Biology subjects

Hartley, B.

Publications and source records attributed to Hartley, B..

6 recordsLinked to original sources

bioRxiv2026

Breaking the bottleneck: self-supervised deep learning framework for fully automated fossil CT segmentation

Semantic segmentation of domain-specific imaging data where labelled training examples are scarce and foreground-background contrast is low remains an open challenge in deep learning applied to science. Palaeontological computed tomography (CT) exemplifies this problem: digitally isolating fossilised bone from surrounding rock matrix is labour-intensive ([≥]100 hrs/dataset), subjective, and often reliant on expensive proprietary software, creating a "segmentation bottleneck" that prevents large-scale and rapid processing of CT data collections. Here we present a self-supervised, end-to-end framework combining SimCLR v1 contrastive pretraining with deterministic pseudo-label generation and U-Net refinement to fully automate fossil CT segmentation without manual annotation. Using 50,626 CT images from the Middle Jurassic Kilmaluag Formation spanning amphibians, reptiles, dinosaurs, and early mammals, the framework achieved a Dice coefficient of 93.66% and IoU of 82.42% on a held-out specimen not seen during training, comparable to the highest Dice and IoU values reported in recent Deep Learning-based fossil CT segmentation studies. Cross-taxon generalisation was validated geometrically on six fully external specimens, achieving sub-voxel mesh agreement with manually thresholded references. By eliminating the annotation requirement that has limited prior deep learning approaches in palaeontology, this framework reduces per-specimen processing from [~]100 person-hours to 6 hrs (one-time UNet training) +1-3 minutes (mesh generation per specimen), an essential first step towards batch processing and analysis of CT data for large-scale comparative and quantitative analyses.

paleontology↗

bioRxiv2026

Matrix metalloproteinases proteolyze RAB proteins and contribute to cisplatin-induced ototoxicity

Matrix metalloproteinases (MMPs) are rapidly expressed and activated in response to oxidative stress and contribute to various pathological conditions. Cisplatin is a highly effective chemotherapeutic agent; however, its clinical use is limited by its associated permanent hearing loss (ototoxicity). While cispwlatin-induced oxidative stress and inner ear cell death are well-established, the contribution of MMPs remains unclear. In this study, we demonstrate that cisplatin exposure triggers activation of MMP-2 and MMP-9 and expression of an intracellular N-terminal-truncated isoform of MMP-2 in mouse inner ear hair cells. Pharmacological inhibition of MMP-2 and genetic knockdown of Mmp-9 enhanced hair cell survival and attenuated cisplatin-induced inflammation and cytotoxicity. Furthermore, proteomic analysis revealed that proteins involved in intracellular trafficking, including RAB proteins, may serve as potential substrates of intracellular MMP-2 upon cisplatin exposure, pointing to a previously unrecognized mechanism of cisplatin-induced hair cell injury. In vitro analysis confirmed that MMP-2 cleaves RAB9A in response to cisplatin, and in silico analyses predicted MMP-2-preferred cleavage sites on RAB9A. Collectively, our findings identify MMP-2 as a promising therapeutic target for mitigating cisplatin-induced ototoxicity.

pharmacology and toxicology↗

bioRxiv2025

Mechanistic insight into the regulation of virulence factor type 1 pili in pathogenic E. coli by rhomboid protease GlpG

With the rise in antimicrobial resistance, understanding the virulence factors utilized by pathogenic E. coli is essential for the development of alternative therapeutics. While previous work has shown that disruption of the E. coli rhomboid protease gene glpG leads to defects in bacterial colonization, here we provide mechanistic insight into the loss of fitness. We show GlpG is essential for the assembly of type 1 pili, a virulence factor required for the colonization of eukaryotic cells. Since pili are critical for biofilm formation and bacterial persistence, the absence of GlpG proteolytic activity reduces the production of biofilm. Working towards new potential antimicrobial targets for treating infections, we show that biofilm formation is hampered by GlpG inhibition. Our data demonstrates that GlpG plays a key role in protein quality control of type 1 pili and alters the paradigm for GlpG proteolysis, previously implicated in the cleavage of only membrane embedded substrates.

biochemistry↗

bioRxiv2024

A repository of Ogden syndrome (NAA10) and NAA15 patient derived iPSC lines and isogenic pairs by X-chromosome screening and genome-editing.

