Latest bioRxiv papers

Category: bioinformatics — Showing 50 items

• Modelling the impacts of imports and non-native subspecies hybridisation in honeybees

  de Carlos, I.; Strachan, L.; McCormack, G. P.; Gorjanc, G.; Obsteter, J.

  Abstract

  Human-mediated movement of organisms for agriculture and ecosystem services often results in hybridisation and introgression between populations of native and non-native species. While introgression may increase genetic diversity, it can erode unique adaptations and reduce fitness, threatening the survival of native lineages. Honeybees offer a good model with extensive records, queen trade and migratory beekeeping facilitating genetic exchange among subspecies. To explore these dynamics, we used SIMplyBee to simulate hybridisation between populations of native Apis mellifera mellifera and non-native A. m. carnica. We adopted the parameters from the Irish honeybee population that maintains relatively low levels of import, but is threatened by commercial imports. The model included colony honey yield and fitness as complex polygenic traits. We simulated varying import rates, genetic correlations between fitness in native and non-native environments, and spatial distributions of introgression over 20 years, measuring levels and rate of introgression and genetic means for both traits. Increased imports accelerated introgression and induced a trade-off between higher honey yield and lower fitness, and decreasing genetic correlations between environments amplified fitness decline. Spatial simulations showed the spread of introgression across the entire simulated area. Halting imports reversed the trend, but purging of introgressed material was slow and varied among replicates. These findings highlight the trade-off between short-term production gains and long-term losses in fitness and adaptation. Our modelling framework provides a reference for exploring introgression in other systems, emphasising that sustainable management of introgression requires restricting imports and breeding locally adapted populations rather than relying on non-native imports.

  bioinformatics2026-07-09v3

• Emergence of Biological Structural Discovery in General-Purpose Language Models

  Wang, L.

  Abstract

  Large language models (LLMs) are evolving into engines for scientific discovery, yet the assumption that biological understanding requires domain-specific pre-training remains largely unchallenged. Here we report that general-purpose LLMs possess an emergent capability for biological structural discovery. Under strict, shortcut-controlled evaluation, a small-scale GPT-2 (124M) fine-tuned solely on English paraphrase discrimination detects protein homology zero-shot at ROC-AUC 0.79 on a shortcut-controlled benchmark. Controls establish that the ability is conferred by pre-training, not architecture: a randomly initialized GPT-2 is at chance (0.52). To exclude the possibility that public checkpoints were contaminated with biological data, we train our own GPT-2 from scratch on an English-only web corpus; it reproduces the transfer (0.76), proving the effect arises from linguistic pre-training alone. Network-based interpretability reveals a deep structural isomorphism: the discriminative signal localizes to deep layers (0.97 at layer 9), and attention analysis surfaces modality-agnostic "difference" operators. Scaling to massive instruction-tuned models further improves performance, including in the remote-homology "twilight zone", which we report as an exploratory upper bound because those models' training corpora are undisclosed. We formalize these tasks through the BioPAWS benchmark. Our controlled results, obtained entirely on models with known training data, establish that abstract logical structures distilled from human language constitute a genuine, if bounded, cognitive prior for decoding the syntax of biology.

  bioinformatics2026-07-09v2

• CircDiscoverer: A multispecies comprehensive resource for circRNA-protein interactions and RNA modification landscapes

  Srinivasan, S.; Kumar, S.; Chatterjee, S.; Chande, A.

  Abstract

  Circular RNAs (circRNAs) regulate various cellular processes by interacting with miRNAs, proteins, and by encoding peptides. Compared to circRNA-miRNA interactions, circRNA-protein interactions (CPIs) and circRNA modifications have been relatively less explored. Here, we developed a comprehensive user-friendly web resource, CircDiscoverer, to explore CPIs and modification landscapes across multiple species, including Homo sapiens, Mus musculus, Drosophila melanogaster, and Arabidopsis thaliana. CircDiscoverer integrates manually curated literature-supported CPIs with computationally predicted interactions. Furthermore, it offers comprehensive insights into protein-binding profiles and detailed information on interacting circRNAs, including qRT-PCR primer sequences, guide RNAs for experimental validation and potential RNA modifications, thereby providing an integrated platform for further research. CircDiscoverer is available at https://aclab.iiserb.ac.in.

  bioinformatics2026-07-09v2

• VicMAG, an open-source tool for visualizing circular metagenome-assembled genomes highlighting bacterial virulence and antimicrobial resistance

  Tsuda, Y.; Tanizawa, Y.; Vu, T. M. H.; Nishimura, Y.; Shintani, M.; Abe, H.; Hasebe, F.; Kasuga, I.; Nagao, M.; Suzuki, M.

  Abstract

  Bacterial pathogens spread in clinical and environmental settings, and mobile genetic elements (MGEs), such as plasmids and phages, mediate the transfer of virulence factor genes (VFGs) and antimicrobial resistance genes (ARGs) among bacterial communities. Metagenomic analysis of environmental and wastewater samples using highly accurate long-read sequencing technologies, such as PacBio HiFi sequencing, provides valuable insights into monitoring the regional spread of VFGs and ARGs, including dissemination mediated by MGEs. No visualization tool is currently available for the comprehensive display of numerous resulting circular metagenome-assembled genomes (cMAGs) with functional gene annotations. Here, we developed VicMAG, a visualization tool for highly complex cMAGs derived from long-read metagenome assemblies annotated using updated databases of VFGs, ARGs, and MGEs. Using 353 cMAGs from PacBio HiFi sequencing of a wastewater sample, we demonstrated the utility of VicMAG for metagenome visualization. VicMAG provides comprehensive, size-aware visualization of cMAGs representing bacterial chromosomes and plasmids, annotated with VFGs, ARGs, and phages. By simultaneously visualizing all cMAGs in a framework, VicMAG facilitates a holistic understanding of the distribution and genomic context of VFGs and ARGs across complex microbial communities. This tool supports integrated surveillance of bacteria associated with virulence and antimicrobial resistance across clinical, environmental, and One Health contexts.

  bioinformatics2026-07-09v2

• Expanding the Landscape of Disordered Flexible Linkers: A Structural and Computational Framework for DLD dataset assembly

  meng, d.; Glavina, J.; Garcia Alvarez, H. M.; Leonetti, C. O.; Pollastri, G.; Chemes, L. B.

  Abstract

  Disordered flexible linkers (DFLs) are functional elements found within intrinsically disordered regions that carry out key functions by connecting domains and/or short linear motifs. Understanding the features of DFLs is limited by the lack of comprehensive datasets and accurate predictive models. In this study, we propose a classification for DFLs that includes linkers joining two domains (DLD), a domain and a motif or two short linear motifs. We developed a workflow that allows the systematic identification of DLD-type linkers from protein structures and created a comprehensive dataset known as the DLD dataset. The DLD dataset includes 1640 independent domain linkers (IDLs) which expands currently available linker datasets and annotates related regions such as dependent-domain linkers, intra-domain loops, and termini. Our data collection process integrates missing residue completion and smoothing of short secondary structure stretches enabling to capture a higher number of longer IDLs. We assessed the features of IDLs using t-SNE analysis and protein language model embedding with a CNN-based classifier as well as PCA analysis. IDLs can be distinguished from other disordered and folded protein regions, and their features highly overlap with DisProt Linkers, considered the gold standard for linker annotation. The DLD dataset offers a valuable resource for researchers seeking to investigate the features of disordered flexible linkers and to improve the accuracy and generalizability of DFL predictive models. The DLD dataset is available via an interactive web server at https://dld.chemeslab.org/ where linkers are annotated with sequence and structural features and can be visualized using a structure viewer.

  bioinformatics2026-07-09v2

• LeafRank: A phylodynamic framework for inferring relative fitness from single-cell phylogenies in chromosomally unstable tumors

  Wu, C.; Leder, K.; Wang, Z.; Sun, R.

  Abstract

  Tumors contain cancer cells with diverse growth potentials that shape evolutionary trajectories, yet this fitness diversity remains difficult to quantify in cases of whole-genome duplication (WGD) and chromosomal instability. We present LeafRank, a mathematical framework that leverages single-cell DNA-seq phylogenies to infer the relative fitness of individual cells. Using a multi-type branching process model, LeafRank integrates full tree topology, including branch lengths and bifurcation patterns, to estimate marginal fitness probabilities under punctuated evolutionary regimes driven by rare driver events. To account for elevated aberration rates following WGD, we introduce a tree-rescaling strategy that adjusts for lineage-specific genomic instability. Unlike methods focused on predefined subclones, LeafRank ranks all sampled cells, enabling flexible assessment of growth heterogeneity. Simulations demonstrate high accuracy across spatial and non-spatial virtual tumors. Applied to ovarian cancer, LeafRank reveals directional and parallel selection in WGD tumors and identifies recurrent copy number events enriched in high-fitness lineages. WGD lineages do not show immediate growth advantages but acquire fitness through subsequent alterations.

  bioinformatics2026-07-09v1

• Coding agents author interpretable single-cell embedding models from the literature

  Brunn, N.; Krissmer, S. M.; Frosch, M.; Frick, M.; Prinz, M.; Binder, H.

