Biology subjects

Elliott, L.

Publications and source records attributed to Elliott, L..

12 recordsLinked to original sources

bioRxiv2026

The Muskox Way: Humans, plants and animals in northeast Greenland through the Holocene

The Muskox Way hypothesis suggests that the first humans entering northeast Greenland from Arctic Canada followed and hunted muskox. Yet whether muskox arrived before people remains debated, as only two bones predate the Independence culture (4500-3900 cal BP). More broadly, the arrival of plants and animals, and how the three periods of human settlement related to ecosystem changes, remain poorly understood. We analysed sedimentary ancient DNA from lakes in Wandel Dal (Late Sommersoe and Midsommersoe) and Zackenberg, identifying 121 native plants and nine animal taxa. In Wandel Dal, the plant assemblage indicates a warmer climate than today about 8200-3900 years ago, followed by an abrupt change in flora and sedimentation rate at 3900 cal BP, especially at Midsommersoe. Muskox DNA is present continuously at Midsommersoe from 7400 cal BP, about three millennia before humans arrived. The Zackenberg record dates to 12,200 cal BP, with arctic forbs, grasses, ferns and crowberry in the earliest layers, willow from 8200 cal BP and birch from 7200 cal BP, but muskox only in the last 2500 years, the latter period coinciding with Greenlandic Dorset culture. The arrival of the Thule culture in 1400 CE coincides with a slight increase in plant richness at Wandel Dal and an increase in mammal detections. Reindeer occurred only in samples older than 6000 years, consistent with their absence from archaeological bone records. Overall, humans arrived long after the first megafauna and appear to have had limited impact on their populations, although cultural periods coincided with environmental changes.

ecology↗

bioRxiv2026

Diapause and Developmental Arrest as Drivers of Population Resilience Through Stress

Dauer formation and L1 arrest are stress-responsive developmental strategies that enable Caenorhabditis elegans to survive unfavorable conditions. These responses are regulated by environmental cues, including food availability and pheromone signals that communicate population density. However, how dauer entry, L1 arrest, and density-dependent signaling collectively influence long-term population dynamics remains poorly understood. In this study, populations of daf-22 and daf-16 mutants with impaired dauer formation, starvation arrest, and pheromone signaling were compared under control and starvation stress. Measurements of developmental stages were used to evaluate how genotype influenced population growth, developmental stage composition, starvation response, and recovery over time. This population-level approach links individual developmental decisions and inter-organismal communication to broader patterns of persistence and population change. The results show that daf-16 and daf-22 mutant populations differed from wild type in their recovery ability following nutrient deprivation as well as in the stage distributions within the population. These findings suggest that dauer signaling contributes broadly to population persistence by coordinating developmental arrest, reproduction, survival, and recovery. Overall, this work supports the interpretation of dauer formation as a larger population level survival program rather than a single isolated developmental outcome.

genetics↗

bioRxiv2026

Few-Shot Classification of C. elegans Developmental Stages via Explainable Hierarchical Hyperbolic Graph Embeddings

Automated, accurate, and fast developmental-stage classification of C. elegans from microscopy-based morphological images is essential for aging research, drug screening, and disease modeling. However, it remains challenging due to morphological similarities between stages and the limited annotated data. In this work, we propose HyperDev, a hyperbolic few-shot learning framework that addresses these limitations by directly encoding developmental hierarchies in the embedding space, unlike conventional Euclidean approaches that treat stages as independent classes. HyperDev uses Poincare ball geometry, combined with a biologically informed developmental prior, to naturally represent stage relationships. We introduce our self-curated C. elegans dataset spanning seven developmental stages (Egg, L1-L4, Adult, Dauer) with extreme class imbalance (6-8 samples per minority class). HyperDev achieves competitive classification accuracy (76.9-88.3%) while providing intrinsic explainability across nine 7-way few-shot evaluation settings. The learned embeddings exhibited strong biological alignment (Pearson r = 0.669, p < 0.001), while significantly outperforming ProtoNet (r = 0.187), MatchingNet (r = 0.235), and RelationNet (r = 0.464). These results establish hyperbolic geometry as a principled approach to explainable few-shot learning in biological imaging, where understanding learned representations is as critical as predictive performance. Clinical RelevanceBy enabling explainable, data-efficient developmental staging from scarce samples, HyperDev supports improved phenotype quantification for aging research, disease modeling, and drug screening.

