Biology subjects

Ascher, D.

Publications and source records attributed to Ascher, D..

13 recordsLinked to original sources

bioRxiv2026

Copine-6 integrates calcium, phosphoinositide and Rab11 signals to coordinate glutamate receptor recycling

Endosomal trafficking is a major pathway that delivers cell-surface proteins, including glutamate receptors, to support neurotransmission and normal brain functions. Activity-dependent insertion of glutamate receptors is essential for synaptic plasticity, learning and memory. Copine-6 is a neuronal-specific calcium (Ca2+) binding protein that mediates activity-induced exocytosis of -amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-type glutamate receptors. The activation of N-methyl-D-aspartate (NMDA) receptors triggers Ca2+-dependent translocation of Copine-6 to intracellular endosomal compartments. However, the mechanisms underlying the activity-dependent accumulation of Copine-6 in endosomes remain unknown. Here, we show that Copine-6 exhibits Ca2+-dependent binding to phosphatidylinositol-3-phosphate (PI(3)P) through the C2B domain and displays enhanced interaction with active Rab11a in a Ca2+-independent manner via the vWA domain. Mutations in the C2B that inhibit binding to PI(3)P not only block the activity-induced translocation of Copine-6 to early endosomes, but it also causes an aberrant accumulation of Copine-6 in recycling endosomes. Consequently, loss of Copine-6 expression impairs the efficient coupling of early and recycling endosomes and blocks activity-dependent delivery of both AMPA and NMDA receptors onto the neuronal plasma membrane. These defects can be restored by re-expressing wild-type Copine-6, but not the C2B phospholipid-binding mutant. Together, our findings establish Copine-6 as a molecular bridge that enhances coupling between the early and recycling endosomal membranes, thereby facilitating the activity-dependent forward trafficking of glutamate receptors to the neuronal plasma membrane to maintain synaptic potentiation.

cell biology↗

bioRxiv2026

Know Your Alphabet: Conformational Noise, Latent-Space Encodings, and the Future of Structural Phylogenetics

Structural alphabets have transformed protein phylogenetics by enabling sequence-style alignment and maximum-likelihood inference to be applied directly to structural data. However, a coordinate-explicit alphabet, in which character states are derived from three-dimensional atomic positions, encodes not only evolutionary signal but also the conformational variability inherent to protein structure. This source of noise has not previously been quantified in a phylogenetic context, and no framework exists for comparing alphabets with respect to their conformational sensitivity. Here, we introduce the Normalised Noise Index (NNI), a Shannon entropy-based metric for quantifying conformational sensitivity in structural alphabet encodings, and apply it alongside ensemble-wide Robinson-Foulds (RF) variance as a framework for characterising the impact of conformational noise on phylogenetic inference. Across 3,749 single-chain NMR ensembles from the Protein Data Bank, we show that 3Di character variability is a pervasive feature of experimentally observed conformational spread, with NNI negatively correlated with within-ensemble structural stability. A 100 ns molecular dynamics simulation of myoglobin confirmed that thermal fluctuations alone are sufficient to generate comparable 3Di character variation and, in 2.9% of cases, to redirect maximum-likelihood tree search away from the expected topology in a 4-taxon globin benchmark with independently established relationships. Exhaustive enumeration of 4,800 conformational replicates across three NMR ensembles revealed that topological variance under 3Di encoding is approximately 1.7-fold greater than under structural distance, based on 11,517,600 pairwise RF comparisons, a source of uncertainty invisible to standard bootstrap analysis. By contrast, TEA, a sequence-derived structure-aware alphabet inferred from ESM-2 embeddings rather than directly from atomic coordinates, is insulated from conformational sampling by construction and yields zero topological variance across all conformational replicates, serving here as a noise-insulated reference rather than a proposed replacement for 3Di. Together, these results demonstrate that alphabet choice is a methodological variable in structural phylogenetics, and that the NNI metric and RF variance frame-work introduced here provide a practical basis for principled noise characterisation as new structural alphabets continue to emerge.

bioinformatics↗

bioRxiv2026

Benchmarking Heritability Estimation Strategies Across 86 Configurations and Their Downstream Effect on Polygenic Risk Score Performance