  Abstract

  The single-cell literature catalogs cell states as validated marker-gene programs - a sparse, compositional prior. Conventional embedding methods do not leverage this prior and learn cell-state structure de novo from the expression matrix, producing dense dimensions needing post-hoc interpretation and batch correction. Here we show coding agents can author single-cell embedding models directly from the literature. Given a scenario that focuses this literature lens on a chosen biological subdomain, the agent edits a structured Python template, curating named, literature-cited gene programs and composing them into axes, without a gene-set database, training, or sight of the data. Across mouse and human tissues these zero-shot embeddings are competitive in biological quality with conventional, foundation-model, and program-informed baselines, batch-robust by construction and reproducible across runs, complementing data-driven embeddings. Because each dimension is a named, cited gene program, the embedding is interpretable and auditable, and its composable axes can be steered into a developmental tree.

  bioinformatics2026-07-09v1

• Mind the Alignment Gap: A Spatial Transcriptomics Benchmark for Scientific Coding Agents

  Chen, Y. T.; Hicks, S. C.

  Abstract

  Scientific coding agents are difficult to benchmark because many research tasks require executable work yet produce ambiguous or hard-to-verify outputs. Because benchmark construction requires substantial time and resources, automation offers a path to accelerating methods evaluation. We introduce an interactive framework for constructing scientific-agent benchmarks from peer-reviewed papers and diagnosing agent behavior through trace inspection. We apply it as a case study in spatial transcriptomics alignment, constructing 40 tasks from SABench in which agents submit coordinate tables aligning pairs of two-dimensional tissue slices. Across 120 runs and three configurations, we compare a basic prompt, a package-aware prompt, and a full prompt with a prebuilt virtual environment. In this setting, richer package and environment context increased tool exploration but reduced the mean alignment score relative to the basic prompt (0.36 vs. 0.43; 95% CI, [-0.11,-0.03]). Trace inspection showed that added scaffolding often induced unnecessary transformations, fragile package-first workflows, and infrastructure failures. These results illustrate how specialized tooling can alter agent behavior and why scientific-agent benchmarks should evaluate agent traces and the workflows that produce them in addition to the final outputs.

  bioinformatics2026-07-09v1

• Rectangle: robust and scalable multiscale deconvolution informed by single-cell RNA sequencing data

  Eder, B.; Rigato, I.; Dietrich, A.; Merotto, L.; Sturm, G.; Treis, T.; List, M.; Theis, F.; Finotello, F.

  Abstract

  Bulk RNA-seq enables effective profiling of large cohorts and complex experimental designs, but current single-cell-informed deconvolution methods incompletely resolve closely related cell phenotypes, do not scale efficiently to large single-cell datasets, or fail to account for cellular content not represented in the reference. Here, we present Rectangle, an scverse Python framework for single-cell-informed deconvolution of bulk RNA-seq data. Rectangle combines multiscale deconvolution, capturing cellular composition across multiple resolution levels, with explicit modeling of unknown cellular content. In a diverse, cross-method benchmark, Rectangle achieved consistently strong performance across all evaluated metrics, demonstrating high accuracy, high resolution, low spillover, strong scalability and efficiency, and robustness to unknown cellular content. By bridging the resolution of single-cell transcriptomics with the scale and cost-efficiency of bulk RNA-seq, Rectangle enables cell-type and cell-state profiling at scale, supporting population-scale cellular biomarker discovery and tracking of cellular dynamics in settings impractical for comprehensive single-cell sequencing.

  bioinformatics2026-07-09v1

• Gene Program Negotiation Defines Cellular Identity in Single-Cell Transcriptomes

  Sung, J.-Y.; Cheong, J.-H.

  Abstract

  Single-cell transcriptomics has transformed the characterization of cellular heterogeneity by enabling systematic analysis of biological gene programs. However, existing computational approaches primarily quantify the activity of individual programs independently and therefore provide limited insight into how multiple simultaneously active programs collectively determine cellular identity. Here we present Gene Program Negotiation (GPN), a graph-based computational framework that models regulatory decision-making among concurrently active biological programs. GPN reconstructs cell-specific program interaction networks from local transcriptional neighborhoods and quantifies regulatory organization using the Gene Program Coherence Index (GPCI) together with measures of local regulatory conflict, program diversity, and dominance. These graph-derived properties enable the classification of individual cells into five regulatory decision states: Consensus, Competition, Negotiation, Dominance, and Low activity. Applying GPN to gastric cancer single-cell transcriptomes revealed that cells sharing the same dominant biological program frequently occupied distinct regulatory decision states, demonstrating that dominant program identity alone does not uniquely define cellular regulatory organization. Competition states consistently exhibited elevated local regulatory conflict and were preferentially enriched among transition-like cells, indicating that regulatory competition is closely associated with transcriptional plasticity. Independent validation using glioblastoma single-cell transcriptomes reproduced these regulatory patterns without modification of the computational framework, supporting the robustness and generalizability of the approach across biologically distinct malignancies. These findings establish regulatory negotiation as an additional layer of cellular organization beyond conventional gene-program activity analysis. By explicitly modeling interactions among simultaneously active biological programs, GPN provides a general computational framework for investigating regulatory coordination, cellular plasticity, and dynamic cell-state organization in single-cell transcriptomic data.

  bioinformatics2026-07-09v1

• IgGM2: An All-Atom Foundation Model for Adaptive Immune Receptor Design

  Ma, J.; Wu, F.; Yao, L.; Gao, J.; Wang, R.; Li, Q.; Yang, N.; Jiang, S.; Huang, D.; Pan, X.; Zhu, Y.; Hou, T.; Yao, J.; Yan, J.

  Abstract

  Accurate immune receptor design requires modeling the coupled variation of amino-acid sequence, full-atom conformation, and target-binding geometry across antibodies, nanobodies, and T-cell receptors (TCRs). Existing methods often address only part of this problem, either by separating structure generation from sequence design, relying on fixed-backbone inverse folding, or focusing on a single receptor class. We introduce IgGM2, a unified all-atom generative framework for immune receptor structure prediction and CDR sequence-structure co-design. IgGM2 follows a structure-to-design strategy: it first learns how immune receptors are positioned around fixed target structures, and then transfers this target-conditioned structural prior to CDR design. Unlike modular design pipelines, IgGM2 jointly generates CDR residue identities and full-atom receptor structures, allowing framework geometry to adapt to designed CDRs without separate inverse folding or external sidechain packing. Unlike continuous residue encodings based on virtual-atom geometry, IgGM2 keeps sequence prediction explicit while using atom14 placeholders only for full-atom representation. On structure prediction benchmarks, IgGM2 better captures receptor-target spatial relationships than AlphaFold3 on FoldBench and achieves strong performance on TCR-pMHC modeling. On sequence design benchmarks, IgGM2 improves amino-acid recovery and Rosetta-based interface preference metrics, suggesting more favorable generated binding interfaces. These results support IgGM2 as a unified all-atom framework for adaptive immune receptor structure prediction and design.

  bioinformatics2026-07-09v1

• Gene-specific exponent-corrected normalization for library size in bulk RNA-seq

  Yin, R.; Li, D.; Zong, W.; Ketchesin, K. D.; Seney, M. L.; McClung, C. A.; Baldoni, P. L.; Tseng, G. C.

