse biochemical and mechanical compositions. Their migration has classically been defined as amoeboid under the assumption that it is integrin independent. Here, we show that activated primary Th1 T cells require both confinement and extracellular matrix proteins to migrate efficiently. This migration is mediated through small and dynamic focal adhesions that are composed of the same proteins associated with canonical mesenchymal cell focal adhesions, such as integrins, talin, and vinculin. These focal adhesions, furthermore, localize to sites of contractile traction stresses, enabling T cells to pull themselves through confined spaces. Finally, we show that Th1 T cells preferentially follow tracks of other T cells, suggesting that these adhesions modify the extracellular matrix to provide additional environmental guidance cues. These results demonstrate not only that the boundaries between amoeboid and mesenchymal migration modes are ambiguous, but that integrin-mediated focal adhesions play a key role in T cell motility.

Cell dynamics revealed by microscopy advances

Cell biology emerges from spatiotemporally coordinated molecular processes. Recent advances in live-cell microscopy, fueled by a surge in optical, molecular, and computational technologies, have enabled dynamic observations from single molecules to whole organisms. Despite technological leaps, there is still an untapped opportunity to fully leverage their capabilities toward biological insight. We highlight how single-molecule imaging has transformed our understanding of biological processes, with a focus on chromatin organization and transcription in the nucleus. We describe how this was enabled by the close integration of new imaging techniques with analysis tools and discuss the challenges to make a comparable impact at larger scales from organelles to organisms. By highlighting recent successful examples, we describe an outlook of ever-increasing data and the need for seamless integration between dataset visualization and quantification to realize the full potential warranted by advances in new imaging technologies.

Inferring cellular contractile forces and work using deep morphology traction microscopy

Traction-force microscopy (TFM) has emerged as a widely used standard methodology to measure cell-generated traction forces and determine their role in regulating cell behavior. While TFM platforms have enabled many discoveries, their implementation remains limited due to complex experimental procedures, specialized substrates, and the ill-posed inverse problem whereby low-magnitude high-frequency noise in the displacement field severely contaminates the resulting traction measurements. Here, we introduce deep morphology traction microscopy (DeepMorphoTM), a deep-learning alternative to conventional TFM approaches. DeepMorphoTM first infers cell-induced substrate displacement solely from a sequence of cell shapes and subsequently computes cellular traction forces, thus avoiding the requirement of a specialized fiduciarily marked deformable substrate or force-free reference image. Rather, this technique drastically simplifies the overall experimental methodology, imaging, and analysis needed to conduct cell-contractility measurements. We demonstrate that DeepMorphoTM quantitatively matches conventional TFM results while offering stability against the biological variability in cell contractility for a given cell shape. Without high-frequency noise in the inferred displacement, DeepMorphoTM also resolves the ill-posedness of traction computation, increasing the consistency and accuracy of traction analysis. We demonstrate the accurate extrapolation across several cell types and substrate materials, suggesting robustness of the methodology. Accordingly, we present DeepMorphoTM as a capable yet simpler alternative to conventional TFM for characterizing cellular contractility in two dimensions.

Interpreting Neural Operators: How Nonlinear Waves Propagate in Nonreciprocal Solids

We present a data-driven pipeline for model building that combines interpretable machine learning, hydrodynamic theories, and microscopic models. The goal is to uncover the underlying processes governing nonlinear dynamics experiments. We exemplify our method with data from microfluidic experiments where crystals of streaming droplets support the propagation of nonlinear waves absent in passive crystals. By combining physics-inspired neural networks, known as neural operators, with symbolic regression tools, we infer the solution, as well as the mathematical form, of a nonlinear dynamical system that accurately models the experimental data. Finally, we interpret this continuum model from fundamental physics principles. Informed by machine learning, we coarse grain a microscopic model of interacting droplets and discover that nonreciprocal hydrodynamic interactions stabilize and promote nonlinear wave propagation.