ObjectiveSNP heritability estimates vary substantially across estimation strategies, yet the downstream consequences for polygenic risk score (PRS) construction remain poorly characterised. We systematically benchmarked heritability estimation configurations and assessed their propagation into downstream PRS performance. MethodsWe benchmarked 86 heritability-estimation configurations spanning six tool families (GEMMA, GCTA, LDAK, DPR, LDSC, SumHer) and ten method groups across 10 UK Biobank phenotypes, yielding 844 configuration-level estimates. Each estimate was propagated into GCTA-SBLUP and LDpred2-lassosum2 PRS frameworks and evaluated across five cross-validation folds using null, PRS-only, and full models. Eleven binary analytical contrasts were tested using Mann-Whitney U tests to identify drivers of heritability variability. ResultsHeritability ranged from -0.862 to 2.735 (mean = 0.134, SD = 0.284), with 133 of 844 estimates (15.8%) negative and concentrated in unconstrained estimation regimes. Ten of eleven analytical contrasts significantly affected heritability magnitude, with algorithm choice and GRM standardisation showing the largest effects. Despite this upstream variability, downstream PRS test performance was only weakly coupled to heritability magnitude: pooled Pearson correlations between h2 and test AUC were r = -0.023 for GCTA-SBLUP and r = +0.014 for LDpred2-lassosum2 (both non-significant). ConclusionSNP heritability is best interpreted as a configuration-sensitive modelling parameter rather than a universally stable scalar input. Heritability estimates should always be reported alongside their full estimation specification, and downstream PRS performance is comparatively robust to moderate variation in the heritability input. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=80 SRC="FIGDIR/small/716079v1_ufig1.gif" ALT="Figure 1"> View larger version (27K): org.highwire.dtl.DTLVardef@112929borg.highwire.dtl.DTLVardef@573c36org.highwire.dtl.DTLVardef@132170borg.highwire.dtl.DTLVardef@1871363_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2026

The Duplicate Monophyly Criterion: An Empirical Approach to Bootstrapping Distance-Based Structural Phylogenies

Distance-based structural phylogenies summarise relationships between proteins using pairwise structural similarity scores. Unlike character-based methods, however, they lack a natural analogue of the non-parametric bootstrap. Because distances derived from continuous, high-dimensional folds provide no discrete sites to resample, rigorous support estimation would ideally rely on conformational ensembles from Molecular Dynamics or Monte Carlo simulations, which is computationally prohibitive at scale. A practical alternative is parametric bootstrapping in distance space, but this introduces a calibration problem: without an objective estimate of the signal-to-noise ratio, the perturbation magnitude cannot be chosen in a principled way. Here, we introduce the Duplicate Monophyly Criterion (DMC), an empirical strategy for calibrating distance-matrix perturbations using synthetic taxon duplicates as internal controls. We expand the observed distance matrix with virtual copies of each taxon, assign each original-duplicate pair a small baseline distance, and apply a floor-augmented heteroscedastic noise model directly in distance space. The central hypothesis is that loss of duplicate monophyly (i.e., failure of taxon-duplicate pairs to form two-tip cherries) marks a regime in which perturbations overwhelm intrinsic phylogenetic signal, thereby defining a conservative operating range for stability-based (bootstrap-like) support estimation. As a practical use case, we select a dataset-specific perturbation level{lambda} * as the largest noise level that retains a target fraction of duplicate pairings (e.g., [≥] 90%), and report split frequencies across replicate trees generated at{lambda} * as support values. We validate the framework in a geometric toy model where evolving two-dimensional shapes provide a known ground-truth topology, and in an empirical globin benchmark using distances derived from 1 - TM-score. Across both settings, duplicate monophyly tracks the erosion of tree structure under distance-matrix noise, establishing an internally calibrated "resolution limit" for assigning confidence in distance-based structural phylogenetics.

bioinformatics↗

bioRxiv2026

A harmonized benchmarking framework for implementation-aware evaluation of 46 polygenic risk score tools across binary and continuous phenotypes