  Abstract

  Correcting for library size is an essential step in bulk RNA-seq analyses, as differences in sequencing depth across samples can obscure biological signal with technical noise. While numerous normalization methods and model-based strategies have been proposed, we demonstrate here that library size-normalized counts and differential expression results obtained from such widely adopted approaches often remain strongly correlated with library size in large-scale RNA-seq experiments. Through a systematic analysis of over 100 publicly available GEO and TCGA RNA-seq datasets with raw count data, we show that library size association is observed for a substantial proportion of genes even after state-of-the-art library size correction approaches recommended by leading normalization tools. To address this issue, we propose gecco, a gene-specific exponent-corrected normalization method for RNA-seq counts that incorporates library size directly into the statistical framework via a gene-specific correction term, rather than applying a uniform adjustment factor across all genes. This formulation generalizes existing normalization approaches and yields normalized counts that are free of residual library size effects. Using both simulation studies and real large-scale RNA-seq datasets, we show that our method mitigates library size bias while preserving biological signal across a range of parameter settings. We further demonstrate that our approach leads to higher detection accuracy and more biologically meaningful pathway enrichment results in downstream differential expression and rhythmicity analyses without compromising false discovery rate control. Our method is implemented in R and is fully compatible with the widely used differential expression analysis methods DESeq2 and edgeR.

  bioinformatics2026-07-09v1

• A five-dimensional functional state space for fingerprinting disease transcriptomes

  Nie, F.; Zhuang, Y.; Chen, K.; Lin, J.; Sun, J.

  Abstract

  High-throughput transcriptomics has transformed disease biology, but its outputs often remain fragmented into gene and pathway lists that are difficult to compare across conditions or use for human-AI interpretation. We developed a five-dimensional (5-D) functional state space that represents disease transcriptomes as coordinated activity patterns across major biological systems. The framework maps transcriptomic signals onto five functional systems, 14 subcategories, and a distinct infrastructure layer, and was implemented as a reproducible pipeline for functional scoring, cross-condition profiling, benchmarking, and large language model (LLM)-assisted interpretation. Applied to wound healing, sepsis, colorectal cancer-related datasets, an extended GEO atlas of 38 complete case-control disease fingerprints spanning diverse disease contexts, and a TCGA-COAD/READ stage benchmark, the approach recovered interpretable disease-state patterns and retained progression-related information under strong compression. It also improved the quantitative grounding of LLM-generated summaries. This framework provides a compact and auditable representation for comparing disease transcriptomes and supporting human-AI biological interpretation.

  bioinformatics2026-07-09v1

• Generative AI Models Reveal Dynamic Views of Aging (DyViA) Phenotypes in Healthy Individuals

  Ray, D.; Ray, M.; Pyne, S.

  Abstract

  Background and objectives: In recent years, the need to develop analytical strategies for healthy aging has assumed great importance. In this study, we introduce DyViA, a generative artificial intelligence (genAI) platform that can construct personalized trajectories capable of predicting the plausible progression of selected phenotypes with advancing age. Research design and methods: DyViA presents a suite of deep learning models covering two major GenAI approaches: DyViA-Diff, a new diffusion model; and DyViA-mGAN, an improved version of a recent Generative Adversarial Network model. It demonstrated the dynamic progression of femoral neck bone mineral density (BMD) using data from a longitudinal cohort study of women in the U.S. of age 65 years or above. Results: Using very few initial measurements, DyViA generated individual-specific continuous trajectories of BMD, with a corresponding region of acceptable predictions, from 66 to 89 years. The results were subjected to rigorous quality-control and comparative analysis across multiple methods. While DyViA-Diff is the superior model with more coherent and accurate predictions, DyViA-mGAN allows for encoding population- and individual-level effects with a better control. Discussion and implications: Given the prevalence of osteoporosis in the aging population, the main impact of DyViAs genAI-driven contribution in the form of personalized, plausible models of BMD progression with age lies in the systematic yet rigorous transition from otherwise static models of inference about a clearly dynamic phenomenon to a continuous one. The foresight offered by DyViAs outputs empowers an individual by conferring a certain degree of strategic preparedness in the course of aging.

  bioinformatics2026-07-09v1

• Learning the spatial cell-cell communication network to decode multi-channel signaling and predict network-hub vulnerabilities with MOSANIC

  Das, D.; Mitra, P.

  Abstract

  Intercellular signaling governs the central decisions of tissue biology, from proliferation and immune recruitment to metabolic adaptation and cell death, and its dysregulation is a hallmark of disease. What matters most are properties of the signaling network as a whole rather than of individual interactions: the hubs that hold the network together and mark rational points of intervention, the relays through which a signal propagates across intermediate cells to reach partners it does not directly contact, the response of tissue-wide communication to the loss of a single node, and the metabolite-mediated axis that operates alongside secreted-protein signalling. Scoring known ligand-receptor pairs yields a ranked interaction list that captures none of these and excludes metabolite signalling. We present MOSANIC (Multi-mOdal Self-Attention Network for Intercellular Communication), which learns a tissue's communication network directly from spatial transcriptomics. MOSANIC represents each tissue as a single heterogeneous graph of cells, genes and metabolites, initialises every node with a frozen foundation-model representation (scVI, ESM-2 and ChemBERTa), and propagates these representations through a self-attention network over biologically typed edges. Supervision is restricted to spatial gene-expression prediction and excludes ligand-receptor annotation, rendering the inferred communication statistically independent of the reference databases used for evaluation. Across five spatial datasets spanning three platforms and two species, MOSANIC attains the highest accuracy on all eight independent ligand-receptor benchmarks (mean AUROC 0.756) against nine established methods, resolves a metabolite-receptor channel statistically orthogonal to the peptide channel, and reconstructs multi-step signal relays that concentrate within a compact rich-core of load-bearing hub genes and cells whose removal fragments the network far beyond a degree-preserving null. In-silico knockout of these hubs recovers experimentally reported phenotypes, and, given no prior oncological input, MOSANIC nominates SCARF1 as a previously unrecognised communication hub in breast cancer whose elevated expression predicts significantly worse survival in an independent cohort after adjustment for tumour stage and age (hazard ratio 1.17, P = 0.043). MOSANIC is released as an open-source Python package (mosanic-ccc).

  bioinformatics2026-07-09v1

• Metabolic Rewiring in Triple-Negative Breast Cancer: Systems Analysis of TCGA-BRCA Transcriptome Reveals Prognostic Hub Genes

  Chandrasekar, S.

  Abstract

  Triple Negative Breast Cancer (TNBC) is the deadliest and most aggressive subtype of breast cancer, with poor prognosis and high rates of metastasis. Despite knowledge of metabolic rewiring in TNBC, the systems-level coordination of these adaptive pathways remains unmapped. This integrative systems-level analysis reveals key metabolic hub genes and identifies ATP1A2 as a significant prognostic marker. Analysis identified 764 differentially expressed genes, with 89 enriched biological processes predominantly involving metabolic pathways. Co-expression network analysis of 261 genes identified metabolic hub genes including LEP, ADIPOQ, and ATP1A2. To evaluate the prognostic framework, survival analysis of the top 10 hubs was performed on synthetic survival data, revealing ATP1A2 as a significant marker (p = 0.03) under Cox regression, with elevated expression associating with altered survival outcomes. By systematically mapping metabolic rewiring in TNBC, this work identifies ATP1A2 as an actionable therapeutic target and establishes a systems-level framework for rational drug discovery and patient stratification in this aggressive malignancy.

  bioinformatics2026-07-09v1

• Assessing tensor decomposition quality of immune profiling data from a dictionary learning perspective

  Konstorum, A.; Xing, J.; Aeron, S.; Kilmer, M.; Kleinstein, S.

  Abstract

  Systems-level immune profiling data arising from longitudinal studies of vaccination or infection has an inherent multi-index array structure. While tensor decomposition of such datasets has gained popularity, choosing a rank and trial for a decomposition is not straightforward. We show that taking into account the experimental data model can inspire the development of new metrics to assess the quality of a Non-negative CANDECOMP/PARAFAC (NCPD) decomposition, and can thus be used to choose a rank and trial for the decomposition. Moreover, we show how framing the results via a dictionary learning framework can better enable interpretation of the components of the decomposition.

  bioinformatics2026-07-09v1

• qg: Configuration-Driven, Multi-Vendor Acquisition Queue Generation with Reproducible Run-Order and QC Control for Mass Spectrometry

  Wolski, W. E.; Schwarz, L.; Trachsel, C.; Zanella, M.; Riedi, C.; Schlapbach, R.; Othman, A.; Tuerker, C.; Nanni, P.; Panse, C.

