Video Microscopy Data Research Articles

Single molecule dynamics in a virtual cell combining a 3-dimensional matrix model with random walks

Recent advances in light microscopy have enabled single molecules to be imaged and tracked within living cells and this approach is impacting our understanding of cell biology. Computer modeling and simulation are important adjuncts to the experimental cycle since they aid interpretation of experimental results and help refine, test and generate hypotheses. Object-oriented computer modeling is particularly well-suited for simulating random, thermal, movements of individual molecules as they interact with other molecules and subcellular structures, but current models are often limited to idealized systems consisting of unit volumes or planar surfaces. Here, a simulation tool is described that combines a 3-dimensional, voxelated, representation of the cell consisting of subcellular structures (e.g. nucleus, endoplasmic reticulum, cytoskeleton, vesicles, and filopodia) combined with numerical floating-point precision simulation of thousands of individual molecules moving and interacting within the 3-dimensional space. Simulations produce realistic time-series video sequences comprising single fluorophore intensities and realistic background noise which can be directly compared to experimental fluorescence video microscopy data sets.

Real-time 3D tracking of swimming microbes using digital holographic microscopy and deep learning.

The three-dimensional swimming tracks of motile microorganisms can be used to identify their species, which holds promise for the rapid identification of bacterial pathogens. The tracks also provide detailed information on the cells' responses to external stimuli such as chemical gradients and physical objects. Digital holographic microscopy (DHM) is a well-established, but computationally intensive method for obtaining three-dimensional cell tracks from video microscopy data. We demonstrate that a common neural network (NN) accelerates the analysis of holographic data by an order of magnitude, enabling its use on single-board computers and in real time. We establish a heuristic relationship between the distance of a cell from the focal plane and the size of the bounding box assigned to it by the NN, allowing us to rapidly localise cells in three dimensions as they swim. This technique opens the possibility of providing real-time feedback in experiments, for example by monitoring and adapting the supply of nutrients to a microbial bioreactor in response to changes in the swimming phenotype of microbes, or for rapid identification of bacterial pathogens in drinking water or clinical samples.

Measuring DNA extension during unfolding in a centrifugal force microscope

DNA in sperm cells can form compact toroids when bound with proteins from the protamine family. To study DNA folding and toroid formation, we have built a centrifugal force microscope (CFM) to unfold DNA and observe the intermediate folding states. In a CFM, micron-sized beads tethered to the sample by a DNA strand are placed into a microscope within a desktop centrifuge. The centrifuge applies a centrifugal force (a pseudoforce caused by inertia) to our sample, unfolding the DNA. Recording and analyzing the bead motion can yield information on the extension of the DNA over time and the intermediate stages of DNA condensation. Our CFM is operational at low centrifuge speeds, but we did not have a method for measuring the extension of the DNA. Here, we will discuss our progress in creating a program that gathers and analyzes video microscopy data from the CFM to measure extension in real time. This program will be useful in collecting data on the pathway of DNA condensation and for other force spectroscopy measurements with the CFM.

Receptor-ligand non-equilibrium kinetics (RLNEK) 1.0: An integrated Trackmate laminar flow chamber analysis

Although parallel plate flow chamber assays are widely performed, extraction of kinetic parameters is limited to specialized labs with mathematical expertise and customized video-microscopy tracking tools. The recent development of Trackmate has increased researcher accessibility to tracking particles in video-microscopy experiments; however, there is a lack of tools that analyze this tracking information. We report a software tool, compatible with Trackmate, that extracts Receptor Ligand Non-Equilibrium Kinetic (RLNEK) parameters from video-microscopy data. This software should be of particular interest to the community of researchers and scientists interrogating the target-specific binding and release of immune cells.

