Biology subjects

Ascher, D. B.

Publications and source records attributed to Ascher, D. B..

At least 19 recordsLinked to original sources

bioRxiv2026

Structural basis of coiled-coil domain-mediated inhibition of A. baumannii AbTir NAD+ nucleosidase activity

TIR (Toll/interleukin-1 receptor) domains are central components of innate immunity and cell-death pathways across diverse forms of life. In bacteria, TIR domain-containing proteins participate in diverse antiphage defence systems, several of which rely on NADase activity carried out by TIR domains. This nucleosidase activity is often regulated by additional domains that modulate activation and specificity in response to infection. AbTir, a TIR domain-containing protein from Acinetobacter baumannii, contains an N-terminal CC domain that negatively regulates its NADase activity, but the structural basis of this regulation remains unknown. Here, we used X-ray crystallography to investigate the molecular basis of CC domain-mediated regulation in AbTir. We determined the crystal structure of full-length AbTir in complex with a nanobody, revealing a symmetric AbTir TIR dimer in which only one of the two symmetry-related CC helices is accommodated at the dimer surface, breaking the overall symmetry of the full-length protein. Structural comparisons with other bacterial TIR domains suggest that such TIR dimers likely represent an inactive state, in which the NAD+-binding site has not yet formed. Phylogenetic analysis places AbTir within a family of bacterial TIR proteins associated with defence-related genomic loci, although phage-spot assays performed in the surrogate host Escherichia coli did not detect measurable antiphage activity. Together, these findings reveal an unusual structural mechanism of CC domain-mediated inhibition of a bacterial TIR NADase and expand our understanding of the mechanistic diversity of bacterial TIR proteins.

biochemistry↗

bioRxiv2026

CharacTERT: A machine learning tool for classifying hTERT missense variants

Missense mutations in TERT, the gene encoding the human telomerase catalytic subunit hTERT, are associated with Telomere Biology Disorders (TBDs). Experimentally elucidating the effects of all possible missense variants would be time-consuming and technically challenging. Moreover, current computational predictors are not hTERT-specific and primarily rely on sequence information, failing to capture the complex biological and structural context of the telomerase enzyme. In this work, we developed three machine learning models integrating both sequence- and structure-based features to account for the biological mechanisms of hTERT. Compared to state-of-the-art methods, our best-performing models achieved a higher Matthews Correlation Coefficient of 0.88 on ClinVar and gnomAD curated variants and demonstrated robust sensitivity (0.75) on a dataset curated according to guidelines from the American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP). Feature interpretation highlighted hTERT residue conservation and changes in hydrophobic and weak polar interactions as critical determinants of pathogenicity. Finally, in silico saturation mutagenesis was performed to present a mutational landscape of TERT, available in a user-friendly web server, CharacTERT, which could offer valuable insights into the molecular mechanisms driving TBDs, aid in early diagnosis, as well as guide personalized treatment strategies. CharacTERT is freely available at https://biosig.lab.uq.edu.au/charactert/.

bioinformatics↗

bioRxiv2026

PhenotypeToGeneDownloaderR: automated multi-source retrieval and validation of phenotype-associated genes

MotivationIdentifying phenotype-associated genes is a common first step in polygenic risk score construction, enrichment testing, target prioritisation and variant interpretation, but relevant evidence is distributed across heterogeneous databases with different interfaces, formats and evidence models. ResultsWe present PhenotypeToGeneDownloaderR, a phenotype-guided R/Python pipeline for automated gene retrieval, harmonisation, symbol validation and cross-source summary analysis. Given a phenotype term, the pipeline queries integrated biological databases, standardises per-source outputs, combines gene lists, validates retrieved symbols against the NCBI human gene reference and generates summary tables and visualisations. Across 13 clinically relevant phenotypes and 13 databases, PhenotypeToGeneDownloaderR generated 136,487 raw gene retrievals, with at least one source returning genes for every phenotype. Across all 13 phenotypes, 100,175 of 114,345 combined input symbols were retained after direct or synonym-based validation, corresponding to an 87.6% validation rate. Cross-source overlap was low, supporting the complementarity of integrated evidence sources. Against an HPO/ClinVar/OMIM-derived gold standard, the pipeline recovered 1,039 of 1,056 known phenotype-associated genes, corresponding to 98.4% recall. PhenotypeToGeneDownloaderR provides a lightweight, reproducible upstream framework for generating candidate gene sets for downstream prioritisation and interpretation. Availability and implementationPhenotypeToGeneDownloaderR is implemented in R and Python, released under the MIT licence, and available at https://github.com/MuhammadMuneeb007/PhenotypeToGeneDownloaderR. Supplementary informationSupplementary data are available online.