  Abstract

  Mass spectrometry laboratories must turn lists of submitted samples into acquisition queues. The run order and the placement of quality-control (QC) injections determine whether a design controls batch effects and signal drift, and whether those effects stay correctable afterward. Yet operators usually set them by hand in vendor worklist editors that neither randomize run order nor offer configurable, pattern-driven QC. We present *qg*, an open-source tool that builds acquisition queues with systematic run-order handling: four run-order modes (none, simple, blocked/randomized-complete-block, and group-uniform blocked), pattern-driven QC and standard injections, and sampler- and plate-aware positioning. Unlike plate-design tools that stop at a generic sample sheet, *qg* writes the native vendor acquisition file directly, for three instrument ecosystems (Thermo Fisher XCalibur, Axel Semrau Chronos, Bruker HyStar) across proteomics, metabolomics, and lipidomics. It separates a small, stateless generation pipeline from a declarative configuration layer, so a laboratory adapts instruments, QC patterns, layouts, and naming by editing version-controlled configuration through a validating editor rather than changing code. *qg* runs from a reactive web interface or a scripted command-line interface, integrated with a LIMS (B-Fabric) or standalone from uploaded tables; randomized runs record their seed and reproduce from exported parameters. On an unbalanced design, group-uniform blocked randomization spreads biological groups evenly across acquisition time, whereas textbook block randomization leaves a tail of the largest group and can track acquisition time worse than a plain shuffle. *qg* is released under the Apache-2.0 license.

  bioinformatics2026-07-09v1

• Thematic Shifts in Early-High-Impact Cancer Genomics and Diagnostics Research: A Bibliometric and Semantic Analysis

  Su, Z.; Li, T.

  Abstract

  Cancer genomics and diagnostics is a rapidly evolving field in which identifying which topics attract early citation prominence can inform laboratory investment, clinical translation, and research strategy. We developed a bibliometric framework to identify and characterize the most influential recent publications in this domain across two consecutive annual cohorts. Using a mathematically exact threshold-expansion algorithm, we ranked over 10,000 OpenAlex-indexed research articles per cohort by 18-month post-publication citation count. Large language model (LLM)-based topical relevance filtering yielded 50 substantively on-topic papers per cohort (100 total). LLM-based concept extraction and a two-stage, embedding-guided normalization pipeline produced 1,853 canonical concepts organized into 103 parent themes, enabling structured cross-cohort comparison of paper-level concept prevalence. The most cited papers in both cohorts were large-scale genomic infrastructure resources rather than single-disease mechanistic studies. Between consecutive cohorts, normalized frequencies increased most for whole-genome sequencing, tumor microenvironment biology, molecular biomarkers, and cancer pharmacotherapy, while liquid biopsy-related themes showed the largest declines. These findings indicate that early citation impact in cancer genomics is shifting toward integrative, population-scale, and microenvironment-aware research, and demonstrate that LLM-augmented citation ranking provides a replicable, semantically enriched lens for monitoring thematic evolution in precision oncology. A web interface for exploring the results is available at https://pri.pepkio.com/.

  bioinformatics2026-07-09v1

• BertST: BERT-based Spatial Domain Identification in Patient Data

  Nnadi, G. O.

  Abstract

  Spatial transcriptomics enables the study of gene expression within its native tissue context, providing critical insights into cellular organization and microenvironment-driven biological processes. A key challenge in this field is spatial domain identification, which aims to partition tissue into coherent regions by jointly leveraging gene expression and spatial information. Existing approaches are predominantly based on Graph Neural Networks (GNNs), and approach based on Transformers particularly, Bidirectional Encoder Reppresentation Transformer (BERT) model for modelling both local and long-range dependencies remains largely unexplored. In this work, we propose BERT for Spatial Transcriptomics (BertST), a transformer-based framework that reformulates spatial transcriptomics as a graph-to-text representation learning problem. Building upon the BERTwalk paradigm, we construct a task-specific multi-graph representation integrating spatial adjacency, pruned gene-expression similarity, and a fully connected gene-expression graph. This design enables the modelling of both local spatial structure and global molecular relationships. Random walks over these graphs are treated as sequences, allowing a BERT model to learn contextualised node embeddings. To further enhance representation quality, we introduce a hierarchical multi-graph propagation strategy, where embedding refinement is performed sequentially: first on the fully connected graph to capture global structure, followed by the pruned graph to refine molecular relationships, and finally on the spatial graph to enforce local smoothness. This ordering ensures that global information is effectively distributed and progressively constrained by biologically meaningful neighbourhoods. We also improve computational efficiency by leveraging \textit{PecanPy}, a fast and scalable implementation of node2vec, enabling efficient random walk generation on dense graphs. Experimental results on multiple 10x Visium datasets, including DLPFC and Human Breast Cancer, demonstrate that BertST consistently outperforms or matches GNN-based methods such as ConST, CCST, and SpaceFlow in terms of Adjusted Rand Index (ARI) and Adjusted Mutual Information (AMI). Overall, BertST highlights the potential of transformer-based architectures for spatial omics analysis by effectively capturing both local and long-range spatial-molecular dependencies, offering a promising alternative to traditional graph-based methods.

  bioinformatics2026-07-09v1

• EZSolver: Template-free prediction of polar enzymatic mechanisms via bidirectional flow matching and search

  Kuo, L.-H.; Yang, J.; Arnold, F.

  Abstract

  Predicting enzymatic reaction mechanisms is critical for understanding enzyme function and for designing and dis-covering new enzymes. Current computational predictors rely on deterministic, rule-based dictionaries, which per-form well on in-distribution tasks but fail to generalize to out-of-distribution (OOD) chemistry. To address this limita-tion, we present EZSolver, a template-free, generative framework for polar enzymatic mechanism prediction. Powered by a flow matching predictor (EZFlow) and navigated by an evaluator-guided bidirectional beam search, EZSolver learns the chemistry of electron redistribution instead of memorizing rigid templates. Evaluated across diverse en-zyme classes, EZSolver achieves a 60.0% accuracy and an 84.6% chemical plausibility rate for full mechanism predic-tion of unseen polar enzymatic reactions. While rule-based models collapse without predefined templates, EZSolver successfully extrapolates chemical knowledge to infer uncatalogued pathways, as demonstrated during rigorous OOD benchmarking. By illuminating enzymatic chemical mechanisms, EZSolver helps pave the way for automated predic-tion of enzyme function and discovery and design of novel biocatalysts for sustainable chemistry.

  bioinformatics2026-07-09v1

• A Spatio-Temporal Analysis Framework for Characterizing Radiation-Induced Genomic Instability

  Chopra, K.; Cucinell, C.; Titov, M.; Weinberg, R.; Forrester, S.; Kilic, O.; Zhu, Y.; Turilli, M.; Jha, S.; Schabacker, D. S.; Brettin, T.; Yoon, B.-J.

  Abstract

  Chronic low-dose ionizing radiation induces complex genomic instability encompassing both structural variants and point mutations, yet these alterations are typically analyzed as independent events limiting detection of mechanistic coupling between rearrangement formation and localized mutagenesis at breakpoint junctions. This gap is particularly consequential given the widespread occupational and environmental exposure contexts; nuclear energy, medical imaging, and environmental contamination, where coupled genomic alterations may contribute to cancer risk through mechanisms invisible to type-agnostic analyses. We developed an integrated analytical framework combining temporal pattern tracking, breakpoint-proximal mutation enrichment analysis, and systematic testing across all structural variant types to resolve these coupled dynamics across dose and time. Applying this framework to whole-genome sequencing data from primary human endothelial cells (HUVEC) exposed to chronic low-dose gamma radiation (0.001 - 2 mGy/hr) over three weeks, we discovered 7.13-fold enrichment of doublet base substitutions (DBS) within 10bp of inversion breakpoints, a signal absent from other structural variant types. This enrichment decayed sharply with distance (to [~]1.9 fold at 100bp), indicating localized mutagenesis at these junctions. Temporal analysis revealed divergent fates: inversions appeared transiently (100% single-timepoint) while DBS showed greater persistence (9.0% multi-timepoint). Among the INV-DBS events identified, affected genes include 16 high-constraint loci (pLI [≥] 0.9) involved in DNA damage response, signal transduction, and chromatin regulation; pathways critical for maintaining genomic stability. Our framework provides a generalizable approach for investigating structural variant-mutation relationships, with applications to radiation biology, cancer genomics, and mechanistic studies of DNA repair fidelity.

  bioinformatics2026-07-08v2

• Modeling patient tissues at molecular resolution with Eva

  Liu, Y.; Sharma, R.; Bieniosek, M.; Kang, A.; Wu, E.; Chou, P.; Li, I.; Rahim, M.; Bauer, E.; Ji, R.; Duan, W.; Qian, L.; Luo, R.; Sharma, P.; Dhanasekaran, R.; Schürch, C. M.; Charville, G.; Mayer, A.; Zou, J.; Trevino, A. E.; Wu, Z.

