Biology subjects

Potter, C.

Publications and source records attributed to Potter, C..

12 recordsLinked to original sources

bioRxiv2026

Non-invasive quantification of hepatic necrosis from circulating alanine aminotransferase kinetics in acetaminophen-treated mice in vivo

Histological necrosis is the reference measure of hepatotoxicity but can only be obtained at terminal cull. A time course requires independent cohorts at every timepoint. Circulating alanine aminotransferase (ALT) can be sampled repeatedly in the same animal, and is released in liver injury. We asked whether serial ALT kinetics can be used to estimate histological necrosis in vivo. Forty-five fasted twelve-week-old male C57BL/6J mice received a single intraperitoneal dose of 350 mg/kg acetaminophen (paracetamol). Plasma ALT and microRNA-122 (miR-122) were measured serially from baseline to 48h as cumulative area-under-the-curve (AUC) to cull, against centrilobular necrosis on haematoxylin and eosin sections as reference (range 0 to 59%, mean 32%). A generalised additive model of necrosis on cumulative ALT AUC and time since dosing predicted necrosis with a leave-one-animal-out cross-validated mean absolute error of 7.7% necrotic area (95% CI 5.6 to 10.0; n=49). A panel of traditional regression and machine-learning models all gave equal or larger error, and adding an additional biomarker or regeneration information did not improve prediction. Serial ALT kinetics therefore provide a calibrated, longitudinal measure of hepatic necrosis in vivo and give a more stable estimate of within-group variance for study planning, while supporting reductions in animal use, because one serially-sampled cohort can replace separate cohorts at each timepoint.

pharmacology and toxicology↗

bioRxiv2026

A novel precision-cut lung slice stretch model using removable inflation materials

Stretch is an important biomechanical stimulus facilitating tissue development in the respiratory system by programming the epithelium, endothelium, and extracellular matrix (ECM). Lung tissue undergoes stretch induced lung differentiation under normal prenatal and postnatal development. Furthermore, supraphysiological and aberrant stretch responses are known mechanisms of acute lung injury and ECM disruption. Current in vitro human tissue cyclic mechanical stretch (CMS) models suffer from significant, well-recognized disadvantages and are poorly validated in vivo for longer-term study. In vitro precision-cut lung slice (PCLS) models are commonly used to study the complex structural arrangement and cellular interactions of human tissue, as well as various lung diseases, including BPD.3 PCLS maintain lung tissue architecture and the variety of cell types present in the lung, allowing for a more realistic imitation of the lung microenvironment.3 Existing agarose-inflated PCLS models are hindered by retention of agarose media in the tissue, affecting material properties and complicating stretch studies. Our novel PCLS approach utilizes several technical innovations including a removable hydrogel for inflation and uses supportive poly(ethylene glycol) (PEG) hydrogels enable improved viability and phenotype retention during cyclic mechanical stretch (CMS). This platform will induce PCLS CMS for biochemical assays (e.g. transcriptomics, proteomics) after exposure.

bioengineering↗

bioRxiv2025

Predicting Organ-Specific Toxicity of Selective Androgen Receptor Modulators, using Transfer Learning on Graph Convolutional Networks

Novel Quantitative Structure-Activity Relationship (QSAR) models were constructed using Graph Convolutional Networks (GCNs), to predict Drug-Induced Liver Injury (DILI), Drug-Induced Renal Injury (DIRI) and Drug-Induced Cardiotoxicity (DICT) of Selective Androgen Receptor Modulators (SARMs) - an emerging class of performance-enhancing drugs. Prior to training on DILI, DIRI and DICT datasets, the GCN QSAR models were first pre-trained on a variety of unrelated biomedical assay datasets, as an attempt to improve model performance via transfer learning. The success of the transfer learning was mixed; model performances were measurably improved via pre-training on certain datasets, by statistically weak increases. The optimal final QSAR models achieved overall accuracy scores of 68% for DILI (no significant improvement via ensemble modelling), 76% for DIRI (improved to 77% via ensemble modelling) and 65% for DICT (improved to 67% via ensemble modelling). Application of the most optimal singular models to a dataset of 25 SARMs predicted that 21 of the 25 SARMs are either DILI-positive, DIRI-positive, or both - which raises concern, given the rising use of SARMs. All SARMs except for one were predicted as DICT-negative. A novel definition of the Applicability Domain (AD) was used, intended for close relevance to the models, via generating three-dimensional graph embeddings, for each model. Convex hulls were fitted around training data embeddings, with a {+/-}10% buffer, defining the AD as the region of embedded chemical space covered by the convex hull, for a given model. Subsequent analysis found that a majority of DILI, DIRI and DICT testing data lay within the AD, alongside a majority of the SARMs - adding consensus to the reliability of the predictions.

