•The most common skin cancers (BCC, SCC and IEC) are amenable to machine learning methods.•Semantic segmentation allows a network to learn the full context of skin tissue types in an interpretable ...way.•High classification accuracy (>93%) can be achieve when classifying whole tissue sections.•Full-characterisation of the tissue allows for the automation of routine pathologist tasks such as surgical margin clearance assessments.
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.
Display omitted
RhoA stimulates cell contractility by recruiting downstream effectors to the cortical plasma membrane. We now show that direct binding by anillin is required for effective signaling: this antagonizes ...the otherwise labile membrane association of GTP-RhoA to promote effector recruitment. However, since its binding to RhoA blocks access by other effectors, we demonstrate that anillin must also concentrate membrane phosphoinositide-4,5-P2 (PIP2) to promote signaling. We propose and test a sequential pathway where GTP-RhoA first binds to anillin and then is retained at the membrane by PIP2 after it disengages from anillin. Importantly, re-binding of membrane GTP-RhoA to anillin, regulated by the cortical density of anillin, creates cycles through this pathway. These cycles repeatedly reset the dissociation kinetics of GTP-RhoA, substantially increasing its dwell time to recruit effectors. Thus, anillin regulates RhoA signaling by a paradigm of kinetic scaffolding that may apply to other signals whose efficacy depends on their cortical dwell times.
Display omitted
•Anillin enhances RhoA signaling by countering its labile cortical association•Anillin concentrates PI(4,5)P2 to retain membrane GTP-RhoA for effector recruitment•Re-binding of RhoA to anillin creates cycles that increase the dwell time of active RhoA•Kinetic scaffolding is sufficient for anillin to support cell contractility
Anillin promotes cell contractility by inhibiting the cortical dissociation of active RhoA. Budnar et al. explain this by a kinetic scaffolding model where anillin also concentrates membrane PI(4,5)P2. Cyclic binding and unbinding of RhoA to anillin allows free RhoA to be retained by PI(4,5)P2, increasing its dwell time for signaling.
Although kidneys of equal size can vary 10-fold in nephron number at birth, discovering what regulates such variation has been hampered by a lack of quantitative parameters defining kidney ...development. Here we report a comprehensive, quantitative, multiscale analysis of mammalian kidney development in which we measure changes in cell number, compartment volumes, and cellular dynamics across the entirety of organogenesis, focusing on two key nephrogenic progenitor populations: the ureteric epithelium and the cap mesenchyme. In doing so, we describe a discontinuous developmental program governed by dynamic changes in interactions between these key cellular populations occurring within a previously unappreciated structurally stereotypic organ architecture. We also illustrate the application of this approach to the detection of a subtle mutant phenotype. This baseline program of kidney morphogenesis provides a framework for assessing genetic and environmental developmental perturbation and will serve as a gold standard for the analysis of other organs.
•A comprehensive temporospatial and multiscale profile of normal kidney development•Analysis of structural stereotypy and branch form across development•Dynamic changes in the cap and tip progenitor cell niches over time•Application of the approach to identify a subtle mutant phenotype
Short et al. describe a comprehensive, quantitative analysis of shape, cell number, and proliferation rate for key progenitor populations across mammalian kidney development. This reveals a morphogenetic program with structural stereotypy, asynchronous branching, substantial remodeling, and discontinuous nephron induction, elucidating how variation in size and nephron number may arise.
We present Omni-ATAC, an improved ATAC-seq protocol for chromatin accessibility profiling that works across multiple applications with substantial improvement of signal-to-background ratio and ...information content. The Omni-ATAC protocol generates chromatin accessibility profiles from archival frozen tissue samples and 50-μm sections, revealing the activities of disease-associated DNA elements in distinct human brain structures. The Omni-ATAC protocol enables the interrogation of personal regulomes in tissue context and translational studies.
Delayed second dose SARS-CoV-2 vaccination trades maximal effectiveness for a lower level of immunity across more of the population. We investigated whether patients with inflammatory bowel disease ...treated with infliximab have attenuated serological responses to a single dose of a SARS-CoV-2 vaccine.
Antibody responses and seroconversion rates in infliximab-treated patients (n=865) were compared with a cohort treated with vedolizumab (n=428), a gut-selective anti-integrin α4β7 monoclonal antibody. Our primary outcome was anti-SARS-CoV-2 spike (S) antibody concentrations, measured using the Elecsys anti-SARS-CoV-2 spike (S) antibody assay 3-10 weeks after vaccination, in patients without evidence of prior infection. Secondary outcomes were seroconversion rates (defined by a cut-off of 15 U/mL), and antibody responses following past infection or a second dose of the BNT162b2 vaccine.
Geometric mean (SD) anti-SARS-CoV-2 antibody concentrations were lower in patients treated with infliximab than vedolizumab, following BNT162b2 (6.0 U/mL (5.9) vs 28.8 U/mL (5.4) p<0.0001) and ChAdOx1 nCoV-19 (4.7 U/mL (4.9)) vs 13.8 U/mL (5.9) p<0.0001) vaccines. In our multivariable models, antibody concentrations were lower in infliximab-treated compared with vedolizumab-treated patients who received the BNT162b2 (fold change (FC) 0.29 (95% CI 0.21 to 0.40), p<0.0001) and ChAdOx1 nCoV-19 (FC 0.39 (95% CI 0.30 to 0.51), p<0.0001) vaccines. In both models, age ≥60 years, immunomodulator use, Crohn's disease and smoking were associated with lower, while non-white ethnicity was associated with higher, anti-SARS-CoV-2 antibody concentrations. Seroconversion rates after a single dose of either vaccine were higher in patients with prior SARS-CoV-2 infection and after two doses of BNT162b2 vaccine.