Quantifying Longitudinal Network Geometry Alterations in Rat Skeletal Microvasculature Subject to Ischemia

Study objectiveIntra‐vital video microscopy (IVVM) data of rat subcutaneous skeletal muscle microvasculature under ischemic conditions at two time points (0 hours, 3 hours) was obtained from our collaborator. The objectives of this study are to elicit network geometries from the IVVM data and quantify the alterations in the network statistics of the geometries due to ischemia.HypothesisWe hypothesize that ischemia causes blood vessel de‐recruitment.MethodsIVVM data from six experiments was obtained, where each experiment provided a video at the start (0 hours) and end (3 hours) of the experiment. Using consecutive still images from the videos, vessel centerlines were manually drawn using Procreate and the NIH software, ImageJ. The included vessels were chosen based on two features: (1) visible red blood cell flow and (2) clearly defined edges. The centerlines were used to guide thresholding by identifying appropriate vessels. This was followed by the generation of binarized representations of the microvascular networks. The network statistics were quantified by measuring vessel segment lengths and diameters, counting the number of bifurcations, and calculating vessel density in ImageJ.ResultsAt 3 hours, the average vessel length changed from 283.91 + 252.14 pixels to 316.75 + 292.37 pixels, the average diameter changed from 16.02 + 7.99 pixels to 15.42 + 7.50 pixels, the average number of bifurcations changed from 32.42 to 18.92 and the average vessel density changed from 0.0085 + 0.0017 to 0.0061 + 0.0024.ConclusionThere is a clear decrease in average diameter, vessel density and the number of bifurcations characterizing the microvascular networks. However, an increase in the average vessel length accompanies these findings. The data indicates that ischemia may cause de‐recruitment of blood vessels.

Ultrasound-assisted membrane permeabilization of endothelial cells under flow conditions

Ultrasound-stimulated microbubbles have been shown a feasible approach for localized therapeutic delivery. As applications of this technique span many anatomical sites, so too do the local fluid dynamics experienced by the circulating microbubbles and the adjacent endothelial cells. Our objective was to assess the relative effectiveness of endothelial cell sonoporation as a function of flow conditions. Human umbilical vein (HUVECs) or human brain endothelial cells (HBECs) were cultured as a monolayer in flow chamber slides connected to a fluidic system and placedupon an acoustically-coupled microscope. A suspension of diluted lipid-encapsulated microbubbles and propidium iodide (PI), used as a sonoporation marker, was constantly perfused over the monolayer at either 5 or 30 ml/min. Cells were treated with 1 MHz ultrasound (PRI= 1 ms, 20 cycles, duration = 2 s), and the video-microscopy data were quantified offline to assess the number of PI-positive cells. Our results demonstrate a marked increase in sonoporation efficiency at 30 ml/min as compared to 5 ml/min in both endothelial cell lines under identical acoustic conditions (9.7-fold increase and 2.3-fold increase for HUVECs and HBECs respectively, p < 0.001). Our results suggest the local fluid flow environment plays a role in US-mediated endothelial perforation efficiency and can modulate treatment strategies.

Novel Analysis Method for Beating Cells Videomicroscopy Data: Functional Characterization of Culture Samples.

Cell culture of cardiac tissue analog is becoming increasingly interesting for regenerative medicine (cell therapy and tissue engineering) and is widely used for high throughput cardiotoxicity. As a cost-effective approach to rapidly discard new compounds with high toxicity risks, cardiotoxicity evaluation is firstly done in vitro requiring cells/tissue with physiological/pathological characteristics (close to in vivo properties). Studying multicellular electrophysiological and contractile properties is needed to assess drug effects. Techniques favoring process automation which could help in simplifying screening drug candidates are thus of central importance. A lot of effort has been made to ameliorate in vitro models including several in vitro platforms for engineering neonatal rat cardiac tissues. However, most of the initial evaluation is done by studying the rate of activity. In this study, we present new approaches that use the videomicroscopy video of monolayer activity to study contractile properties of beating cells in culture. Two new variables are proposed which are linked to the contraction dynamics and are dependent on the rhythm of activity. Methods for evaluation of regional synchronicity within the image field of view are also presented that can rapidly determine regions with abnormal activity or heterogeneity in contraction dynamics.

Quantitative digital microscopy with deep learning

Video microscopy has a long history of providing insight and breakthroughs for a broad range of disciplines, from physics to biology. Image analysis to extract quantitative information from video microscopy data has traditionally relied on algorithmic approaches, which are often difficult to implement, time-consuming, and computationally expensive. Recently, alternative data-driven approaches using deep learning have greatly improved quantitative digital microscopy, potentially offering automatized, accurate, and fast image analysis. However, the combination of deep learning and video microscopy remains underutilized primarily due to the steep learning curve involved in developing custom deep-learning solutions. To overcome this issue, we introduce software, DeepTrack 2.0, to design, train, and validate deep-learning solutions for digital microscopy. We use this software to exemplify how deep learning can be employed for a broad range of applications, from particle localization, tracking, and characterization, to cell counting and classification. Thanks to its user-friendly graphical interface, DeepTrack 2.0 can be easily customized for user-specific applications, and thanks to its open-source, object-oriented programing, it can be easily expanded to add features and functionalities, potentially introducing deep-learning-enhanced video microscopy to a far wider audience.