pharmacology and toxicology↗

bioRxiv2025

Machine learning model determines optimal coast redwood restoration sites in Santa Clara County, California

Many governmental and non-governmental organizations are planting trees, often in developing countries where costs are lower, to offset carbon emissions in industrial countries and slow global warming. These efforts often fail to achieve carbon sequestration goals, frequently related to the selection of unsuitable planting sites, use of tree species with inadequate biomass or vulnerability to wildfire. Reforestation of the worlds tallest trees, which attain heights above 80 m and standing biomasses orders of magnitude greater than other trees, could optimize removal of atmospheric carbon. Wildfires have consumed almost 30% of Californias forests since 2000. To optimize carbon sequestration by tree planting in Santa Clara County, California, we selected coast redwoods, a tree species that is highly resilient to fire, rapidly accumulates the largest biomass of any tree in the world, has exceptional longevity, and is historically native to the westernmost portion of the county, where they support a unique ecosystem. Because historical redwoods range maps are conflicting, and because global warming may change the range of suitable habitat, we developed a machine learning model to determine optimal habitat available for reforestation. The optimal habitat identified included the current and known historical coast redwood forest and excluded lands with no historical records of redwoods, validating model accuracy. The opportunity to capture and store atmospheric carbon dioxide (CO2) is significant, as the model found 33,969 ha (131 mi2) of optimal redwood habitat, of which nearly 77% [26,051 ha (101 mi2)] currently has no conifer cover. Restoring the historical coast redwood forest in this single county alone could sequester 2.3% of the entire State of Californias 2020 carbon emissions. Similar studies should be conducted in ocean-adjacent counties which likely have order of magnitude greater optimal coast redwood planting areas than inland Santa Clara County, with correspondingly greater potential carbon sequestration impacts.

ecology↗

bioRxiv2024

Annotating CryoET Volumes: A Machine Learning Challenge

Cryo-electron tomography (cryoET) has emerged as a powerful structural biology tool for understanding protein complexes in their native cellular environments. Presently, 3D volumes of cellular environments can be acquired in the thousands in a few days where each volume provides a rich and complex cellular landscape. Despite numerous innovations, localizing and identifying the vast majority of protein species in these volumes remains prohibitively difficult. Machine learning based methods provide an opportunity to automate the process of labeling and annotating cryoET volumes. Due to current bottlenecks in the annotation process, and a lack of large standardized datasets, training datasets for machine learning algorithms have been scarce. Here, we present a defined "phantom" sample, along with "ground truth" annotations, that will be the basis of a machine learning challenge to bring cryoET and ML experts together and spur creativity to address this annotation problem. We have also set up a cryoET data portal that provides additional diverse sets of annotated 3D volumes from cryoET experts across the world for the machine learning challenge.

cell biology↗

bioRxiv2024

Open-source Tools for CryoET Particle Picking Machine Learning Competitions

We are launching a machine learning (ML) competition focused on particle picking in cryo-electron tomography (cryoET) data, a crucial task in structural biology. To support this, we have created a comprehensive suite of open-source tools to develop resources for our competition, including copick for dataset management, napari plugins for interactive visualization, utilities for converting particle picks to segmentation masks, and PyTorch tools for custom dataset sampling. These resources streamline the processes of data handling, labeling, and visualization, allowing participants to focus on model development. By leveraging these tools, competitors will be better equipped to tackle the unique challenges of cryoET data and push forward advancements in particle picking techniques.

bioinformatics↗

bioRxiv2023

Mitochondrial protein FKBP8 captures PDZD8 to form mitochondria-ER contacts

Mitochondria-ER membrane contact sites (MERCS) represent a fundamental ultrastructural feature underlying unique biochemistry and physiology in eukaryotic cells. The ER protein PDZD8 is required for the formation of MERCS in many cell types, however, its tethering partner on the outer mitochondrial membrane (OMM) is currently unknown. Here we identified the OMM protein FKBP8 as the tethering partner of PDZD8 using a combination of unbiased proximity proteomics, CRISPR-Cas9 endogenous protein tagging, Cryo-Electron Microscopy (Cryo-EM) tomography, and correlative light-EM (CLEM). Single molecule tracking revealed highly dynamic diffusion properties of PDZD8 along the ER membrane with significant pauses and capture at MERCS. Overexpression of FKBP8 was sufficient to narrow the ER-OMM distance, whereas independent versus combined deletions of these two proteins demonstrated their interdependence for MERCS formation. Furthermore, PDZD8 enhances mitochondrial complexity in a FKBP8-dependent manner. Our results identify a novel ER-mitochondria tethering complex that regulates mitochondrial morphology in mammalian cells.

