Non-melanoma skin cancer segmentation for histopathology dataset.

Densely labelled segmentation data for digital pathology images is costly to produce but is invaluable to training effective machine learning models. We make available 290 hand-annotated histopathology tissue sections of the 3 most common skin cancers; basal cell carcinoma (BCC), squamous cell carcinoma (SCC) and intraepidermal carcinoma (IEC). These non-melanoma skin cancers constitute over 90% of all skin cancer diagnoses and hence this dataset gives an opportunity to the scientific community to benchmark analytic methodologies on a significant portion of the dermatopathology workflow. The data represents typical cases of the three cancer types (not requiring a differential diagnosis) across shave, punch and excision biopsy contexts. Each image is accompanied with a segmentation mask which characterizes the section into 12 tissue types, specifically: keratin, epidermis, papillary dermis, reticular dermis, hypodermis, inflammation, glands, hair follicles and background, as well as BCC, SCC and IEC. Included also are cancer margin measurements to work towards automated assessment of surgical margin clearance and tumour invasion. This leaves open many opportunities for researchers to utilize or extend the dataset, building upon recent work on image analysis problems in skin cancer (Thomas et al., 2021).

LLAMA: a robust and scalable machine learning pipeline for analysis of large scale 4D microscopy data: analysis of cell ruffles and filopodia.

BACKGROUND: With recent advances in microscopy, recordings of cell behaviour can result in terabyte-size datasets. The lattice light sheet microscope (LLSM) images cells at high speed and high 3D resolution, accumulating data at 100 frames/second over hours, presenting a major challenge for interrogating these datasets. The surfaces of vertebrate cells can rapidly deform to create projections that interact with the microenvironment. Such surface projections include spike-like filopodia and wave-like ruffles on the surface of macrophages as they engage in immune surveillance. LLSM imaging has provided new insights into the complex surface behaviours of immune cells, including revealing new types of ruffles. However, full use of these data requires systematic and quantitative analysis of thousands of projections over hundreds of time steps, and an effective system for analysis of individual structures at this scale requires efficient and robust methods with minimal user intervention. RESULTS: We present LLAMA, a platform to enable systematic analysis of terabyte-scale 4D microscopy datasets. We use a machine learning method for semantic segmentation, followed by a robust and configurable object separation and tracking algorithm, generating detailed object level statistics. Our system is designed to run on high-performance computing to achieve high throughput, with outputs suitable for visualisation and statistical analysis. Advanced visualisation is a key element of LLAMA: we provide a specialised tool which supports interactive quality control, optimisation, and output visualisation processes to complement the processing pipeline. LLAMA is demonstrated in an analysis of macrophage surface projections, in which it is used to i) discriminate ruffles induced by lipopolysaccharide (LPS) and macrophage colony stimulating factor (CSF-1) and ii) determine the autonomy of ruffle morphologies. CONCLUSIONS: LLAMA provides an effective open source tool for running a cell microscopy analysis pipeline based on semantic segmentation, object analysis and tracking. Detailed numerical and visual outputs enable effective statistical analysis, identifying distinct patterns of increased activity under the two interventions considered in our example analysis. Our system provides the capacity to screen large datasets for specific structural configurations. LLAMA identified distinct features of LPS and CSF-1 induced ruffles and it identified a continuity of behaviour between tent pole ruffling, wave-like ruffling and filopodia deployment.

Interpretable deep learning systems for multi-class segmentation and classification of non-melanoma skin cancer.

We apply for the first-time interpretable deep learning methods simultaneously to the most common skin cancers (basal cell carcinoma, squamous cell carcinoma and intraepidermal carcinoma) in a histological setting. As these three cancer types constitute more than 90% of diagnoses, we demonstrate that the majority of dermatopathology work is amenable to automatic machine analysis. A major feature of this work is characterising the tissue by classifying it into 12 meaningful dermatological classes, including hair follicles, sweat glands as well as identifying the well-defined stratified layers of the skin. These provide highly interpretable outputs as the network is trained to represent the problem domain in the same way a pathologist would. While this enables a high accuracy of whole image classification (93.6-97.9%), by characterising the full context of the tissue we can also work towards performing routine pathologist tasks, for instance, orientating sections and automatically assessing and measuring surgical margins. This work seeks to inform ways in which future computer aided diagnosis systems could be applied usefully in a clinical setting with human interpretable outcomes.

Branching morphogenesis in the developing kidney is not impacted by nephron formation or integration.

Branching morphogenesis of the ureteric bud is integral to kidney development; establishing the collecting ducts of the adult organ and driving organ expansion via peripheral interactions with nephron progenitor cells. A recent study suggested that termination of tip branching within the developing kidney involved stochastic exhaustion in response to nephron formation, with such a termination event representing a unifying developmental process evident in many organs. To examine this possibility, we have profiled the impact of nephron formation and maturation on elaboration of the ureteric bud during mouse kidney development. We find a distinct absence of random branch termination events within the kidney or evidence that nephrogenesis impacts the branching program or cell proliferation in either tip or progenitor cell niches. Instead, organogenesis proceeds in a manner indifferent to the development of these structures. Hence, stochastic cessation of branching is not a unifying developmental feature in all branching organs.

Branching morphogenesis in the developing kidney is governed by rules that pattern the ureteric tree.