Polygenic risk score (PRS) tools differ substantially in statistical assumptions, input requirements, and implementation complexity, making direct comparison difficult. We developed a harmonized, implementation-aware benchmarking framework to evaluate 46 PRS tools across seven binary UK Biobank phenotypes and one continuous trait under three model configurations: null, PRS-only, and PRS plus covariates. The framework integrates standardized preprocessing, tool-specific execution, hyperparameter exploration, and unified downstream evaluation using five-fold cross-validation on high-performance computing infrastructure. In addition to predictive performance, we assessed runtime, memory use, input dependencies, and failure modes. A Friedman test across 40 phenotype-fold combinations confirmed significant differences in tool rankings ({chi}2 = 102.29, p = 2.57 x 10-11), with no single method universally optimal. These findings provide a reproducible framework for comparative PRS evaluation and demonstrate that tool performance is shaped not only by statistical methodology but also by phenotype architecture, preprocessing choices, covariate structure, computational demands, software robustness, and practical implementation constraints.

bioinformatics↗

bioRxiv2026

Benchmarking 80 binary phenotypes from the openSNP dataset using deep learning algorithms and polygenic risk score tools

Genotype-phenotype prediction plays a crucial role in identifying disease-causing single nucleotide polymorphisms and precision medicine. In this manuscript, we benchmark the performance of various machine/deep learning algorithms and polygenic risk score tools on 80 binary phenotypes extracted from the openSNP dataset. After cleaning and extraction, the genotype data for each phenotype is passed to PLINK for quality control, after which it is transformed separately for each of the considered tools/algorithms. To compute polygenic risk scores, we used the quality control measures for the test data and the genome-wide association studies summary statistic file, along with various combinations of clumping and pruning. For the machine learning algorithms, we used p-value thresholding on the training data to select the single nucleotide polymorphisms, and the resulting data was passed to the algorithm. Our results report the average 5-fold Area Under the Curve (AUC) for 29 machine learning algorithms, 80 deep learning algorithms, and 3 polygenic risk scores tools with 675 different clumping and pruning parameters. Machine learning outperformed for 44 phenotypes, while polygenic risk score tools excelled for 36 phenotypes. The results give us valuable insights into which techniques tend to perform better for certain phenotypes compared to more traditional polygenic risk scores tools.

bioinformatics↗

bioRxiv2026

Identifying genes associated with phenotypes using machine and deep learning

Identifying disease-associated genes enables the development of precision medicine and the understanding of biological processes. Genome-wide association studies (GWAS), gene expression data, biological pathway analysis, and protein network analysis are among the techniques used to identify causal genes. We propose a machine-learning (ML) and deep-learning (DL) pipeline to identify genes associated with a phenotype. The proposed pipeline consists of two interrelated processes. The first is classifying people into case/control based on the genotype data. The second is calculating feature importance to identify genes associated with a particular phenotype. We considered 30 phenotypes from the openSNP data for analysis, 21 ML algorithms, and 80 DL algorithms and variants. The best-performing ML and DL models, evaluated by the area under the curve (AUC), F1 score, and Matthews correlation coefficient (MCC), were used to identify important single-nucleotide polymorphisms (SNPs), and the identified SNPs were compared with the phenotype-associated SNPs from the GWAS Catalog. The mean per-phenotype gene identification ratio (GIR) was 0.84. These results suggest that SNPs selected by ML/DL algorithms that maximize classification performance can help prioritise phenotype-associated SNPs and genes, potentially supporting downstream studies aimed at understanding disease mechanisms and identifying candidate therapeutic targets.

bioinformatics↗

bioRxiv2026

GWAS Summary Statistic Tool: A Meta-Analysis and Parsing Tool for Polygenic Risk Score Calculation