Motility Analysis of Trypanosomatids.

Motility analysis of microswimmers has long been limited to a few model cell types and broadly restricted by technical challenges of high-resolution in vivo microscopy. Recently, interdisciplinary interest in detailed analysis of the motile behavior of various species has gained momentum. Here we describe a basic protocol for motility analysis of an important, highly diverse group of eukaryotic flagellate microswimmers, using high spatiotemporal resolution videomicroscopy. Further, we provide a special, time-dependent tomographic approach for the proof of rotational locomotion of periodically oscillating microswimmers, using the same data. Taken together, the methods describe part of an integrative approach to generate decisive information on three-dimensional in vivo motility from standard two-dimensional videomicroscopy data.

Dynamical analysis of bacteria in microscopy movies.

Recent advances in microscopy, computing power and image processing have enabled the analysis of ever larger datasets of movies of microorganisms to study their behaviour. However, techniques for analysing the dynamics of individual cells from such datasets are not yet widely available in the public domain. We recently demonstrated significant phenotypic heterogeneity in the adhesion of Escherichia coli bacteria to glass surfaces using a new method for the high-throughput analysis of video microscopy data. Here, we present an in-depth analysis of this method and its limitations, and make public our algorithms for following the positions and orientations of individual rod-shaped bacteria from time-series of 2D images to reconstruct their trajectories and characterise their dynamics. We demonstrate in detail how to use these algorithms to identify different types of adhesive dynamics within a clonal population of bacteria sedimenting onto a surface. The effects of measurement errors in cell positions and of limited trajectory durations on our results are discussed.

It Does Not Do to Dwell on Single Components and Forget the Importance of Complete Networks: Optimizing an Integrated Hemodynamic Model Derived from Experimental Data

In our previous study, we demonstrated the important role of capillary resistance and venular network geometry on arteriolar blood flow and red blood cell (RBC) distribution in terminal arterioles (TAs). Using arteriolar and corresponding venular networks reconstructed from intravital videomicroscopy (IVVM) data obtained in rat gluteus maximus preparations, as well as mathematical modelling and simulation, we showed that for RBC flow, adding the same resistance to all TAs to represent downstream capillary beds significantly decreased the coefficient of variation of TA RBC flow (CVTA(RBC), standard deviation/mean, n = 8 networks) by 37%, whereas, adding the venular network did not further significantly change CVTA(RBC). However, we found that adding constant resistive elements to the TAs did not significantly decrease the coefficient of variation of TA tube hematocrit, whereas, adding the venular network significantly decreased CVTA(HT) by 20%. Given the need for this network‐oriented approach, we are optimizing our arteriolar network analysis to include as much detail as possible through experimental acquisition and data reconstruction of arterioles and venules, as well as estimation of missing data using theoretical methods. Our goal is to develop an analysis technique that accounts for the interconnectivity of microvascular systems, and that can be applied to networks in a wide range of situations. In our current methods, we reconstructed corresponding arteriolar and venular networks from experimental data, and estimated and applied total capillary resistance for each network, based on the arteriolar network resistance and the relative pressure drop between the arteriolar and capillary sections of the network. The capillary resistance is now distributed to each TA segment according to its diameter, and therefore, variable. We acquired fluorescent streaks for experimental blood flow data in an arteriolar network, which we used to validate flow values predicted by our model. Using the experimental flow data, we also calculated a Murray's law exponent of approximately a=2.8, to which we compared predicted values. For 3 networks, we found a=2.78±0.30 using variable capillary resistance vs. a=1.95±0.17 using constant capillary resistance. Our results, using improved theoretical methods and newly acquired flow data, show that our network‐oriented approach is moving towards more accurately predicting hemodynamic properties of arteriolar networks under normal baseline conditions. We are currently working to extend this approach and apply it to networks under different experimental conditions.Support or Funding InformationThis work was supported by Natural Sciences and Engineering Research Council of Canada (NSERC) Grants R4081A03 (DG) and R4218A03 (DNJ).This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