bioinformatics↗

bioRxiv2026

AnnotateMissense: a genome-wide annotation and benchmarking framework for missense pathogenicity prediction

MotivationMissense variant interpretation remains challenging because pathogenicity depends on heterogeneous evidence, including population frequency, evolutionary conservation, transcript context, amino acid substitution severity, prior pathogenicity predictors and protein-language-model-derived features. Although these resources are individually useful, there remains a need for reproducible workflows that integrate them at genome-wide scale, benchmark their contribution and provide accessible outputs for downstream research. ResultsWe present AnnotateMissense, a scalable annotation, benchmarking and genome-wide prediction framework for missense variant interpretation. AnnotateMissense integrates chromosome-wise hg38 missense variants derived from dbNSFP v5.1 with ANNOVAR-based gene-, region- and filter-based annotations, dbNSFP transcript and protein descriptors, AlphaMissense scores, ESM-derived features, conservation metrics, population-frequency variables, established pathogenicity predictors and engineered amino acid/codon-context features. Using 132,714 ClinVar-labelled missense variants, we benchmarked machine-learning and deep-learning models under controlled feature configurations. The full 303-feature benchmark set achieved the strongest performance with XGBoost, reaching mean Matthews correlation coefficient (MCC) = 0.9411 and ROC-AUC = 0.9950 across stratified five-fold cross-validation. Restricted naive and location-oriented feature sets achieved substantially lower best MCC values of 0.4989 and 0.5113, respectively. Circularity-controlled ablations showed that removing prior-predictor, population-frequency and clinically overlapping evidence reduced performance, whereas excluding AlphaMissense and ESM-derived features alone had minimal effect. Temporal ClinVar validation on newly observed pathogenic/benign variants achieved MCC = 0.7613, accuracy = 0.8798 and F1-score = 0.8750. The final genome-wide model was applied to 90,643,830 hg38 missense variants to generate AnnotateMissense pathogenicity scores and binary prediction labels. Availability and implementationSource code, workflow scripts and command files are available at https://github.com/MuhammadMuneeb007/CAGI7_Annotate_All_Missense. Genome-wide prediction outputs and the compressed DuckDB database are available at https://doi.org/10.5281/zenodo.19981867. Supplementary informationSupplementary data are available online.

bioinformatics↗

bioRxiv2026

UshEffect-3D: Structure-informed Classification of USH2A Missense Variants for Inherited Retinal Disease

USH2A encodes Usherin, a 5,202-residue multi-domain protein whose missense variation is a major cause of Usher syndrome type II. Interpretation remains challenging because the protein lacks a full-length experimental structure and thousands of reported missense variants remain unresolved. In an attempt to address this gap, we developed UshEffect-3D, a gene-specific machine-learning-based variant effect predictor that integrates sequence-based evolutionary measures, structure-based biochemical properties, local interaction terms, and predicted protein stability changes to prioritise USH2A missense variants of unknown significance (VUS). To ensure that the structure-derived features were extracted from reliably modelled predicted structures, the AlphaFold2 models of 47 annotated Usherin domains were compared against experimental structures of representative domain types. The predicted structures showed strong fold concordance with representative experimental structures, with 43 domains achieving TM-scores of at least 0.5. Among binary classifiers trained using a residue position-grouped cross-validation strategy, Logistic Regression was selected as the most stable performer and achieved an MCC of 0.75, AUC of 0.96, sensitivity of 0.93 and specificity of 0.81 on a group-aware held-out test set. On a benchmarking task, UshEffect-3D performed comparably to six general-purpose variant-effect predictors evaluated on the held out test set, scoring the highest on every balanced metric but no statistically resolvable advantage over any major competitors such as PolyPhen-2, VESPAl, ESM-1b and AlphaMissense. Feature ablation displayed that sequence derived evolutionary constraints were the sole detectable source of discrimination - withholding the conservation features reduced cross-validated MCC by 0.35 (paired 95% CI -0.45 to -0.25), whereas removing structural descriptors and predicted stability change negligibly affected model performance. Applying our model to the 2,639 ClinVar VUS, we reclassified 981 (37.2%) as likely pathogenic highlighting the extent of disease-relevant missense variation yet to be uncovered. Thus, UshEffect-3D serves as an interpretable protein-specific framework and a supplementary ranked resource that can be used to flag unresolved variants for further clinical and experimental review.