Metanephric kidney development is orchestrated by the iterative branching morphogenesis of the ureteric bud. We describe an underlying patterning associated with the ramification of this structure and show that this pattern is conserved between developing kidneys, in different parts of the organ and across developmental time. This regularity is associated with a highly reproducible branching asymmetry that is consistent with locally operative growth mechanisms. We then develop a class of tip state models to represent elaboration of the ureteric tree and describe rules for 'half-delay' branching morphogenesis that describe almost perfectly the patterning of this structure. Spatial analysis suggests that the observed asymmetry may arise from mutual suppression of bifurcation, but not extension, between the growing ureteric tips, and demonstrates that disruption of patterning occurs in mouse mutants in which the distribution of tips on the surface of the kidney is altered. These findings demonstrate that kidney development occurs by way of a highly conserved reiterative pattern of asymmetric bifurcation that is governed by intrinsic and locally operative mechanisms.

Self-organisation after embryonic kidney dissociation is driven via selective adhesion of ureteric epithelial cells.

Human pluripotent stem cells, after directed differentiation in vitro, can spontaneously generate complex tissues via self-organisation of the component cells. Self-organisation can also reform embryonic organ structure after tissue disruption. It has previously been demonstrated that dissociated embryonic kidneys can recreate component epithelial and mesenchymal relationships sufficient to allow continued kidney morphogenesis. Here, we investigate the timing and underlying mechanisms driving self-organisation after dissociation of the embryonic kidney using time-lapse imaging, high-resolution confocal analyses and mathematical modelling. Organotypic self-organisation sufficient for nephron initiation was observed within a 24 h period. This involved cell movement, with structure emerging after the clustering of ureteric epithelial cells, a process consistent with models of random cell movement with preferential cell adhesion. Ureteric epithelialisation rapidly followed the formation of ureteric cell clusters with the reformation of nephron-forming niches representing a later event. Disruption of P-cadherin interactions was seen to impair this ureteric epithelial cell clustering without affecting epithelial maturation. This understanding could facilitate improved regulation of patterning within organoids and facilitate kidney engineering approaches guided by cell-cell self-organisation.

Analysed cap mesenchyme track data from live imaging of mouse kidney development.

This article provides detailed information on manually tracked cap mesenchyme cells from timelapse imaging of multiple ex vivo embryonic mouse kidneys. Cells were imaged for up to 18 h at 15 or 20 min intervals, and multiple cell divisions were tracked. Positional data is supplemented with a range of information including the relative location of the closest ureteric tip and a correction for drift due to bulk movement and tip growth. A subset of tracks were annotated to indicate the presence of processes attached to the ureteric epithelium. The calculations used for drift correction are described, as are the main methods used in the analysis of this data for the purpose of describing cap cell motility. The outcomes of this analysis are discussed in "Cap mesenchyme cell swarming during kidney development is influenced by attraction, repulsion, and adhesion to the ureteric tip" (A.N. Combes, J.G. Lefevre, S. Wilson, N.A. Hamilton, M.H. Little, 2016) [1].

Cap mesenchyme cell swarming during kidney development is influenced by attraction, repulsion, and adhesion to the ureteric tip.

Morphogenesis of the mammalian kidney requires reciprocal interactions between two cellular domains at the periphery of the developing organ: the tips of the epithelial ureteric tree and adjacent regions of cap mesenchyme. While the presence of the cap mesenchyme is essential for ureteric branching, how it is specifically maintained at the tips is unclear. Using ex vivo timelapse imaging we show that cells of the cap mesenchyme are highly motile. Individual cap mesenchyme cells move within and between cap domains. They also attach and detach from the ureteric tip across time. Timelapse tracks collected for >800 cells showed evidence that this movement was largely stochastic, with cell autonomous migration influenced by opposing attractive, repulsive and cell adhesion cues. The resulting swarming behaviour maintains a distinct cap mesenchyme domain while facilitating dynamic remodelling in response to underlying changes in the tip.

Cortical F-actin stabilization generates apical-lateral patterns of junctional contractility that integrate cells into epithelia.

E-cadherin cell-cell junctions couple the contractile cortices of epithelial cells together, generating tension within junctions that influences tissue organization. Although junctional tension is commonly studied at the apical zonula adherens, we now report that E-cadherin adhesions induce the contractile actomyosin cortex throughout the apical-lateral axis of junctions. However, cells establish distinct regions of contractile activity even within individual contacts, producing high tension at the zonula adherens but substantially lower tension elsewhere. We demonstrate that N-WASP (also known as WASL) enhances apical junctional tension by stabilizing local F-actin networks, which otherwise undergo stress-induced turnover. Further, we find that cells are extruded from monolayers when this pattern of intra-junctional contractility is disturbed, either when N-WASP redistributes into lateral junctions in H-Ras(V12)-expressing cells or on mosaic redistribution of active N-WASP itself. We propose that local control of actin filament stability regulates the landscape of intra-junctional contractility to determine whether or not cells integrate into epithelial populations.

Independent contrasts and PGLS regression estimators are equivalent.

We prove that the slope parameter of the ordinary least squares regression of phylogenetically independent contrasts (PICs) conducted through the origin is identical to the slope parameter of the method of generalized least squares (GLSs) regression under a Brownian motion model of evolution. This equivalence has several implications: 1. Understanding the structure of the linear model for GLS regression provides insight into when and why phylogeny is important in comparative studies. 2. The limitations of the PIC regression analysis are the same as the limitations of the GLS model. In particular, phylogenetic covariance applies only to the response variable in the regression and the explanatory variable should be regarded as fixed. Calculation of PICs for explanatory variables should be treated as a mathematical idiosyncrasy of the PIC regression algorithm. 3. Since the GLS estimator is the best linear unbiased estimator (BLUE), the slope parameter estimated using PICs is also BLUE. 4. If the slope is estimated using different branch lengths for the explanatory and response variables in the PIC algorithm, the estimator is no longer the BLUE, so this is not recommended. Finally, we discuss whether or not and how to accommodate phylogenetic covariance in regression analyses, particularly in relation to the problem of phylogenetic uncertainty. This discussion is from both frequentist and Bayesian perspectives.