cell biology↗

bioRxiv2023

Microbiota epigenetically direct tuft cell differentiation to control type 2 immunity

Allergy and anti-helminth immunity are driven by type 2 responses in mucosal tissues. Tuft cells are key regulators of type 2 immunity, however the factors that control these cells remain poorly understood. Here we find that butyrate-producing commensal bacteria decrease tuft cells in the intestine. Butyrate suppression of tuft cells required the epigenetic modifying enzyme histone deacetylase 3 (HDAC3), suggesting that HDAC3 may promote tuft cell-dependent immunity. Consistent with this, epithelial-intrinsic HDAC3 actively regulated tuft cell expansion in vivo and was required to induce type 2 immune responses during helminth infection. Interestingly, butyrate epigenetically restricted stem cell differentiation into tuft cells, and inhibition of HDAC3 in adult mice and human intestinal organoids was sufficient to block tuft cell expansion. Collectively, these data reveal an epigenetic pathway in stem cells that directs tuft cell differentiation, and highlight a new level of regulation through which commensal bacteria calibrate intestinal immunity.

immunology↗

bioRxiv2018

The impact of data reduction and lossy image formats on electron microscope images

Recent advances in instrumentation and software for cryoEM have increased the applicability and utility of this method. Coupled with the adoption of automated pipelines, significant infrastructure support is required to sustain high throughput workflows. In particular, data generation rates may outpace the ability to deploy data storage and archival solutions. We have investigated what effects data compression and conversion to different file formats have on the ability to obtain high resolution cryoEM reconstructions. Standard lossless data compression strategies have a high impact on reducing the size of direct detector electron counting movie stacks, but provide more modest gains for aligned summed images. We show that EM images can be compressed using standard lossy methods to reduce file storage to 5-10% of the size of the original aligned sum or movie stack file and yet still retain enough information such that modern image processing pipelines will provide sub-2[A] reconstructions from the compressed data.

biophysics↗

bioRxiv2018

Temperate grass allergy season defined by spatio-temporal shifts in airborne pollen communities

Grass pollen is the worlds most harmful outdoor aeroallergen and sensitivity varies between species. Different species of grass flower at different times, but it is not known how airborne communities of grass pollen change in time and space. Persistence and high mobility of grass pollen could result in increasingly diverse seasonal pollen communities. Conversely, if grass pollen does not persist for an extended time in the air, shifting pollen communities would be predicted throughout the summer months. Here, using targeted high throughput sequencing, we tracked the seasonal progression of airborne Poaceae pollen biodiversity across Britain, throughout the grass allergy season. All grass genera displayed discrete, temporally restricted peaks of pollen incidence which varied with latitude, revealing that the taxonomic composition of grass pollen exposure changes substantially across the allergy season. By developing more refined aeroallergen profiling, we predict that our findings will facilitate the exploration of links between taxon-specific exposure of harmful grass pollen and disease, with concomitant socio-economic benefits.

ecology↗

bioRxiv2018

Benchmarking cryo-EM single particle analysis workflow

Cryo electron microscopy facilities running multiple instruments and serving users with varying skill levels need a robust and reliable method for benchmarking both the hardware and software components of their single particle analysis workflow. The workflow is complex, with many bottlenecks existing at the specimen preparation, data collection and image analysis steps; the samples and grid preparation can be of unpredictable quality, there are many different protocols for microscope and camera settings, and there is a myriad of software programs for analysis that can depend on dozens of settings chosen by the user. For this reason, we believe it is important to benchmark the entire workflow, using a standard sample and standard operating procedures, on a regular basis. This provides confidence that all aspects of the pipeline are capable of producing maps to high resolution. Here we describe benchmarking procedures using a test sample, rabbit muscle aldolase.

molecular biology↗

bioRxiv2017

A GAL80 collection to nullify transgenes in Drosophila olfactory sensory neurons

Fruit flies recognize hundreds of ecologically relevant odors and respond appropriately to them. The complexity, redundancy and interconnectedness of the olfactory machinery complicate efforts to pinpoint the functional contributions of any component neuron or receptor to behavior. Some contributions can only be elucidated in flies that carry multiple mutations and transgenes, but the production of such flies is currently labor-intensive and time-consuming. Here, we describe a set of transgenic flies that express the Saccharomyces cerevisiae GAL80 in specific olfactory sensory neurons (OrX-GAL80s). The GAL80s effectively and specifically subtract the activities of GAL4-driven transgenes that impart anatomical and physiological phenotypes. OrX-GAL80s can allow researchers to efficiently activate only one or a few types of functional neurons in an otherwise nonfunctional olfactory background. Such experiments will improve our understanding of the mechanistic connections between odorant inputs and behavioral outputs at the resolution of only a few functional neurons.

neuroscience↗