bioinformatics↗

bioRxiv2025

LDLR variant classification through activity-normalized prime editing screening

BackgroundInherited variants in the LDL receptor (LDLR) gene are the most common cause of familial hypercholesterolemia (FH), significantly increasing coronary artery disease risk. Early identification of pathogenic LDLR variants enables prompt intervention with lipid-lowering therapies; however, the majority of LDLR variants observed in the population have uncertain or absent clinical classifications, limiting the potential to improve clinical management. MethodsWe developed an innovative, activity-normalized prime editing screening pipeline to measure the impact of 5,184 LDLR coding variants on LDL-cholesterol (LDL-C) uptake. Through pairing a genotypic outcome reporter with every prime editing guide RNA (pegRNA), we adjust phenotypic measurements to account for variable editing efficiency, extending activity normalization to prime editing for the first time at this scale. Further, we use a statistical estimation approach that leverages measurements for all missense variants at a given position to denoise the resulting scores. ResultsWe show that prime editing-mediated reporter editing correlates with endogenous variant installation frequency, allowing activity normalization to improve imputation of LDLR variant effect. Our optimized prime editing assay identifies a broad, continuous spectrum of variant functional effects. We achieve robust separation of pathogenic vs. benign ClinVar variants and concordance between experimentally derived functional scores and LDL-C levels measured in UK Biobank participants. Further, when calibrating the strength of evidence provided by this functional screening data to align with the ACMG/AMP variant interpretation guidelines, and integrating additional sources of evidence, a majority of currently unclassified rare LDLR variants meet evidence thresholds for reclassification. We use the broad coverage of this screen to gain insight into how apolipoproteins bind to LDLR. In particular, we identify and characterize rare LDLR variants that enhance LDL-C uptake through increased interaction with apolipoprotein B. Finally, we compare prime editing-based functional scores with those derived from recent base editing and cDNA-based LDLR variant screens, showing that these approaches all show robust correlation with clinically observed LDL-C levels and computational scores, while prime editing identifies candidate splice-altering coding variants that are not modeled by cDNA screening. ConclusionsAltogether, our approach demonstrates the power of prime editing to significantly improve understanding of how variants in LDLR impact function and contribute to FH.

genetics↗

bioRxiv2025

Risk Based Prediction of Novel AMR Variants Using Protein Language Models

Antimicrobial resistance (AMR) is among the most pressing global health threats of the 21st century, with the potential to thrust modern medicine back into a pre-antibiotic era. Resistance can arise through diverse mechanisms, including genomic mutations that prevent antibiotics from reaching or acting on their targets. To limit the spread of AMR, surveillance systems must detect both known and emerging resistance markers. Here we present AMRscope, a model trained on ESM2 protein language model embeddings of single mutations for prediction of resistance likelihood, combined with a rigorous evaluation framework. This tool is applied across antibiotic-interacting proteins of different bacterial species, including WHO priority pathogens, such as rifampicin-resistant M. tuberculosis and carbapenem-resistant P. Aeruginosa. Performance on random splits achieves a competitive accuracy, F1 and MCC of 0.88, 0.87 and 0.75, respectively, while additional splitting strategies demonstrate transfer of predictive power to unseen organisms or genes. Moreover, in silico deep mutational scanning and structural mapping across these targets reveals the tool can recover known resistance-associated regions and highlight new candidates. The risk-based outputs complement database matching and resistance element detection tools, providing clinicians and public health agencies with an interpretable and scalable system for AMR surveillance and proactive response.

bioinformatics↗

bioRxiv2025

Transfer learning across molecular graphs for predicting protein-ligand affinities and their changes upon mutations

Predicting protein-ligand binding affinity and mutation-induced affinity changes ({Delta}{Delta}G) remains challenging due to limited data and complex interaction mechanisms. Here we present DDMuffin, a deep learning framework that integrates structural, sequence, and interaction graph features, employing transfer learning and stringent dataset partitioning to achieve reliable generalization. DDMuffin demonstrates strong predictive accuracy on the rigorous LP-PDBBind benchmark (Pearson r up to 0.70 after excluding top 10% outliers; RMSE = 1.48 kcal mol-{superscript 1}). In evaluating mutation-induced affinity changes for clinically relevant kinase inhibitors, DDMuffin achieves competitive average performance (mean Spearman{rho} = 0.39), outperforming or matching several established methods. The approach provides valuable insights into mutation-driven drug resistance and interpretable ligand binding mechanisms, particularly advantageous for guiding inhibitor design and personalized therapeutic strategies. To facilitate broad application, we deployed DDMuffin as an accessible web server at https://biosig.lab.uq.edu.au/ddmuffin/, promoting systematic exploration of protein-ligand interactions in drug discovery research.