The Microvascular Lattice: An Updated Paradigm of Flow Distribution Through Capillary Networks

BackgroundCapillaries, as the smallest vessels in the microcirculation, represent the most essential site for oxygen delivery; yet, given their overwhelming density and complicated geometry, understanding how they integrate within the broader microvascular network has been challenging.Classical microvascular flow modelling has used linear branching networks such that each capillary bundle (CB) corresponds to a unique arteriolar‐venular pair with one‐directional pathway of flow. However, it has long been observed in vivo that terminal arterioles supply multiple CBs and similarly, post‐capillary venules collect from multiple CBs. Such a network forms a lattice‐like structure with an alternating pattern of arterioles, CBs, and venules.ObjectiveWe aim to develop a mathematical model for the microvascular lattice that can be applied to study blood flow properties through capillary networks in skeletal muscle.Methods/ResultsMicrovascular lattice networks are generated from our own rodent intravital videomicroscopy data and compared to data reported in the literature. Distinct CBs are selected and their anatomical features are delineated with 3‐D reconstruction software. Capillary hemodynamics (RBC velocity, RBC supply rate, hematocrit) are extracted from video data and used to characterize the flow within and between multiple CBs. The generalized architecture of a microvascular lattice is then extrapolated from our experimental data and applied to a dual‐phase steady‐state blood flow model. Flow properties of the lattice are compared to those of a classical linear branching network ‐ both containing the same number of CBs.Conclusions/ImpactThe microvascular lattice represents an updated paradigm for describing blood flow through capillary networks. This lattice structure raises important questions regarding how flow can be regulated to support local oxygen demand and how larger microvessels interact with capillaries within the microcirculation. Furthermore, CB functional data for a variety of microvascular conditions can be inputted into this lattice model to determine how alterations to capillary architecture affect overall flow through the microvascular network.This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.

Differential Variance Analysis: a direct method to quantify and visualize dynamic heterogeneities

Many amorphous materials show spatially heterogenous dynamics, as different regions of the same system relax at different rates. Such a signature, known as Dynamic Heterogeneity, has been crucial to understand the nature of the jamming transition in simple model systems and is currently considered very promising to characterize more complex fluids of industrial and biological relevance. Unfortunately, measurements of dynamic heterogeneities typically require sophisticated experimental set-ups and are performed by few specialized groups. It is now possible to quantitatively characterize the relaxation process and the emergence of dynamic heterogeneities using a straightforward method, here validated on video microscopy data of hard-sphere colloidal glasses. We call this method Differential Variance Analysis (DVA), since it focuses on the variance of the differential frames, obtained subtracting images at different time-lags. Moreover, direct visualization of dynamic heterogeneities naturally appears in the differential frames, when the time-lag is set to the one corresponding to the maximum dynamic susceptibility. This approach opens the way to effectively characterize and tailor a wide variety of soft materials, from complex formulated products to biological tissues.

Waves of DNA: Propagating excitations in extended nanoconfined polymers.

We use a nanofluidic system to investigate the emergence of thermally driven collective phenomena along a single polymer chain. In our approach, a single DNA molecule is confined in a nanofluidic slit etched with arrays of embedded nanocavities; the cavity lattice is designed so that a single chain occupies multiple cavities. Fluorescent video-microscopy data shows fluctuations in intensity between cavities, including waves of excess fluorescence that propagate across the cavity-straddling molecule, corresponding to propagating fluctuations of contour overdensity in the cavities. The transfer of DNA between neighboring pits is quantified by examining the correlation in intensity fluctuations between neighboring cavities. Correlations grow from an anticorrelated minimum to a correlated maximum before decaying, corresponding to a transfer of contour between neighboring cavities at a fixed transfer time scale. The observed dynamics can be modeled using Langevin dynamics simulations and a minimal lattice model of coupled diffusion. This study shows how confinement-based sculpting of the polymer equilibrium configuration, by renormalizing the physical system into a series of discrete cavity states, can lead to new types of dynamic collective phenomena.

Analysis-preserving video microscopy compression via correlation and mathematical morphology.