Myosin forces elicit an F-actin structural landscape that mediates mechanosensitive protein recognition

Cells mechanically interface with their surroundings through cytoskeleton-linked adhesions, allowing them to sense physical cues that instruct development and drive diseases such as cancer. Contractile forces generated by myosin motor proteins mediate these mechanical signal transduction processes through unclear protein structural mechanisms. Here, we show that myosin forces elicit structural changes in actin filaments (F-actin) that modulate binding by the mechanosensitive adhesion protein α-catenin. Using correlative cryo-fluorescence microscopy and cryo-electron tomography, we identify F-actin featuring domains of nanoscale oscillating curvature at cytoskeleton-adhesion interfaces enriched in zyxin, a marker of actin-myosin generated traction forces. We next introduce a reconstitution system for visualizing F-actin in the presence of myosin forces with cryo-electron microscopy, which reveals morphologically similar superhelical F-actin spirals. In simulations, transient forces mimicking tugging and release of filaments by motors produce spirals, supporting a mechanistic link to myosin's ATPase mechanochemical cycle. Three-dimensional reconstruction of spirals uncovers extensive asymmetric remodeling of F-actin's helical lattice. This is recognized by α-catenin, which cooperatively binds along individual strands, preferentially engaging interfaces featuring extended inter-subunit distances while simultaneously suppressing rotational deviations to regularize the lattice. Collectively, we find that myosin forces can deform F-actin, generating a conformational landscape that is detected and reciprocally modulated by a mechanosensitive protein, providing a direct structural glimpse at active force transduction through the cytoskeleton.

In silico labeling in cell biology: Potential and limitations

In silico labeling is the computational cross-modality image translation where the output modality is a subcellular marker that is not specifically encoded in the input image, for example, in silico localization of organelles from transmitted light images. In principle, in silico labeling has the potential to facilitate rapid live imaging of multiple organelles with reduced photobleaching and phototoxicity, a technology enabling a major leap toward understanding the cell as an integrated complex system. However, five years have passed since feasibility was attained, without any demonstration of using in silico labeling to uncover new biological insight. In here, we discuss the current state of in silico labeling, the limitations preventing it from becoming a practical tool, and how we can overcome these limitations to reach its full potential.

The evolution of developmental biology through conceptual and technological revolutions

Developmental biology-the study of the processes by which cells, tissues, and organisms develop and change over time-has entered a new golden age. After the molecular genetics revolution in the 80s and 90s and the diversification of the field in the early 21st century, we have entered a phase when powerful technologies provide new approaches and open unexplored avenues. Progress in the field has been accelerated by advances in genomics, imaging, engineering, and computational biology and by emerging model systems ranging from tardigrades to organoids. We summarize how revolutionary technologies have led to remarkable progress in understanding animal development. We describe how classic questions in gene regulation, pattern formation, morphogenesis, organogenesis, and stem cell biology are being revisited. We discuss the connections of development with evolution, self-organization, metabolism, time, and ecology. We speculate how developmental biology might evolve in an era of synthetic biology, artificial intelligence, and human engineering.

Focal adhesions are controlled by microtubules through local contractility regulation

Microtubules regulate cell polarity and migration via local activation of focal adhesion turnover, but the mechanism of this process is insufficiently understood. Molecular complexes containing KANK family proteins connect microtubules with talin, the major component of focal adhesions. Here, local optogenetic activation of KANK1-mediated microtubule/talin linkage promoted microtubule targeting to an individual focal adhesion and subsequent withdrawal, resulting in focal adhesion centripetal sliding and rapid disassembly. This sliding is preceded by a local increase of traction force due to accumulation of myosin-II and actin in the proximity of the focal adhesion. Knockdown of the Rho activator GEF-H1 prevented development of traction force and abolished sliding and disassembly of focal adhesions upon KANK1 activation. Other players participating in microtubule-driven, KANK-dependent focal adhesion disassembly include kinases ROCK, PAK, and FAK, as well as microtubules/focal adhesion-associated proteins kinesin-1, APC, and αTAT. Based on these data, we develop a mathematical model for a microtubule-driven focal adhesion disruption involving local GEF-H1/RhoA/ROCK-dependent activation of contractility, which is consistent with experimental data.

MechanoBase: a comprehensive database for the mechanics of tissues and cells

Mechanical aspects of tissues and cells critically influence a myriad of biological processes and can substantially alter the course of diverse diseases. The emergence of diverse methodologies adapted from physical science now permits the precise quantification of the cellular forces and the mechanical properties of tissues and cells. Despite the rising interest in tissue and cellular mechanics across fields like biology, bioengineering and medicine, there remains a noticeable absence of a comprehensive and readily accessible repository of this pertinent information. To fill this gap, we present MechanoBase, a comprehensive tissue and cellular mechanics database, curating 57 480 records from 5634 PubMed articles. The records archived in MechanoBase encompass a range of mechanical properties and forces, such as modulus and tractions, which have been measured utilizing various technical approaches. These measurements span hundreds of biosamples across more than 400 species studied under diverse conditions. Aiming for broad applicability, we design MechanoBase with user-friendly search, browsing and data download features, making it a versatile tool for exploring biomechanical attributes in various biological contexts. Moreover, we add complementary resources, including the principles of popular techniques, the concepts of mechanobiology terms and the cellular and tissue-level expression of related genes, offering scientists unprecedented access to a wealth of knowledge in this field of research. Database URL: https://zhanglab-web.tongji.edu.cn/mechanobase/ and https://compbio-zhanglab.org/mechanobase/.