  Abstract

  Tissue structure is essential to function and homeostasis in all organs, and disruptions to structure usually indicate disease. Modeling relationships between structural, molecular, and clinical aspects of tissues could advance new diagnostics and treatment strategies. Although profiling techniques like spatial proteomics can capture these relationships, the data remain challenging to extract insight from. Here, we present Eva, a foundation model for tissue imaging data that learns multi-scale spatial representations of tissues at the molecular, cellular, and sample level. Eva uses a novel vision transformer architecture and is pre-trained on masked reconstruction of matched spatial proteomics and histopathology images. We show that Eva excels at a variety of tasks, including cross-modal inference, quality control, data annotation, zero-shot retrieval, survival modeling, and patient stratification. Extensive evaluations on held-out validation data demonstrate the versatility and generalizability of the learned embeddings. We anticipate that Eva will accelerate translational science by bridging basic research and clinical practice.

  bioinformatics2026-07-08v2

• Decoding UTRs by applying explainable AI to a genomic foundation model

  Brase, L.; Creamer, D. R.; Shapovalova, Y.; Ashe, M. P.; Ashe, H. L.; Rattray, M.

  Abstract

  Background The regulation of mRNA decay and translation is crucial for cellular function and development; however, the complex interplay of RNA-binding proteins (RBPs) regulating these processes remains incompletely understood. Recent advances in genomic foundation models present new opportunities for decoding the regulatory grammar embedded within mRNA untranslated regions (UTRs). Here, we leverage explainable artificial intelligence to systematically identify RBP motifs that influence translation and mRNA decay during Drosophila melanogaster development. Results We extended the training of GENA-LM Fly, a genomic foundation model, on all 5' and 3' UTR pairs from the D. melanogaster genome. We separately fine-tuned it using ribosome density (RD) and mRNA decay (half-life) data from the embryonic maternal-to-zygotic transition (MZT). Using SHapley Additive exPlanations (SHAP) analysis, we identified the sequence regions most influential for prediction and performed motif enrichment analysis to discover associated RBP binding sites. We identified 42 unique RBPs associated with increasing (n=23; e.g., Rnp4f, Mxt) or decreasing (n=19; e.g., Aret/Bruno, Rox8, Sxl, Orb2) RD and 18 unique RBPs associated with increasing (n=6) and decreasing (n=12; e.g., Cnot4, Rbp9, Rox8) mRNA half-life. Using publicly available PAR-CLIP data, we validated our Orb2 signal in a Drosophila cell line. Furthermore, feature ablation and shuffling experiments revealed the contributions of different sequence components to model performance. Our approach significantly outperformed naive high-versus-low RD comparisons, demonstrating the power of model explainability in biological discovery. Conclusions This study demonstrates that genomic foundation models, when combined with explainability methods, can discover meaningful biology even without drastically improving the underlying prediction accuracy. The identified RBP motifs provide new insights into post-transcriptional regulatory elements that govern RD and decay during early development.

  bioinformatics2026-07-08v2

• Design of a Multi-epitope Vaccine Against Human Glanders Targeting Outer Membrane β-barrel Proteins of Burkholderia mallei

  Kapoor, J.; Panda, A.; Kumar, S.; Bandyopadhyay, A.

  Abstract

  Burkholderia mallei, a facultative intracellular Gram-negative pathogen, is the causative agent of glanders that primarily affects solipeds and sporadically transmitted to humans. Current interventions mainly rely on antibiotics; however, increasing resistance and the lack of a licensed vaccine further complicate disease management. In the present study, a consensus-based computational framework was employed on the B. mallei turkey2 proteome. Total 59 proteins - including porins, TonB receptors, autotransporters, and efflux components - were identified as surface exposed outer membrane {beta}-barrel (OMBB) proteins that were used to design a multi-epitope vaccine (MEV) construct. B- and T-cell epitopes were predicted from 59 proteins, and ten epitopes each of cytotoxic T-lymphocyte (CTL), helper T-lymphocyte (HTL), and B-cell were chosen based on their antigenicity, non-allergenicity, non-toxicity, surface accessibility, and conservation across 32 B. mallei strains. The MEV was included with suitable adjuvants at the N-terminus to enhance its immunogenicity. The 780 amino acid MEV construct was predicted to be antigenic, and soluble upon overexpression with 62.69% random coils, while the rest formed -helices and {beta}-strands. The tertiary structure of the MEV was generated and subsequently validated, indicating good structural quality. Molecular docking of the MEV with toll-like receptor 4 (TLR4) demonstrated strong affinity, and molecular dynamics simulation confirmed the structural stability of the MEV-TLR4 complex. In-silico immune simulation showed the capability of MEV to induce a strong immune response. Codon optimization and in-silico cloning were performed for efficient protein expression in the E. coli host. The study proposes an MEV construct by utilizing surface exposed OMBB proteins which directly interact with the host and serve as effective immunogenic targets against B. mallei infection.

  bioinformatics2026-07-08v2

• A systematic analysis of machine learning pipelines for robust antimicrobial resistance prediction

  Aselstyne, A.; Karthik, E. N.; El Azami, M.; Pogorelcnik, R.; Fournier, Q.; Chandar, S.

  Abstract

  Motivation: Antimicrobial resistance (AMR) has been identified as a top global public health threat. Accurate AMR phenotype prediction from whole-genome sequencing data is an essential tool for accelerating clinical decision-making and mitigating resistance spread. Although many previous works have explored the use of tree-based machine learning (ML) models to predict resistance, the field lacks a systematic evaluation of the training pipeline across a variety of pathogenic species and antibiotics. Results: Using nine clinically relevant species-antibiotic combinations from the NCBI antimicrobial susceptibility testing database, we present a detailed analysis of the ML pipeline and identify key factors affecting model performance and evaluation. We begin by relabelling all isolates using current CLSI minimum inhibitory concentration breakpoints to resolve inconsistencies and increase available data, resulting in up to a 19% label swap and 56% data enlargement per species-antibiotic combination. We identify several key training parameters including k-mer length, which can increase classification F1 scores by over 20 points compared to commonly used k-values, feature matrix truncation, which can induce polynomial time reductions with limited performance reduction, and ML model class. By comparing 5-fold cross-validation with evaluation on an unseen clinical dataset, we show that random cross-validation splits--often criticized as overly optimistic--can act as a strong proxy for downstream clinical performance, yielding closer F1 scores than phylogeny-aware splits in all cases. We finally present an interpretability study which shows that over 95% of k-mers used by our models are associated with identifiable genomic features. Our results highlight the importance of feature design, evaluation protocol, and biological analysis in genomic AMR prediction, and support tree-based models as a robust and interpretable method.

  bioinformatics2026-07-08v1

• CSGDA: A Cell State-Guided Graph Domain Adaptation Network for Single-Cell Drug Response Prediction

  Yan, F.; Cao, X.; Mao, F.; You, Z.; Chen, Y.; Du, Z.; Huang, Y.-A.

  Abstract

  Intratumoral heterogeneity drives cancer recurrence and metastasis, yet single-cell drug response prediction faces severe "cross-domain" challenges, such as applying in vitro models to in vivo tissues or inferring metastatic resistance from primary tumors. These scenarios trigger distribution shifts arising from heterogeneous sequencing platforms, distinct tissue microenvironments, and metastatic evolution - problems rarely addressed by existing methods. We introduce CSGDA, a cell state-guided graph domain adaptation framework designed to predict drug responses across these biological heterogeneities. CSGDA incorporates biological priors to map gene expression into functional cell states, guiding a structure learning module to construct robust cell topology. To conquer distribution shifts, the model employs graph domain adaptation combined with a novel overlap penalty mechanism. Extensive benchmarks on five scRNA-seq datasets demonstrate that CSGDA outperforms state-of-the-art methods, achieving an average gain of ~6% in ACC and AUPR. Beyond prediction accuracy, we employed integrated gradients to effectively pinpoint key genes involved in drug resistance within a challenging cross-metastasis cisplatin dataset. These findings underscore CSGDA's superior performance in single-cell drug response prediction and its potential in resolving single-cell heterogeneity, paving the way for precision medicine.

  bioinformatics2026-07-08v1

• Interpretable and scalable spatial gene set activity analysis with GESSO uncovers functional tissue architecture

  Yang, A. J.; Tan, C.; Ma, Y.