Amino-terminal (Nt-) acetylation (NTA) is a common protein modification, affecting 80% of cytosolic proteins in humans. The human essential gene, NAA10, encodes the enzyme NAA10, as the catalytic subunit for the N-terminal acetyltransferase A (NatA) complex, including the accessory protein, NAA15. The first human disease directly involving NAA10 was discovered in 2011, and it was named Ogden syndrome (OS), after the location of the first affected family residing in Ogden, Utah, USA. Since that time, other variants have been found in NAA10 and NAA15. Here we describe the generation of 31 iPSC lines, with 16 from females and 15 from males. This cohort includes CRISPR-mediated correction to the wild-type genotype in 4 male lines, along with editing one female line to generate homozygous wild-type or mutant clones. Following the monoclonalizaiton and screening for X-chromosome activation status in female lines, 3 additional pairs of female lines, in which either the wild type allele is on the active X chromosome (Xa) or the pathogenic variant allele is on Xa, have been generated. Subsets of this cohort have been successfully used to make cardiomyocytes and neural progenitor cells (NPCs). These cell lines are made available to the community via the NYSCF Repository.

genetics↗

bioRxiv2024

Deep Learning Analysis on Images of iPSC-derived Motor Neurons Carrying fALS-genetics Reveals Disease-Relevant Phenotypes

Amyotrophic lateral sclerosis (ALS) is a devastating condition with very limited treatment options. It is a heterogeneous disease with complex genetics and unclear etiology, making the discovery of disease-modifying interventions very challenging. To discover novel mechanisms underlying ALS, we leverage a unique platform that combines isogenic, induced pluripotent stem cell (iPSC)-derived models of disease-causing mutations with rich phenotyping via high-content imaging and deep learning models. We introduced eight mutations that cause familial ALS (fALS) into multiple donor iPSC lines, and differentiated them into motor neurons to create multiple isogenic pairs of healthy (wild-type) and sick (mutant) motor neurons. We collected extensive high-content imaging data and used machine learning (ML) to process the images, segment the cells, and learn phenotypes. Self-supervised ML was used to create a concise embedding that captured significant, ALS-relevant biological information in these images. We demonstrate that ML models trained on core cell morphology alone can accurately predict TDP-43 mislocalization, a known phenotypic feature related to ALS. In addition, we were able to impute RNA expression from these image embeddings, in a way that elucidates molecular differences between mutants and wild-type cells. Finally, predictors leveraging these embeddings are able to distinguish between mutant and wild-type both within and across donors, defining cellular, ML-derived disease models for diverse fALS mutations. These disease models are the foundation for a novel screening approach to discover disease-modifying targets for familial ALS.

neuroscience↗

bioRxiv2017

An Efficient Platform For Astrocyte Differentiation From Human Induced Pluripotent Stem Cells

Growing evidence implicates the importance of glia, particularly astrocytes, in neurological and psychiatric diseases. Here, we describe a rapid and robust method for the differentiation of highly pure populations of replicative astrocytes from human induced pluripotent stem cells (hiPSCs), via a neural progenitor cell (NPC) intermediate. Using this method, we generated hiPSC-derived astrocyte populations (hiPSC-astrocytes) from 42 NPC lines (derived from 30 individuals) with an average of [~]90% S100{beta}-positive cells. Transcriptomic analysis demonstrated that the hiPSC-astrocytes are highly similar to primary human fetal astrocytes and characteristic of a non-reactive state. hiPSC-astrocytes respond to inflammatory stimulants, display phagocytic capacity and enhance microglial phagocytosis. hiPSC-astrocytes also possess spontaneous calcium transient activity. Our novel protocol is a reproducible, straightforward (single media) and rapid (<30 days) method to generate homogenous populations of hiPSC-astrocytes that can be used for neuron-astrocyte and microglia-astrocyte co-cultures for the study of neuropsychiatric disorders.\n\nABBREVIATIONS

neuroscience↗