MotivationGWAS (genome-wide association study) summary statistic files are essential inputs for polygenic risk score (PRS) calculation, yet identifying suitable files across thousands of catalog entries requires downloading large files and manually inspecting their column structures--a process that is time-consuming and storage-intensive. ResultsWe present GWASPoker, a phenotype-driven, GWAS-Catalog-specific pre-download triage tool that scans candidate GWAS files for PRS column availability by partial download and header detection, without requiring full-file transfer. Analysing 60,499 records from the GWAS Catalog, 60,281 (99.6%) contained accessible download links, of which 54,026 (89.6%) were successfully partially downloaded and parsed across 20 file formats, yielding 724 unique header signatures. Across 13 phenotypes, 84 of 85 manually curated GWAS files (98.8%) were automatically retrieved and processed. Header validation against fully downloaded files showed exact agreement in 23 of 28 cases (82.1%). Availability and implementationGWASPoker is implemented in Python 3 and freely available at https://github.com/MuhammadMuneeb007/GWASPokerforPRS under the MIT licence. Example outputs and documentation are provided in the repository. The tool was tested on Linux (HPC cluster) with Python 3.8+. The LLM-based code-generation step is entirely optional; a rules-based column-mapping template is provided for fully offline use.

bioinformatics↗

bioRxiv2025

An Empirical Analysis of Fine-Tuning Large Language Models on Bioinformatics Literature: PRSGPT and BioStarsGPT

Large language models (LLMs) often lack specialized knowledge for complex bioinformatics applications. We present a reproducible pipeline for fine-tuning LLMs on specialized bioinformatics data, demonstrated through two use cases: PRSGPT, focused on polygenic risk score (PRS) tools, and BioStarsGPT, trained on community forum discussions. The nine-step pipeline integrates diverse data sources, structured preprocessing, prompt-based question-answer (QA) generation (via Google Gemini), natural language inference (NLI) for quality control, semantic deduplication, clustering-based data splitting, and parameter-efficient fine-tuning using LoRA. We fine-tuned three LLMs (LLaMA-3.2-3B, Qwen2.5-7B, Gemma) and benchmarked them on over 14 lexical and semantic metrics. Qwen2.5-7B emerged as the best performer, with BLEU-4 and ROUGE-1 improvements of 82% and 70% for PRSGPT and 6% and 18% for BioStarsGPT, respectively. The open-source datasets produced include over 28,000 QA pairs for PRSGPT and 154,282 for BioStarsGPT. Human evaluation of PRSGPT yielded 61.9% accuracy on the PRS tools comparison task, comparable to Google Gemini (61.4%), but with richer methodological detail and accurate citations. BioStarsGPT demonstrated 59% conceptual accuracy across 142 curated bioinformatics questions. Our pipeline enables scalable, domain-specific fine-tuning of LLMs. It enables privacy-preserving, locally deployable bioinformatics assistants, explores their practical applications, and addresses the challenges, limitations, and mitigation strategies associated with their development and use. Key PointsO_LIWe present a comprehensive 9-step pipeline for fine-tuning large language models on bioinformatics data, demonstrated through PRSGPT (PRS tools) and BioStarsGPT (forum discussions). C_LIO_LIThe pipeline generated over 28,000 QA pairs for PRSGPT from diverse sources (PDFs, GitHub repositories, research articles) and 154,282 QA pairs for BioStarsGPT from community forum discussions. C_LIO_LIQwen2.5-7B emerged as the best-performing model, achieving BLEU-4 improvements of 82% for PRSGPT and 6% for BioStarsGPT, with ROUGE-1 improvements of 70% and 18% respectively. C_LIO_LIThe methodology integrates automated QA generation via Google Gemini, NLI for quality control, semantic deduplication, and parameter-efficient fine-tuning using LoRA, providing a scalable framework for domain-specific LLM adaptation. C_LI