Functional regimes define the response of the soil microbiome to environmental change

The metabolic activity of soil microbiomes plays a central role in carbon and nitrogen cycling. Given the changing climate, it is important to understand how the metabolism of natural communities responds to environmental change. However, the ecological, spatial, and chemical complexity of soils makes understanding the mechanisms governing the response of these communities to perturbations challenging. Here, we overcome this complexity by using dynamic measurements of metabolism in microcosms and modeling to reveal regimes where a few key mechanisms govern the response of soils to environmental change. We sample soils along a natural pH gradient, construct >1500 microcosms to perturb the pH, and quantify the dynamics of respiratory nitrate utilization, a key process in the nitrogen cycle. Despite the complexity of the soil microbiome, a minimal mathematical model with two variables, the quantity of active biomass in the community and the availability of a growth-limiting nutrient, quantifies observed nitrate utilization dynamics across soils and pH perturbations. Across environmental perturbations, changes in these two variables give rise to three functional regimes each with qualitatively distinct dynamics of nitrate utilization over time: a regime where acidic perturbations induce cell death that limits metabolic activity, a nutrient-limiting regime where nitrate uptake is performed by dominant taxa that utilize nutrients released from the soil matrix, and a resurgent growth regime in basic conditions, where excess nutrients enable growth of initially rare taxa. The underlying mechanism of each regime is predicted by our interpretable model and tested via amendment experiments, nutrient measurements, and sequencing. Further, our data suggest that the long-term history of environmental variation in the wild influences the transitions between functional regimes. Therefore, quantitative measurements and a mathematical model reveal the existence of qualitative regimes that capture the mechanisms and dynamics of a community responding to environmental change.

Establishing a conceptual framework for holistic cell states and state transitions

Cell states were traditionally defined by how they looked, where they were located, and what functions they performed. In this post-genomic era, the field is largely focused on a molecular view of cell state. Moving forward, we anticipate that the observables used to define cell states will evolve again as single-cell imaging and analytics are advancing at a breakneck pace via the collection of large-scale, systematic cell image datasets and the application of quantitative image-based data science methods. This is, therefore, a key moment in the arc of cell biological research to develop approaches that integrate the spatiotemporal observables of the physical structure and organization of the cell with molecular observables toward the concept of a holistic cell state. In this perspective, we propose a conceptual framework for holistic cell states and state transitions that is data-driven, practical, and useful to enable integrative analyses and modeling across many data types.

Learning dynamical models of single and collective cell migration: a review

Single and collective cell migration are fundamental processes critical for physiological phenomena ranging from embryonic development and immune response to wound healing and cancer metastasis. To understand cell migration from a physical perspective, a broad variety of models for the underlying physical mechanisms that govern cell motility have been developed. A key challenge in the development of such models is how to connect them to experimental observations, which often exhibit complex stochastic behaviours. In this review, we discuss recent advances in data-driven theoretical approaches that directly connect with experimental data to infer dynamical models of stochastic cell migration. Leveraging advances in nanofabrication, image analysis, and tracking technology, experimental studies now provide unprecedented large datasets on cellular dynamics. In parallel, theoretical efforts have been directed towards integrating such datasets into physical models from the single cell to the tissue scale with the aim of conceptualising the emergent behaviour of cells. We first review how this inference problem has been addressed in both freely migrating and confined cells. Next, we discuss why these dynamics typically take the form of underdamped stochastic equations of motion, and how such equations can be inferred from data. We then review applications of data-driven inference and machine learning approaches to heterogeneity in cell behaviour, subcellular degrees of freedom, and to the collective dynamics of multicellular systems. Across these applications, we emphasise how data-driven methods can be integrated with physical active matter models of migrating cells, and help reveal how underlying molecular mechanisms control cell behaviour. Together, these data-driven approaches are a promising avenue for building physical models of cell migration directly from experimental data, and for providing conceptual links between different length-scales of description.