  Abstract

  Recent advances in spatially resolved transcriptomics (SRT) enabled measurement of sets of pathway genes activity within tissues. However, existing gene set activity scoring methods overlook spatial dependencies among tissue locations, restricting their ability to capture region-specific pathway activities associated with disease pathology or cellular communication. Moreover, these methods lack significance-level inference for activity scores, provide limited interpretability of gene-level contribution to a pathway, and scale poorly to advanced large-size SRT datasets. To address these limitations, we present GESSO (Gene sEt activity Score analysis with Spatial lOcation), a spatially informed gene set scoring method adaptable to diverse SRT platforms. GESSO models gene set activity levels through a graph-regularized matrix decomposition algorithm, jointly inferring spatially coherent gene set activity scores (GASs) and interpretable metagene weights that capture gene-level contributions. It further implements a permutation-based local significance test and a stratified low-resolution approximation that scales to high-resolution SRT datasets such as Visium HD, Stereo-seq, and Xenium Prime. Across 13 datasets from five SRT platforms, GESSO outperformed all existing methods in accuracy, calibration, interpretability, and scalability. Applications revealed novel biological programs, including spatially confined EMT activation within tumor-stroma interfaces, developmental signaling gradients across embryonic tissues, and coordinated B-cell, T-cell, and signaling pathways within germinal centers of human lymph node tissue, revealing the spatial organization of immune function at subregional resolution.

  bioinformatics2026-07-08v1

• FEATMAP: Targeted Correction of Acquisition Signatures Harmonizes Medical Foundation Model Embeddings and Enables Robust Task Generalization

  Donle, L.; Phillips, M.; Gaber, F.; Ramesh, S.; Sacco, M.; Hautaniemi, S.; Virtanen, A.; Bressem, K.; Adams, L.; Goon, K.; Nevins, E.; Robinett, R. A.; Kochanny, S.; Hassan, S.; Dolezal, J.; Pearson, A. T.; Lengyel, E.

  Abstract

  Medical foundation models compress biomedical data into embeddings that support diverse downstream clinical tasks. However, successful model deployment is hampered by performance degradation on external data. It is recognized that embeddings capture acquisition signatures, such as hardware and technical differences, in addition to biology. Effective harmonization must remove the acquisition signature while preserving biological signals, a trade-off that current methods fail to balance adequately. Input-level normalization fails to eliminate acquisition signatures from embeddings, whereas embedding-level methods adjust features in an untargeted manner. We present FEATMAP, a harmonization approach that models acquisition signatures as geometric distortions between manifolds of similarly arranged embeddings. Using paired data that isolate the effect of acquisition signatures, FEATMAP fits a single global affine transformation per foundation model to correct acquisition signatures directly in the embedding space. This targeted, reusable correction aims to preserve biological and demographic variation while harmonizing across acquisition signatures. Across scanner and foundation-model harmonization in digital pathology and field-strength harmonization in brain MRI, FEATMAP improves cross-condition embedding similarity, reduces performance gaps without retraining, and suggests potential for the alignment of disparate embedding spaces.

  bioinformatics2026-07-08v1

• PredHLM: quantitative and interpretable prediction of metabolic half-life in human liver microsomes

  Jang, J.; Cho, N.-C.; Oh, K.-S.

  Abstract

  Motivation: Human liver microsome (HLM)-based metabolic stability assays are fundamental in early drug discovery, shaping pharmacokinetic profiles and oral bioavailability. However, these experimental assays are labor-intensive and time-consuming, limiting their application in large-scale virtual screening. Computational models can prioritize compounds at scale, yet most are classification-based, leaving quantitative and interpretable prediction of HLM half-life limited. Results: In this study, we developed a quantitative machine learning model for the direct prediction of HLM half-life (T1/2) by integrating 11,790 compounds combining in-house and curated public data. Among various combinations of molecular features and learning algorithms, the XGBoost model with RDKit 2D descriptors achieved the best predictive performance, with an RMSE of 0.507 and an R2 of 0.431 on an independent test set. Shapley Additive Explanations (SHAP) analysis identified lipophilicity and known metabolic soft-spot features as the primary contributors to the predictions. These results suggest that this quantitative approach provides a practical framework for defining metabolic stability margins, thereby supporting rapid Go/No-go decisions in preclinical drug discovery. Availability: The source code, data, and trained model are available at https://github.com/joshua-416/PredHLM.

  bioinformatics2026-07-08v1

• Immunoinformatics-Guided Design and In Silico Evaluation of a Multi-Epitope Vaccine Against Influenza A H10N5 and H3N2 Strains Based on Hemagglutinin and Neuraminidase Proteins

  Shabbir, M. Z.; Kumar, P.; Rehman, M. A. U.; Kumar, J.; Urooj, U.; Batool, S. I.; Sourav, C.; Ghazanfar, R.; Nagari, Z.; Hameed, D.; Wahid, A.; Atique, A.; Siddique, M. D.

  Abstract

  Influenza A viruses H3N2 and H10N5 represent, respectively, a persistently dominant seasonal pathogen and a newly documented zoonotic threat with the latter strain variants responsible for the first confirmed human fatality in January 2024, yet no vaccine platform currently addresses co-protection against both subtypes within a unified immunogen. We report here the immunoinformatics based vaccine design and multi-layered computational validation of a 419-amino-acid multi-epitope subunit vaccine construct targeting conserved hemagglutinin (HA) and neuraminidase (NA) antigens identified through multiple sequence alignment of the avian H10N5 (A/swine/Hubei/10/2008) and H3N2 human reference strain sequences to identify viral agents undergoing mammalian adaptations. Linear B-cell, cytotoxic T lymphocyte (CTL), and helper T lymphocyte (HTL) epitopes were predicted using ABCpred, BCEpred, BepiPred 2.0, NetMHCpan 2.1, and NetMHCpan 4.0, then filtered through VaxiJen 3.0, AllerTOP v2.1, and ToxinPred to retain only antigenic, non-allergenic, non-toxic candidates. The final construct, incorporating an avian {beta}-defensin N-terminal adjuvant with GPGPG, AAY, and EAAAK linkers, exhibited a molecular weight of 43.9 kDa, instability index of 31.15, and SOLPro solubility probability of 0.763. Tertiary structure modeling via I-TASSER and GalaxyRefine achieved 84.4% Ramachandran-favored residues. Molecular docking against TLR3 and TLR7 yielded binding free energies of -16.1 and -16.8 kcal/mol with picomolar dissociation constants. Molecular dynamics simulations confirmed complex stability over extended trajectories. Furthermore, codon optimization produced a Codon Adaptation Index of 1.0 for E. coli K12 expression. In silico immune simulation demonstrated robust activation of humoral and cellular immunity including elevated IgG1, IgM, IFN-{gamma}, IL-2, rapid NK cell expansion, and broad B-cell clonal diversity. These findings establish a computationally validated candidate capable of providing protection against influenza in multiple host organisms, warranting experimental advancement.

  bioinformatics2026-07-08v1

• Navigating the pangenome coordinate system with Shredtools

  Shivakumar, V. S.; Langmead, B.

  Abstract

  Existing notions of pangenome coordinates rely on hard-to-compute multiple sequence alignments. On the other hand, pangenome-wide exact unique matches (multi-MUMs) can be computed efficiently, and represent conserved stretches of columns in the underlying MSA. We introduce Shredtools, which uses multi-MUMs as pangenome waypoints and allows for sophisticated queries in pangenome coordinates. Its primary query is extract, which takes an interval of one sequence and extracts the smallest window containing it that is syntenic pangenome-wide. Shredtools' extract query can extract a gene region from 476 human genomes in half a second. Other queries help to refine these results, by finding local exact matches to improve the density of multi-MUM coverage ("enhance") and by selectively discarding sequences to improve the precision of the syntenic region ("zoom"). The Shredtools web interface (available at https://vikshiv.github.io/shredtools) allows for client-side handling of extract queries with index queries handled via simple and fast HTTP Range requests, simplifying usage and enabling pangenome-scale discoveries.

  bioinformatics2026-07-08v1

• Residual Multi-Modal Learning for Pan-Breast-Cancer Drug Response Prediction

  Huang, B.; Tasaka, L.; Li, J.; Islam, T.; Zhang, S.