bioinformatics↗

bioRxiv2025

Structome-AlignViewer: On Confidence Assessment in Structure-Aware Alignments

Protein structure-based comparison provides a framework for uncovering deep evolutionary relationships that can escape conventional sequence-based approaches. Encoding three-dimensional protein structures using a simplified structure-aware alphabet, can lead to compact, comparable strings that retain key spatial relationships. Although this enables comparison, structure-aware alignments can experience misaligned regions, particularly when comparing proteins with substantial divergence in fold architecture. To address this, a web-based resource, Structome-AlignViewer, is introduced in this work for evaluating the quality of structure-aware alignments through both spatial mapping of alignment columns to protein structures and quantitative confidence scoring. Confidence is computed from pairwise structural substitutions between adjacent inputs and normalized within each alignment to highlight relatively well-supported columns. To provide broader context, thousands of alignments from established structural classification systems were analysed, allowing for an empirical comparative statistic to be derived to assess alignment quality. Option to exclude gap-rich regions enable users to refine alignments and focus on conserved structural cores. This approach provides an interpretable method for assessing structural alignment quality and supports more robust comparative and evolutionary analyses. Structome-AlignViewer is freely available at https://biosig.lab.uq.edu.au/structome_alignviewer/.

bioinformatics↗

bioRxiv2025

On interactive spatial visualisation of pathogenicity predictions

The past few years have revolutionised the utility of protein structural information in assessing disease risk. Such efforts have been pioneered by the DeepMind and Meta teams, who initially developed accurate protein prediction tools AlphaFold and ESMFold, respectively. More recently, both have been readapted to create AlphaMissene and ESM1b, through which, the pathogenicity potential of all possible missense mutations within the human proteome has been predicted. Similarly, VESPA was created as an alignment-agnostic predictor, which infers conservation from protein-language model embeddings to assign an effect for each variant. To facilitate their utility, we present a unified dashboard, MissenseViewer, which permits direct predictor comparison, and visualisation within the protein 3-dimensional structural context. Our interactive resource is freely available at: https://biosig.lab.uq.edu.au/missenseviewer/.

bioinformatics↗

bioRxiv2025

Structome-TM: Complementing dataset assembly for structural phylogenetics by addressing size-based biases

Harnessing the explosion of protein structure data to uncover deep evolutionary relation-ships requires effective comparison methods. While widely used global alignment techniques are powerful, they can fail to identify homologous structures that differ significantly in size or domain architecture. To address this limitation, we introduce Structome-TM, a web resource for assembling datasets for distance-based phylogenetic reconstruction. Making use of the TM-score to prioritize local structural similarity, Structome-TM excels at identifying these otherwise obscured relationships, allowing users to build a comprehensive structural neighbourhood of proteins suitable for comparison. To facilitate this dataset assembly, the resource accepts both RCSB PDB identifiers and protein sequences as inputs. When querying using a protein sequence, protein structures are predicted in real-time and their respective neighbourhoods determined, enabling analysis where experimentally determined structures may not be available. Through its user-friendly interface, Structome-TM provides a powerful and necessary approach for a more comprehensive exploration of protein evolution. This resource is freely available at: https://biosig.lab.uq.edu.au/structome_tm/.

bioinformatics↗

bioRxiv2024

AI-m6ARS: Machine learning-driven m6A RNA methylation site discovery with integrated sequence, conservation, and geographical descriptors