bioinformatics↗

bioRxiv2025

Quantome: A Quantum Surrogate Model for Biophysical Landscapes

Accurately modelling the potential energy landscapes that govern molecular interactions is a central challenge in computational biophysics. While quantum computers promise to solve such problems with high fidelity, a key bottleneck is the encoding of complex spatial information into low-qubit Hamiltonians suitable for near-term devices. Here, we introduce a generalisable framework for creating quantum surrogate models of 2D biophysical landscapes. Our method translates a discrete, classically-derived potential energy grid into a continuous quantum Hamiltonian by fitting it to a high-degree polynomial, where the polynomials coefficients directly define the potential energy operator. We demonstrate this pipeline on a custom-designed landscape featuring two asymmetric potential wells. By systematically varying a kinetic hopping term in the Hamiltonian, our variational quantum eigensolver simulations, averaged over 100 runs, successfully reproduce the physical transition from a localised ground state to a delocalised state governed by tunnelling. The entire framework is made accessible through the Quantome web application, a pedagogical platform that allows users to both design custom landscapes and explore pre-calculated results on EF-hand protein binding sites, which serve as a simplified real-world test bed for this framework. The web app is freely available at: https://biosig.lab.uq.edu.au/quantumlabs/quantome.

biophysics↗

bioRxiv2025

On a quantum-inspired kernel for classifying protein torsion angles

Algorithms grounded in quantum principles need to demonstrate they are at least as expressive as established classical methods before any hardware advantage can be sought. Here we demonstrate that a quantum-inspired kernel reaches the same accuracy and Matthews correlation coefficient as carefully tuned radial basis function and degree-2 polynomial support vector machines when tasked with separating geometric regions in Ramachandran space. On a rigorously curated, balanced dataset of 10,000 torsion angle pairs derived from DSSP-annotated residues, the model achieves 98% accuracy with a Matthews correlation coefficient of 0.96, comparable to the top-performing classical models. By achieving true predictive parity on a well-characterised structural benchmark, the quantum-inspired kernel establishes that quantum-informed similarity measures already match classical baselines, laying a firm groundwork for future quantum-native bioinformatics workflows once suitable hardware becomes available.

bioinformatics↗

bioRxiv2023

Tertiary-interaction characters enable fast, model-based structural phylogenetics beyond the twilight zone

Protein structure is more conserved than protein sequence, and therefore may be useful for phylogenetic inference beyond the "twilight zone" where sequence similarity is highly decayed. Until recently, structural phylogenetics was constrained by the lack of solved structures for most proteins, and the reliance on phylogenetic distance methods which made it difficult to treat inference and uncertainty statistically. AlphaFold has mostly overcome the first problem by making structural predictions readily available. We address the second problem by redeploying a structural alphabet recently developed for Foldseek, a highly-efficient deep homology search program. For each residue in a structure, Foldseek identifies a tertiary interaction closest-neighbor residue in the structure, and classifies it into one of twenty "3Di" states. We test the hypothesis that 3Dis can be used as standard phylogenetic characters using a dataset of 53 structures from the ferritin-like superfamily. We performed 60 IQtree Maximum Likelihood runs to compare structure-free, PDB, and AlphaFold analyses, and default versus custom model sets that include a 3DI-specific rate matrix. Analyses that combine amino acids, 3Di characters, partitioning, and custom models produce the closest match to the structural distances tree of Malik et al. (2020), avoiding the long-branch attraction errors of structure-free analyses. Analyses include standard ultrafast bootstrapping confidence measures, and take minutes instead of weeks to run on desktop computers. These results suggest that structural phylogenetics could soon be routine practice in protein phylogenetics, allowing the re-exploration of many fundamental phylogenetic problems.

evolutionary biology↗

bioRxiv2019

Missense mutations in the MLKL brace region lead to lethal neonatal inflammation in mice and are present in high frequency in humans.

We have isolated a mouse strain with a single missense mutation in the gene encoding MLKL, the essential effector of necroptotic cell death. The resulting substitution lies within the two-helix brace and confers constitutive, RIPK3 independent, killing activity to MLKL. Mice homozygous for MlklD139V develop lethal inflammation within days of birth, implicating the salivary glands and pericardium as hotspots for necroptosis and inflammatory infiltration. The normal development of MlklD139V homozygotes until birth, and the absence of any overt phenotype in heterozygotes provides important in vivo precedent for the capacity of cells to clear activated MLKL. These observations offer an important insight into the potential disease-modulating roles of three common human MLKL polymorphisms that encode amino acid substitutions within or adjacent to the brace region. Compound heterozygosity of these variants is found at up to 12-fold the expected frequency in patients that suffer from a pediatric autoinflammatory disease, CRMO.

cell biology↗