  Abstract

  Predicting drug sensitivity across diverse cancer cell lines remains a fundamental challenge in precision oncology, particularly for data-scarce cell lines where per-cell-line models overfit and lookup-table approaches cannot generalise to unseen biological contexts. We present DL4DR, a Two Tower Residual Late Fusion deep learning model that addresses this challenge through content-based, identity-free genomic conditioning. The Cell Line Tower encodes each cell line as a 3 x 139 x 139 genomic image - encoding gene expression, mutation severity, and copy-number variation as RGB channels - using a convolutional encoder that maps directly from biological content, never from a cell line ID. The Compound Tower combines three complementary molecular representations: D-MPNN graph message passing, ORNN octave convolutional image features, and an ECFP hard-memorization head that preserves activity-cliff resolution. Predictions are composed as a residual sum: f = fhard + {lambda}(zc). fresidual, where the learned gate $\lambda$ modulates how much interaction signal supplements the memorization baseline. Evaluated across 51 breast cancer cell lines(136,342 records), Residual Fusion outperforms the ECFP-Only baseline in 48/51 cell lines (94.1%), with {Delta}R2 > 0.02 in 26/51 (51.0%). On the leave-cell-line-out split - the decisive test of genomic generalisation - the mean {Delta} R2 = 0.016 across all 51 lines demonstrates that the genomic encoder learns transferable biological signal beyond cell line identity. External validation on 601 cell lines across 27 cancer tissue types (CellTiter-Glo dataset; 0 cell line overlap with training) achieves median R2 = 0.627, within the range of the internal random-split performance (R2 = 0.61--0.69), confirming pan-cancer generalisation. GradCAM interpretability on the Cell Line Tower recovers TP53 among the top-five cross-cell-line genomic activators (5/51 cell lines) alongside several uncharacterised candidate genes (e.g.FSIP2, 6/51) - without any prior pathway annotation - providing partial biological validation of the learned representation, while also indicating that a substantial share of the encoder's top-ranked signal corresponds to genes with no current annotation as breast cancer drivers. Code and data are available at https://github.com/bayjuan5/DL4DR.

  bioinformatics2026-07-08v1

• Gene regulatory co-expression networks decipher potential lncRNA-miRNA-mRNA interactions modulating transcription regulation in neurodegeneration

  Venkatesan, A.; Sinha, P.; Basak, J.; Bahadur, R.

  Abstract

  Neurodegenerative diseases are complex disorders characterised by progressive neuronal loss and widespread transcriptomic dysregulation; however, the coordinated interactions among coding and non-coding RNAs that contribute to disease progression remain incompletely understood. In this study, RNA-seq datasets from disease-relevant neuronal populations and brain regions representing Alzheimer's disease (AD), Parkinson's disease (PD) and amyotrophic lateral sclerosis (ALS) were analysed using an integrative network-based framework. Differential expression analysis coupled with weighted gene co-expression network analysis identified modules significantly correlated with disease and prioritised highly connected hub genes. Integration of these hub genes with curated RNA interaction database enabled the construction of candidate lncRNA-miRNA-mRNA regulatory networks. Functional enrichment analysis revealed Gene Ontology biological processes associated with synaptic signalling, mitochondrial function, RNA metabolism and neuroinflammatory responses across neurodegenerative conditions. The inferred regulatory networks suggested both disease-specific and shared post-transcriptional regulatory modules involving key hub genes and non-coding RNAs. Additionally, putative sequence variants were identified within untranslated regions of selected hub genes, suggesting potential alterations in miRNA-mediated regulations. Therefore, this study provides a systems-level view of transcriptomic dysregulation across major neurodegenerative diseases and identifies candidate regulatory interactions and molecular targets for future functional investigation

  bioinformatics2026-07-08v1

• Beyond infinite sites: Generalized ABBA-BABA statistic for deeper phylogenies

  Zhang, C.; Nielsen, R.

  Abstract

  The Patterson's D statistic detects gene flow from ABBA-BABA site patterns, but its biallelic site patterns fail under deeper divergences where multiple hits cause false positives. We propose two extensions, D+ and D*. Both incorporate multiallelic site patterns to reduce saturation bias under JC and F84 model. Simulations show that D+ and D* both remain correctly null under all conditions and detect gene flow effectively, with distinct advantages: D+ guarantees non-negativity of the denominator, while D* provides greater robustness when mutation rates vary across genomic regions. The source code and binary files are publicly available at https://github.com/chaoszhang/ASTER.

  bioinformatics2026-07-08v1

• An Integrated Knowledge Graph and Network Medicine Pipeline for Drug Repurposing: Benchmarking Across Human Diseases and Application to Amyotrophic Lateral Sclerosis

  Jiang, A.; Hu, J.; Abdulle, Y.; Pain, O.; Iacoangeli, A.

  Abstract

  Drug repurposing offers a practical strategy to identify new therapeutic uses for approved drugs, potentially reducing the time and cost associated with conventional drug development. We present a novel three-stage drug repurposing pipeline that integrates knowledge graph-based gene prediction, network-based drug-disease association analysis, and systematic classification of candidate drugs by therapeutic class. The pipeline integrates DGLinker to predict novel disease-associated genes, SAveRUNNER to identify drug repurposing candidates, and ATC Category Enrichment Analysis (ATCEA) to prioritise candidates by pharmacological class. We benchmarked the pipeline across twelve diseases using DrugBank and MEDI2-HPS as validation resources. Utilising DGLinker-expanded disease-gene sets as input increased the number of predicted repurposed drugs, while overall discriminative performance remained stable across diseases (AUROC 0.71-0.77). Application of ATCEA consistently improved precision, F1-score, and specificity, while reducing recall, reflecting a conservative prioritisation strategy that contracts the candidate space while retaining pharmacologically coherent drug-disease candidates. We further applied the pipeline to amyotrophic lateral sclerosis (ALS), a neurodegenerative disease with limited therapeutic options, and performed a deeper literature-based validation of the results. Incorporation of DGLinker-predicted genes substantially increased the number of significant candidate drugs and uncovered enriched ATC categories not identified using known ALS genes alone, including antidepressants and antipsychotics. Moreover, several drugs with supporting evidence available in the literature were identified only when DGLinker-predicted genes were used. Overall, 77 candidate drugs were prioritised within significantly enriched ATC categories, several of which are supported by previously published studies. To provide exploratory real-world support for these findings, we further evaluated candidate drugs in a longitudinal electronic health record (EHR) dataset of 2361 patients with ALS from King's College Hospital. Although the number of evaluable drugs was limited due to sample size, the EHR analysis provided additional clinically relevant context for selected prioritised drugs and pharmacological classes. Our pipeline demonstrates potential to accelerate drug repurposing by integrating complementary computational approaches to each step of the process, providing an end-to-end framework that showed robust performance across benchmarking experiments and use cases.

  bioinformatics2026-07-08v1

• AllTheBacteria: a community resource empowers biology and discovers novel peptide antibiotics

  Hunt, M.; Torres, M. D. T.; Alikhan, N.-F.; Anderson, D.; Andreani, M. L.; Blom, J.; Bouras, G.; Brinkman, F.; Carroll, L. M.; Croxen, M. A.; Floto, A.; Hall, M. B.; Hawkey, J.; Horsfield, S. T.; Jia, B.; Lacey, J. A.; Lee, H.-S.; Lima, L.; MacAlasdair, N.; Mallawaarachchi, S.; Matlock, W.; Moustafa, A. M.; Petit, R.; Raghuram, V.; Ramnath, V.; Russell, M. J.; Sanderson, T.; Saratto, T.; Schwengers, O.; Seemann, T.; Shaw, L. P.; Shen, W.; Thomson, N.; Tonkin-Hill, G.; Toussaint, J.; Viet, T. L.; Wachsmann, J. v.; Wan, F.; Weimann, A.; Wheatley, R. M.; Wiatrak, M.; Xie, O.; Fuente-Nunez, C. d.

  Abstract

  Public microbial genomes encode an immense record of biological diversity, evolution and molecular function, but much of this information remains difficult to reuse because raw sequencing data are not uniformly assembled, quality controlled, annotated or searchable at scale. Here we present AllTheBacteria, an open, community-built resource that transforms public bacterial short-read whole-genome sequencing reads into a uniformly processed discovery platform. The current analysed release contains 2,440,377 high-quality bacterial and archaeal genomes from 11,273 species, together with standardized taxonomic assignments, genome annotations, antimicrobial resistance calls, antiphage-defence annotations, protein structure predictions and AI-ready sequence tables. We show that this infrastructure enables applications that would otherwise be impractical, from global sequence search and outbreak contextualization to pangenome method development, antimicrobial resistance reservoir mapping and antiphage-defence ecology. As a stringent experimental demonstration, we mined 3,919,096 encrypted peptide fragments from AllTheBacteria proteomes using our deep learning model APEX 1.1, identifying 1,867 candidates with predicted antimicrobial activity. We synthesized 24 representative peptides and tested them against 20 clinically relevant bacterial strains, including antibiotic-resistant pathogens. Multiple peptides showed low-micromolar activity, membrane-responsive conformational transitions and selective envelope perturbation. A lead molecule, ATB20, reduced Acinetobacter baumannii burden in a murine skin abscess model with efficacy comparable to polymyxin B and no overt toxicity. Together, these results establish AllTheBacteria as both a foundational community resource for microbiology and a renewable engine for AI-guided antimicrobial discovery.

  bioinformatics2026-07-07v8

• The RdRp Thumb-1 Pocket is a Conserved Target for Broad-Spectrum Antiviral Development

  Woods, V.; Umansky, T.; Russell, S. M.; Gallay, P.; Smith, D.; Haders, D.