N6-Methyladenosine (m6A) is a predominant type of human RNA methylation, regulating diverse biochemical processes and being associated with the development of several diseases. Despite its significance, an extensive experimental examination across diverse cellular and transcriptome contexts is still lacking due to time and cost constraints. Computational models have been proposed to prioritise potential m6A methylation sites, although having limited predictive performance due to inadequate characterisation and modelling of m6A sites. This work presents AI-m6ARS, a novel model that utilises integrated sequence, conservation, and geographical descriptive features to predict human m6A methylation sites. The model was trained using the Light Gradient Boosting Machine (LightGBM) algorithm, which was coupled with comprehensive feature selection to improve the data quality. AI-m6RS demonstrates strong predictive capabilities, achieving an impressive area under the receiver operating characteristic curve of 0.87 on cross-validation. Consistent results on unseen transcripts in a blind test highlight the AI-m6ARS generalisability. AI-m6ARS also demonstrates comparable performance to state-of-the-art models, but offers two significant benefits: the model interpretability and the availability of a user-friendly web server. The AI-m6ARS web server offers valuable insights into the distribution of m6A sites within the human genome, thereby facilitating progress in medical applications. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=77 SRC="FIGDIR/small/599439v1_ufig1.gif" ALT="Figure 1"> View larger version (13K): org.highwire.dtl.DTLVardef@12d7502org.highwire.dtl.DTLVardef@15cf6b5org.highwire.dtl.DTLVardef@490699org.highwire.dtl.DTLVardef@5046c1_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2024

From Mutations to Disease: Computational Analysis and Interpretation of GPCR-Associated Pathogenicity

G protein-coupled receptors (GPCRs) perform critical roles in numerous physiological processes and their mutations are, therefore, associated with various human diseases. Hence, understanding the molecular consequences of pathogenic mutations in GPCRs is essential for elucidating disease mechanisms and developing effective therapeutic strategies. In this study, we employed computational approaches to explore the impact of mutations on GPCRs using two distinct datasets: ClinVar and MutHTP. We first evaluated the performance of available pathogenicity predictors. Beyond that, we used statistical analysis to identify key characteristics of mutations in GPCRs leading to diseases. We first evaluated available computational predictors, such as SIFT, PolyPhen-2, PROVEAN, ESM1b, and AlphaMissense in classifying GPCR mutations. During this task, we observed that all predictors performed with reliability when assessing GPCR mutations leading to diseases in the ClinVar dataset. On the other hand, when dealing with the MutHTP dataset, all predictors demonstrated poor performance, emphasising the importance of dataset characteristics and the need for comprehensive evaluation when selecting mutation predictive tools for GPCR analysis. The statistical analysis of mutations on GPCRs and disease development suggests that mutations occurring in conserved regions or regions with stronger intermolecular interactions are more likely to disrupt protein function and contribute to disease pathogenesis. Additionally, regarding our analysis, we also obtained insights into the importance of hydrophobic interactions and hydrogen bonding patterns in mutations in GPCRs and pathogenicity. Overall, our study enhances our understanding of the molecular mechanisms underlying GPCR-associated diseases and provides valuable insights for future research and clinical diagnostics in this field.

bioinformatics↗

bioRxiv2024

PRIMITI: a computational approach for accurate prediction of miRNA-target mRNA interaction

Current medical research has been demonstrating the roles of miRNAs in a variety of cellular mechanisms, lending credence to the association between miRNA dysregulation and multiple diseases. Understanding the mechanisms of miRNA is critical for developing effective diagnostic and therapeutic strategies. miRNA-mRNA interactions emerge as the most important mechanism to be understood despite their experimental validation constraints. Accordingly, several computational models have been developed to predict miRNA-mRNA interactions, albeit presenting limited predictive capabilities, poor characterisation of miRNA-mRNA interactions and low usability. To address these drawbacks, we developed PRIMITI, a PRedictive model for the Identification of novel MIRNA-Target mRNA Interactions. PRIMITI is a novel machine learning model that utilises CLIP-seq and expression data to characterise functional target sites in 3-untranslated regions (3-UTRs) and predict miRNA-target mRNA repression activity. The model was trained using a reliable negative sample selection approach and the robust extreme gradient boosting (XGBoost) model, which was coupled with newly introduced features, including sequence and genetic variation information. PRIMITI achieved an area under the receiver operating characteristic (ROC) curve (AUC) up to 0.96 for a prediction of functional miRNA-target site binding and 0.96 for a prediction of miRNA-target mRNA repression activity on cross-validation and an independent blind test. Additionally, the model outperformed state-of-the-art methods in recovering miRNA-target repressions in an unseen microarray dataset and in a collection of validated miRNA-mRNA interactions, highlighting its utility for preliminary screening. PRIMITI is available on a reliable, scalable and user-friendly web server at https://biosig.lab.uq.edu.au/primiti.

bioinformatics↗

bioRxiv2024

On use of tertiary structure characters in hidden Markov models for protein fold prediction