The large amount video data produced by multi-channel, high-resolution microscopy system drives the need for a new high-performance domain-specific video compression technique. We describe a novel compression method for video microscopy data. The method is based on Pearson's correlation and mathematical morphology. The method makes use of the point-spread function (PSF) in the microscopy video acquisition phase. We compare our method to other lossless compression methods and to lossy JPEG, JPEG2000, and H.264 compression for various kinds of video microscopy data including fluorescence video and brightfield video. We find that for certain data sets, the new method compresses much better than lossless compression with no impact on analysis results. It achieved a best compressed size of 0.77% of the original size, 25× smaller than the best lossless technique (which yields 20% for the same video). The compressed size scales with the video's scientific data content. Further testing showed that existing lossy algorithms greatly impacted data analysis at similar compression sizes.

Abstract 5337: An in silico platform for characterizing ADC bystander effects

Abstract Introduction The “bystander effect” is the hypothesized process by which neighboring cancer cells are killed when free drug is released into the tumor environment by dying cells. The extent to which an antibody drug conjugate (ADC) elicits the bystander effect can have significant impact on the magnitude and dynamics of efficacy. Understanding the bystander effect though experimental methods (e.g. video-microscopy) are challenging; however, when supplemented with in silico modeling, mechanistic details can be elucidated. The purpose of this work is to develop a spatial-temporal in silico platform describing tumor cell progression in the presence of an ADC in order to quantify the bystander effect. Methods The model employs spatial-stochastic methods to simulate the time evolution of cell death resulting from ADCs and freely diffusing payload based on following parameters: drug to antibody ratio (DAR), linker stability, antigen concentration, ADC cell membrane permeability, ADC-antigen internalization, payload permeability, and payload potency. Specifically, a 2-D rectangular grid of tumor cells, ADCs, and payload are numerically implemented through binary pixel occupation of the grid. Governing rules are then assigned for: mobility for freely diffusing ADCs and payload, reaction rate for ADC binding to antigen-expressing cells, rate of cell death dependent on payload concentration, and internalization rate of payload into non-antigen expressing cells. The simulation is run until stopping criteria are met, e.g. all cells are killed. Baseline parameters can be optimized by fitting the model to in vitro experimental video-microscopy data. Results Experimental conditions were simulated with the model to characterize the influence of the aforementioned ADC and biological parameters on the bystander effect. Conclusions By incorporating the spatial aspect of ADC mechanics, spatially dependent metrics can be formulated to quantify the bystander effect. The model allows for wide adjustability of the ADC design parameters as well as relevant biological parameters. Thus, by understanding the relative parameter sensitivities, the model can explore design optimization of ADCs that maximize the bystander effect. Through such optimization, the magnitude and dynamics of drug exposure at the site of action can be controlled to maximize efficacy and improve the therapeutic window. Citation Format: Jackson Burton, Shu-Wen Teng, Christopher J. Zopf, Ryan Nolan, Arijit Chakravarty. An in silico platform for characterizing ADC bystander effects. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 5337. doi:10.1158/1538-7445.AM2015-5337

γδ Intraepithelial Lymphocyte Migration Limits Transepithelial Pathogen Invasion and Systemic Disease in Mice

Intraepithelial lymphocytes that express the γδ T-cell receptor (γδ IELs) limit pathogen translocation across the intestinal epithelium by unknown mechanisms. We investigated whether γδ IEL migration and interaction with epithelial cells promote mucosal barrier maintenance during enteric infection. Salmonella typhimurium or Toxoplasma gondii were administered to knockout (KO) mice lacking either the T cell receptor δ chain (Tcrd) or CD103, or control TcrdEGFP C57BL/6 reporter mice. Intravital microscopy was used to visualize migration of green fluorescent protein (GFP)-tagged γδ T cells within the small intestinal mucosa of mice infected with DsRed-labeled S typhimurium. Mixed bone marrow chimeras were generated to assess the effects of γδ IEL migration on early pathogen invasion and chronic systemic infection. Morphometric analyses of intravital video microscopy data showed that γδ IELs rapidly localized to and remained near epithelial cells in direct contact with bacteria. Within 1 hour, greater numbers of T gondii or S typhimurium were present within mucosae of mice with migration-defective occludin KO γδ T cells, compared with controls. Pathogen invasion in Tcrd KO mice was quantitatively similar to that in mice with occludin-deficient γδ T cells, whereas invasion in CD103 KO mice, which have increased migration of γδ T cells into the lateral intercellular space, was reduced by 63%. Consistent with a role of γδ T-cell migration in early host defense, systemic salmonellosis developed more rapidly and with greater severity in mice with occludin-deficient γδ IELs, relative to those with wild-type or CD103 KO γδ IELs. In mice, intraepithelial migration to epithelial cells in contact with pathogens is essential to γδ IEL surveillance and immediate host defense. γδ IEL occludin is required for early surveillance that limits systemic disease.