bioinformatics↗

bioRxiv2026

DPI-score: A deep learning-based metric for assessing protein-protein interfaces in cryo-EM derived assemblies

Advances in cryoEM have led to a surge in high-resolution structures; however, model building and refinement at resolutions (>=3 [A]) and in regions with variable local resolution remain challenging, making validation essential. CryoEM derived assemblies often contain extensive protein-protein interfaces, however, most existing validation metrics focus on density fit or overall geometry without directly assessing interface quality. To address this, we present DPI-Score, a deep learning-based metric for assessing protein-protein interfaces in cryoEM derived complexes. The method uses only raw structural coordinates of interface atoms, without requiring engineered features, and achieves 87.53% validation accuracy. DPI-Score was applied to 29,120 interfaces from 6,011 fitted entries with resolutions worse than 3 [A] in the Electron Microscopy Data Bank. Here, we show that DPI-Score provides complementary information to existing validation metrics and can identify interface errors in modelled assemblies that are not detected by density-based scores alone.

bioinformatics↗

bioRxiv2024

A Rab-Kinesin12-Fused kinase module couples vesicle delivery and phragmoplast remodelling during cytokinesis in Arabidopsis

Cytokinesis is a key process in the development of multicellular organisms, through both formative and proliferative divisions. In land plants, successful cytokinesis requires precise targeting of Golgi-derived vesicles to the future division plane by the phragmoplast, a cytoskeletal structure that undergoes continuous remodelling. Endomembrane trafficking and phragmoplast remodelling are tightly coupled, indicating the existence of active crosstalk. However, although many molecular regulators of membrane trafficking and cytoskeletal organisation are known, it remains poorly understood how these processes are co-ordinated. Here, we describe a regulatory module consisting of the membrane-associated GTPase RAB-A2a, cytoskeleton-associated Class II Kinesin-12s, and the Fused kinase orthologue TIO. We provide evidence that the interaction between these molecules at the midzone is essential for cytokinesis in Arabidopsis through simultaneously targeting vesicles to the midzone and coupling phragmoplast remodelling to vesicle delivery.

plant biology↗

bioRxiv2023

CASP15 cryoEM protein and RNA targets: refinement and analysis using experimental maps

CASP assessments primarily rely on comparing predicted coordinates with experimental reference structures. However, errors in the reference structures can potentially reduce the accuracy of the assessment. This issue is particularly prominent in cryoEM-determined structures, and therefore, in the assessment of CASP15 cryoEM targets, we directly utilized density maps to evaluate the predictions. A method for ranking the quality of protein chain predictions based on rigid fitting to experimental density was found to correlate well with the CASP assessment scores. Overall, the evaluation against the density map indicated that the models are of high accuracy although local assessment of predicted side chains in a 1.52 [A] resolution map showed that side-chains are sometimes poorly positioned. The top 136 predictions associated with 9 protein target reference structures were selected for refinement, in addition to the top 40 predictions for 11 RNA targets. To this end, we have developed an automated hierarchical refinement pipeline in cryoEM maps. For both proteins and RNA, the refinement of CASP15 predictions resulted in structures that are close to the reference target structure, including some regions with better fit to the density. This refinement was successful despite large conformational changes and secondary structure element movements often being required, suggesting that predictions from CASP-assessed methods could serve as a good starting point for building atomic models in cryoEM maps for both proteins and RNA. Loop modeling continued to pose a challenge for predictors with even short loops failing to be accurately modeled or refined at times. The lack of consensus amongst models suggests that modeling holds the potential for identifying more flexible regions within the structure.

bioinformatics↗

bioRxiv2022

Edge-based growth control in Arabidopsis involves two cell wall-associated Receptor-Like Proteins

Morphogenesis of multicellular organs requires coordination of cellular growth. In plants, 3D growth is driven by undirected turgor pressure, whereas growth directionality is controlled by cell wall mechanical properties at 2D cell faces. Their shared cell wall also fixes cells in their position, and plants thus have to integrate tissue-scale mechanical stresses arising due to growth in a fixed tissue topology. This implies a need to monitor cell wall mechanical and biochemical status and to adapt growth accordingly. Here, we propose that plant cells use their 1D cell edges to monitor cell wall status. We describe two Receptor-Like Proteins, RLP4 and RLP4-L1, which occupy a unique polarity domain at cell edges established through a targeted secretory transport pathway. We show that at cell edges, RLP4s associate with the cell wall via their extracellular domain, and contribute to directional growth control in Arabidopsis.