While advances in protein structure prediction have opened up insights into arcane proteins, weak sequence homology makes functional characterisation challenging. To overcome this challenge, we use structure-based hidden Markov models of groupings in SCOP, CATH and ECOD to predict folds in proteins and thereby infer function. Conservation of structure and ability of hidden Markov models to detect remote signals make this a powerful resource for complete characterisation of arcane proteins.

bioinformatics↗

bioRxiv2024

Targeting the Plasmodium falciparum UCHL3 ubiquitin hydrolase using chemically constrained peptides

The ubiquitin-proteasome system is essential to all eukaryotes and has been shown to be critical to parasite survival as well, including Plasmodium falciparum, the causative agent of the deadliest form of malarial disease. Despite the central role of the ubiquitin-proteasome pathway to parasite viability across its entire life-cycle, specific inhibitors targeting the individual enzymes mediating ubiquitin attachment and removal do not currently exist. The ability to disrupt P. falciparum growth at multiple developmental stages is particularly attractive as this could potentially prevent both disease pathology, caused by asexually dividing parasites, as well as transmission which is mediated by sexually differentiated parasites. The deubiquitinating enzyme PfUCHL3 is an essential protein, transcribed across both human and mosquito developmental stages. PfUCHL3 is considered hard to drug by conventional methods given the high level of homology of its active site to human UCHL3 as well as to other UCH domain enzymes. Here, we apply the RaPID mRNA display technology and identify constrained peptides capable of binding to PfUCHL3 with nanomolar affinities. The two lead peptides were found to selectively inhibit the deubiquitinase activity of PfUCHL3 versus HsUCHL3. NMR spectroscopy revealed that the peptides do not act by binding to the active site but instead block binding of the ubiquitin substrate. We demonstrate that this approach can be used to target essential protein-protein interactions within the Plasmodium ubiquitin pathway, enabling the application of chemically constrained peptides as a novel class of anti-malarial therapeutics.

biophysics↗

bioRxiv2023

A broad-spectrum family GH13 α-glucosidase from Marinovum sp., a member of the Roseobacter clade

Glycoside hydrolases (GHs) are a diverse group of enzymes that catalyze the hydrolysis of glycosidic bonds. The Carbohydrate-Active enZymes (CAZy) classification organizes GHs into families based on sequence data and function, with fewer than 1% of the predicted proteins characterized biochemically. Consideration of genomic context can provide clues to infer possible enzyme activities for proteins of unknown function. We used the MultiGeneBLAST tool to discover a gene cluster in Marinovum sp., a member of the marine Roseobacter clade, that encodes homologues of enzymes belonging to the sulfoquinovose monooxygenase pathway for sulfosugar catabolism. This cluster lacks a gene encoding a classical family GH31 sulfoquinovosidase candidate, but which instead includes an uncharacterized family GH13 protein (MsGH13) that we hypothesized could be a non-classical sulfoquinovosidase. Surprisingly, recombinant MsGH13 lacks sulfoquinovosidase activity and is a broad spectrum -glucosidase that is active on a diverse array of -linked disaccharides, including: maltose, sucrose, nigerose, trehalose, isomaltose, and kojibiose. Using AlphaFold, a 3D model for the MsGH13 enzyme was constructed that predicted its active site shared close similarity with an -glucosidase from Halomonas sp. H11 of the same GH13 subfamily that shows narrower substrate specificity.

biochemistry↗

bioRxiv2022

epitope1D: Accurate Taxonomy-Aware B-Cell Linear Epitope Prediction

The ability to identify B-cell epitopes is an essential step in vaccine design, immunodiagnostic tests, and antibody production. Several computational approaches have been proposed to identify, from an antigen protein, which residues are likely to be part of an epitope, but have limited performance on relatively homogeneous data sets and lack interpretability, limiting biological insights that could be derived. To address these limitations, we have developed epitope1D, an explainable machine learning method capable of accurately identifying linear B-cell epitopes, leveraging two new descriptors: a graph-based signature representation of protein sequences, based on our well established CSM (Cutoff Scanning Matrix) algorithm and Organism Ontology information. Our model achieved Area Under the ROC curve of up to 0.935 on cross-validation and blind tests, demonstrating robust performance and outperforming state-of-the-art tools. epitope1D has been made available as a user-friendly web server interface and API at http://biosig.lab.uq.edu.au/epitope1d.

bioinformatics↗