Machine-learning approach to holographic particle characterization.

Holograms of colloidal dispersions encode comprehensive information about individual particles' three-dimensional positions, sizes and optical properties. Extracting that information typically is computationally intensive, and thus slow. Here, we demonstrate that machine-learning techniques based on support vector machines (SVMs) can analyze holographic video microscopy data in real time on low-power computers. The resulting stream of precise particle-resolved tracking and characterization data provides unparalleled insights into the composition and dynamics of colloidal dispersions and enables applications ranging from basic research to process control and quality assurance.

Measuring Ligand-Receptor Binding Rates with K-Space Image Correlation Spectroscopy: Theory and Experimental Applications

The ability to measure accurate kinetic binding rates of ligand-receptor and receptor-receptor interactions is important in order to study the triggering of signaling events in cell membranes. We apply an image analysis technique, k-space image correlation spectroscopy (kICS), to measure the kinetic binding rates of ligand-receptor interactions from fluorescence video-microscopy data occurring within a thin plane of illumination. We consider two different kinds of experiments where: (1) fluorescently tagged ligands in solution bind and unbind dynamically to unlabeled receptors which are allowed to diffuse in the imaging plane and, (2) fluorescently tagged, diffusing receptors in the imaging plane interact with other (possibly non-fluorescing) receptors and interconvert between two diffusive states. For system (1), by measuring image intensity fluctuations of fluorescently labeled human doublecortin proteins in solution (2.5 nM), binding to microtubules adhering on a sample substrate, the kICS method accurately determines kinetic dissociation rates (kd = 1.6 ± 0.2 s−1) as well as diffusion coefficients (0.38 ± 0.19 μm2/s) of the proteins consistent with single-particle tracking data. For system (2), we perform drug-trial experiments on living COS-7 cells to show that kICS can quantify the degree to which the drugs will affect kinetic binding/unbinding rates. By labeling the glycolipid membrane receptor GM1 with a fluorescent cholera toxin B-subunit, we show that cytoskeleton perturbations with the drug cytochalasin D result in significant changes (ku = 0.013 ± 0.004 s−1 to 0.08 ± 0.02 s−1) to the undocking rate of GM1 to the actin cytoskeleton. With these two proof-of-concept experiments, we pave the way to using kICS as an efficient and reliable method to elucidate kinetic binding rates and transport dynamics parameters for studies in biomembranes.

The Neutrophil's Eye-View: Inference and Visualisation of the Chemoattractant Field Driving Cell Chemotaxis In Vivo

As we begin to understand the signals that drive chemotaxis in vivo, it is becoming clear that there is a complex interplay of chemotactic factors, which changes over time as the inflammatory response evolves. New animal models such as transgenic lines of zebrafish, which are near transparent and where the neutrophils express a green fluorescent protein, have the potential to greatly increase our understanding of the chemotactic process under conditions of wounding and infection from video microscopy data. Measurement of the chemoattractants over space (and their evolution over time) is a key objective for understanding the signals driving neutrophil chemotaxis. However, it is not possible to measure and visualise the most important contributors to in vivo chemotaxis, and in fact the understanding of the main contributors at any particular time is incomplete. The key insight that we make in this investigation is that the neutrophils themselves are sensing the underlying field that is driving their action and we can use the observations of neutrophil movement to infer the hidden net chemoattractant field by use of a novel computational framework. We apply the methodology to multiple in vivo neutrophil recruitment data sets to demonstrate this new technique and find that the method provides consistent estimates of the chemoattractant field across the majority of experiments. The framework that we derive represents an important new methodology for cell biologists investigating the signalling processes driving cell chemotaxis, which we label the neutrophils eye-view of the chemoattractant field.