Infliximab is associated with attenuated immunogenicity to a single dose of the BNT162b2 and ChAdOx1 nCoV-19 SARS-CoV-2 vaccines. Vaccination after SARS-CoV-2 infection, or a second dose of vaccine, led to seroconversion in most patients. Delayed second dosing should be avoided in patients treated with infliximab.
ISRCTN45176516.
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. 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. 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.
Yeasts, which have been a component of the human diet for at least 7,000 years, possess an elaborate cell wall α-mannan. The influence of yeast mannan on the ecology of the human microbiota is ...unknown. Here we show that yeast α-mannan is a viable food source for the Gram-negative bacterium Bacteroides thetaiotaomicron, a dominant member of the microbiota. Detailed biochemical analysis and targeted gene disruption studies support a model whereby limited cleavage of α-mannan on the surface generates large oligosaccharides that are subsequently depolymerized to mannose by the action of periplasmic enzymes. Co-culturing studies showed that metabolism of yeast mannan by B. thetaiotaomicron presents a 'selfish' model for the catabolism of this difficult to breakdown polysaccharide. Genomic comparison with B. thetaiotaomicron in conjunction with cell culture studies show that a cohort of highly successful members of the microbiota has evolved to consume sterically-restricted yeast glycans, an adaptation that may reflect the incorporation of eukaryotic microorganisms into the human diet.
A genetic mutation in the Vps35 subunit of retromer has recently been linked to late onset Parkinson's disease. We observed that the distribution and maturation of Vps35 D620N positive endosomes are ...altered. While Vps35 D620N containing retromer still binds CI‐M6PR, its trafficking is perturbed as shown by secretion of its ligand cathepsin D. As cathepsin D is involved in processing of α‐synuclein, a well‐established causative agent of Parkinson's disease, its altered trafficking may therefore represent the underlying cause of disease.
The retromer is a trimeric cargo‐recognition protein complex composed of Vps26, Vps29 and Vps35 associated with protein trafficking within endosomes. Recently, a pathogenic point mutation within the Vps35 subunit (D620N) was linked to the manifestation of Parkinson's disease (PD). Here, we investigated details underlying the molecular mechanism by which the D620N mutation in Vps35 modulates retromer function, including examination of retromer's subcellular localization and its capacity to sort cargo. We show that expression of the PD‐linked Vps35 D620N mutant redistributes retromer‐positive endosomes to a perinuclear subcellular localization and that these endosomes are enlarged in both model cell lines and fibroblasts isolated from a PD patient. Vps35 D620N is correctly folded and binds Vps29 and Vps26A with the same affinity as wild‐type Vps35. While PD‐linked point mutant Vps35 D620N interacts with the cation‐independent mannose‐6‐phosphate receptor (CI‐M6PR), a known retromer cargo, we find that its expression disrupts the trafficking of cathepsin D, a CI‐M6PR ligand and protease responsible for degradation of α‐synuclein, a causative agent of PD. In summary, we find that the expression of Vps35 D620N leads to endosomal alterations and trafficking defects that may partly explain its action in PD.
The genomic revolution has led to rapid growth in sequencing of genes and proteins, and attention is now turning to the function of the encoded proteins. In this respect, microscope imaging of a ...protein's sub-cellular localisation is proving invaluable, and recent advances in automated fluorescent microscopy allow protein localisations to be imaged in high throughput. Hence there is a need for large scale automated computational techniques to efficiently quantify, distinguish and classify sub-cellular images. While image statistics have proved highly successful in distinguishing localisation, commonly used measures suffer from being relatively slow to compute, and often require cells to be individually selected from experimental images, thus limiting both throughput and the range of potential applications. Here we introduce threshold adjacency statistics, the essence which is to threshold the image and to count the number of above threshold pixels with a given number of above threshold pixels adjacent. These novel measures are shown to distinguish and classify images of distinct sub-cellular localization with high speed and accuracy without image cropping.
Threshold adjacency statistics are applied to classification of protein sub-cellular localization images. They are tested on two image sets (available for download), one for which fluorescently tagged proteins are endogenously expressed in 10 sub-cellular locations, and another for which proteins are transfected into 11 locations. For each image set, a support vector machine was trained and tested. Classification accuracies of 94.4% and 86.6% are obtained on the endogenous and transfected sets, respectively. Threshold adjacency statistics are found to provide comparable or higher accuracy than other commonly used statistics while being an order of magnitude faster to calculate. Further, threshold adjacency statistics in combination with Haralick measures give accuracies of 98.2% and 93.2% on the endogenous and transfected sets, respectively.
Threshold adjacency statistics have the potential to greatly extend the scale and range of applications of image statistics in computational image analysis. They remove the need for cropping of individual cells from images, and are an order of magnitude faster to calculate than other commonly used statistics while providing comparable or better classification accuracy, both essential requirements for application to large-scale approaches.