plant biology↗

bioRxiv2022

Red harvester ant (Order: Hymenoptera) preference for cover crop seeds in South Texas

Harvester ants are known to selectively forage seeds, potentially impacting nearby plant community composition. In agricultural areas, harvester ants may be viewed as pests by foraging on crop seeds or as beneficials by preferentially removing weed seeds. However, little work has been done on harvester ant preferences for cover crop seeds. Local observations suggest that ants may take cover crop seeds, but no studies have evaluated ant agricultural impacts or seed preferences in the Lower Rio Grande Valley (LRGV). We examined red harvester ant (Pogonomyrmex barbatus Smith) preferences for commonly used cover crop seeds in the LRGV (vetch, oat, fescue, sunn hemp, and radish with wheatgrass as a control) and a commonly used bacterial seed inoculation treatment meant to increase root nodulation. We tested seed sets using choice tests housed in seed depots located within the foraging range of ant colonies with no prior exposure to the selected seeds. Of the evaluated cover crop seeds, wheatgrass and oat were the first to be removed entirely from the depot, with vetch remaining after 24 h. When we inoculated the two most preferred seeds to determine if there was a preference for non-inoculated seeds, we found no difference between inoculated and non-inoculated seeds. There were also significant changes in activity over time for both trials. These data indicate that harvester ant foraging preferences and activity can inform grower management recommendations regarding the risks of using certain cover crops and months sowing should be conducted in fields with known harvester ant presence.

ecology↗

bioRxiv2019

Proteomic studies of the Arabidopsis TRAPP complexes reveal conserved organization and a novel plant-specific component with a role in plant development

How the membrane trafficking system spatially organizes intracellular activities and intercellular signaling networks is not well understood in plants. The Transport Protein Particle (TRAPP) complexes are known to play key roles in selective delivery of membrane vesicles to various subcellular compartments in yeast and animals, but remain to be fully characterized in plants. Here we interrogate the TRAPP complexes in Arabidopsis using quantitative proteomic approaches. TRS33 is a component shared by all TRAPP complexes in yeast and animals, and the Arabidopsis AtTRS33 is essential for the subcellular dynamics of other TRAPP components. Affinity purification of AtTRS33 followed by quantitative mass spectrometry identified fourteen interacting proteins; these include not only thirteen homologs of all known TRAPP components in yeast and mammals but also a novel protein we named TRAPP-interacting plant protein (TRIPP), which is conserved in multi-cellular photosynthetic organisms. Proteomic and molecular analyses showed that TRIPP specifically associates with the TRAPPII complex in vivo and directly interacts with the TRAPPII-specific subunits but not the subunits shared with TRAPPIII. TRIPP co-localizes with a subset of TRS33 compartments, and its localization is disrupted in the trs33 mutant. Loss-of-function tripp mutation caused growth and reproductive development defects, including partial photomorphogenesis in the dark. Our study demonstrates that plants possess at least two distinct TRAPP complexes similar to metazoan, and identifies TRIPP as a novel plant-specific component of the TRAPPII complex with important functions in plant growth and development.

plant biology↗

bioRxiv2018

Analysis of head and neck carcinoma progression reveals novel and relevant stage-specific genetic changes associated with immortalisation and malignancy.

Head and neck squamous cell carcinoma (HNSCC) is a widely prevalent cancer globally with high mortality and morbidity. We report here changes in the genomic landscape in the development of HNSCC from potentially premalignant lesions (PPOLS) to malignancy and lymph node metastases. Frequent likely pathological mutations are restricted to a relatively small set of genes including TP53, CDKN2A, FBXW7, FAT1, NOTCH1 and KMT2D; these arise early in tumour progression and are present in PPOLs with NOTCH1 mutations restricted to cell lines from lesions that subsequently progressed to HNSCC. The most frequent genetic changes are of consistent somatic copy number alterations (SCNA). The earliest SCNAs involved deletions of CSMD1 (8p23.2), FHIT (3p14.2) and CDKN2A (9p21.3) together with gains of chromosome 20. CSMD1 deletions or promoter hypermethylation were present in all of the immortal PPOLs and occurred at high frequency in the immortal HNSCC cell lines (promoter hypermethylation ~63%, hemizygous deletions ~75%, homozygous deletions ~18%). Forced expression of CSMD1 in the HNSCC cell line H103 showed significant suppression of proliferation (p=0.0053) and invasion in vitro (p=5.98X10-5) supporting a role for CSMD1 inactivation in early head and neck carcinogenesis. In addition, knockdown of CSMD1 in the CSMD1-expressing BICR16 cell line showed significant stimulation of invasion in vitro (p=1.82 x 10-5) but not cell proliferation (p=0.239). HNSCC with and without nodal metastases showed some clear differences including high copy number gains of CCND1, hsa-miR-548k and TP63 in the metastases group. GISTIC peak SCNA regions showed significant enrichment (adj P<0.01) of genes in multiple KEGG cancer pathways at all stages with disruption of an increasing number of these involved in the progression to lymph node metastases. Sixty-seven genes from regions with statistically significant differences in SCNA/LOH frequency between immortal PPOL and HNSCC cell lines showed correlation with expression including 5 known cancer drivers.\n\nLay SummaryCancers affecting the head and neck region are relatively common. A large percentage of these are of one particular type; these are generally detected late and are associated with poor prognosis. Early detection and treatment dramatically improve survival and reduces the damage associated with the cancer and its treatment. Cancers arise and progress because of changes in the genetic material of the cells. This study focused on identifying such changes in these cancers particularly in the early stages of development, which are not fully known. Identification of these changes is important in developing new treatments as well as markers of behaviour of cancers and also the early or premalignant lesions. We used a well-characterised panel of cell lines generated from premalignant lesions as well as cancers, to identify mutations in genes, and an increase or decrease in number of copies of genes. We mapped new and previously identified changes in these cancers to specific stages in the development of these cancers and their spread. We additionally report here for the first time, alterations in CSMD1 gene in early premalignant lesions; we further show that this is likely to result in increased ability of the cells to spread and possibly, multiply faster as well.

cancer biology↗

bioRxiv2017

Directed functional connectivity using dynamic graphical models

There are a growing number of neuroimaging methods that model spatio-temporal patterns of brain activity to allow more meaningful characterizations of brain networks. This paper proposes dynamic graphical models (DGMs) for dynamic, directed functional connectivity. DGMs are a multivariate graphical model with time-varying coefficients that describe instantaneous directed relationships between nodes. A further benefit of DGMs is that networks may contain loops and that large networks can be estimated. We use network simulations, human resting-state fMRI (N = 500) to investigate the validity and reliability of the estimated networks. We simulate systematic lags of the hemodynamic response at different brain regions to investigate how these lags potentially bias directionality estimates. In the presence of such lag confounds (0.4-0.8 seconds offset between connected nodes), our method has a sensitivity of 72%-77% to detect the true direction. Stronger lag confounds have reduced sensitivity, but do not increase false positives (i.e., directionality estimates of the opposite direction). In human resting-state fMRI, we find the DMN has consistent influence on the cerebellar, the limbic and the auditory/temporal network, as well a consistent reciprocal relationship between the visual medial and visual lateral network. Finally, we apply the method in a small mouse fMRI sample and discover a highly plausible relationship between areas in the hippocampus feeding into the cingulate cortex. We provide a computationally efficient implementation of DGM as a free software package for R.

neuroscience↗

bioRxiv2017

The genetic basis of human brain structure and function: 1,262 genome-wide associations found from 3,144 GWAS of multimodal brain imaging phenotypes from 9,707 UK Biobank participants

The genetic basis of brain structure and function is largely unknown. We carried out genome-wide association studies of 3,144 distinct functional and structural brain imaging derived phenotypes in UK Biobank (discovery dataset 8,428 subjects). We show that many of these phenotypes are heritable. We identify 148 clusters of SNP-imaging associations with lead SNPs that replicate at p<0.05, when we would expect 21 to replicate by chance. Notable significant and interpretable associations include: iron transport and storage genes, related to changes in T2* in subcortical regions; extracellular matrix and the epidermal growth factor genes, associated with white matter micro-structure and lesion volume; genes regulating mid-line axon guidance development associated with pontine crossing tract organisation; and overall 17 genes involved in development, pathway signalling and plasticity. Our results provide new insight into the genetic architecture of the brain with relevance to complex neurological and psychiatric disorders, as well as brain development and aging. The full set of results is available on the interactive Oxford Brain Imaging Genetics (BIG) web browser.

genetics↗