  Abstract

  Single-stranded RNA (ssRNA) viruses cause human diseases ranging from mild colds to deadly pandemics. Broad-spectrum non-nucleoside antivirals have been characterized as impossible to develop because allosteric binding sites are poorly conserved. The Thumb-1 allosteric site identified in HCV's RNA-dependent RNA polymerase (RdRp) governs an essential conformational change in the {Lambda}1-loop required for polymerase initiation. The only approved Thumb-1 inhibitor, beclabuvir, has been shown to be inactive against a broad panel of non-HCV viruses, including poliovirus, rhinovirus, coronavirus, coxsackievirus, influenzavirus, and HIV. It subsequently failed to inhibit SARS-CoV-2 despite favorable docking predictions. A conserved, homologous allosteric site on RdRp that spans multiple viral families has not been reported. Here, we demonstrate that the Thumb-1 pocket and its associated {Lambda}1-loop are conserved across ssRNA viral families through comparative structural analysis and multiple sequence alignments. We demonstrate that beclabuvir's dependence on its indole C6 carbonyl to interact with the HCV-specific residue R503 and its C3 cyclohexyl chemistry restricts its activity to HCV. We validate the target discovery with MDL-001, which does not contain a C6 carbonyl or a C3 cycloalkyl substituent. MDL-001 directly blocks viral RNA synthesis in isolated replication complexes and selects for the canonical Thumb-1 resistance mutation P495S in HCV. MDL-001 demonstrates broad-spectrum in vitro inhibition of both HCV and SARS-CoV-2. Preclinical proof of concept and development of MDL-001 across HCV, HBV, HDV, influenza, SARS-CoV-2, and RSV are reported in a companion manuscript. These findings establish RdRp Thumb-1 as a conserved allosteric pocket and a druggable target for broad-spectrum antiviral development.

  bioinformatics2026-07-07v6

• PLANET-MD: Ultra-fast Proteome-scale Prediction of Allosteric Networks in Proteins

  Sledzieski, S.; Hanson, S. M.

  Abstract

  Proteins are dynamic molecules that depend on conformational flexibility to carry out functions in the cell, yet despite significant advances in the modeling of static protein structure, prediction of these dynamics remains challenging. We introduce PLANET-MD, a machine learning model that predicts dynamic protein properties from sequence or static structure with unprecedented speed and accuracy. Trained on thousands of molecular dynamics trajectories spanning diverse protein families, PLANET-MD simultaneously models multiple dynamics features: root-mean-square fluctuations (RMSF), generalized correlation coefficients (GCC-LMI), and a novel structural heterogeneity profile (SHP) based on recent structure quantization methods. PLANET-MD significantly outperforms existing methods in predicting simulation-derived dynamics. We reduce RMSF prediction error by 57% compared to BioEmu and calibrated Dyna-1 predictions, including an up to 73% error reduction for long proteins. We validate these predictions with experimental hetNOE data, and we demonstrate the ability to adapt predictions to different physical temperatures. We highlight PLANET-MD's utility in constructing allosteric networks in the oncogene KRAS and identify structural sub-modules with correlated motions, and we validate PLANET-MD by showing that changes in node centrality within predicted KRAS allosteric networks correlate with changes of folding free energy in experimental DMS data. Our approach makes predictions in seconds rather than hours or days, enabling us to perform the first comprehensive dynamics analysis of the entire human proteome. PLANET-MD bridges the gap between static structural biology and dynamic functional understanding, enabling dynamics-aware structural analysis and variant effect prediction at scales previously unavailable. PLANET-MD is available as free and open-source software at https://github.com/flatironinstitute/PLANET-MD.

  bioinformatics2026-07-07v2

• Identifying intervention strategies from machine learning models with COALA: a counterfactual optimization framework

  Han, B.; Duan, Q.; Hu, T.

  Abstract

  Motivation: Machine learning models in biomedicine have become increasingly complex, often functioning as black boxes. However, understanding contributors to disease and making actionable health interventions requires interpretable models. Common explainable AI methods like SHAP focus on feature importance but fall short in explaining why features contribute in certain patterns or what interventions to take. Counterfactual explanations address this by proposing "what if" scenarios but current tools focus on individual predictions and fail to generalize complex trends. Results: We introduce the framework Counterfactual Optimization for Actionable interpretabiLity in AI (COALA). COALA interprets models by identifying optimal counterfactuals across user-defined mutable feature subsets and constraining remaining features to reveal how constraint features determine what interventions are optimal. By analyzing counterfactual profiles of features rather than individual features, COALA reveals holistic patterns. Using synthetic and real datasets, COALA reveals simple and complex model trends and provides more intuitive, multi-feature interventions than SHAP. Availability and Implementation: Code for COALA implementation, synthetic data, models trained on synthetic data, and code to replicate results and figures are available at https://github.com/brt-solo/COALA.

  bioinformatics2026-07-07v2

• PHI: A Galaxy-based workflow for reproducible prophage-host interaction analysis and standardized viral-genomics reporting

  Saraiva, J. P.; Borim Correa, F.; Bernt, M.; Ghanem, N.; Nieto, E.; Brizola Toscan, R.; Y. Wick, L.; Chatzinotas, A.

  Abstract

  Background: Viruses that infect bacteria, known as bacteriophages or phages, are widespread in nature and play important roles in shaping microbial communities and ecosystem functions. Some phages can integrate into bacterial genomes as "prophages", where they may influence the biology of their host by carrying genes that affect metabolism, virulence, or environmental adaptation. Despite their importance, studying prophages and their interactions with bacterial hosts remains challenging because it typically requires combining many complex computational tools and can be resource-intensive. Results: In this study, we introduce the Prophage-Host Interaction Toolkit (PHI), a user-friendly and automated workflow available through the Galaxy platform. PHI brings together multiple established tools into a single, reproducible pipeline that identifies candidate prophages, evaluates their quality, predicts host relationships, and characterizes key functional genes. Importantly, all results are summarized in an interactive report that simplifies interpretation. When applied to a mock community composed of 22 bacteria as a workflow demonstration, PHI detected 41 prophages across 14 hosts, classifying them into high- and medium-quality phage genomes. Host assemblies exhibited > 99 % completeness and < 1 % contamination for most genomes, while DefenseFinder revealed between 3 and 24 antiviral systems per genome. Conclusions: By removing installation barriers and consolidating the outputs of multiple established tools, PHI lowers the barrier to advanced phage analysis, enabling both specialists and non-experts to explore phage-host interactions and their implications in areas such as microbiome research, biotechnology, and environmental science.

  bioinformatics2026-07-07v2

• Artificial intelligence aided design of peptides with custom secondary structure motifs and reduced amino acid alphabets

  Brown, S. M.; Cohen, A. B.; Dean, S. N.

  Abstract

  Proteins are highly diverse functional polymers where the specific sequence of amino acids, selected from a standard genetically-encoded alphabet of twenty (C20), determines the structure and ultimately the function of the resulting folded protein. This standard alphabet has been identified to be non-randomly distributed in physicochemical properties crucial to both structure-formation and function, often referred to as coverage theory. While machine learning models have drastically improved protein structure prediction, success of protein design models lags structure prediction, particularly for custom secondary structure motifs and amino acid alphabets. Here we therefore bridge contemporary biological theory with recent advancements in artificial intelligence (AI) to develop and evaluate a generative AI protein design model, trained on hundreds of thousands of proteins within the RSCB PDB, for custom secondary structure motifs using reduced amino acid alphabets (RAAs). Results indicate an overall success in designing novel proteins with desired secondary structure motifs for a broad range of amino acid alphabets and complexity of designs. Interestingly, this tool often captures the full three-dimensional tertiary structure of a target protein despite training only on physicochemical sequence space and secondary structure information. The development of this model advances research across multiple disciplines, from general scientific AI architecture development to protein design for biotechnology, astrobiology, and early-Earth evolutionary biology.

  bioinformatics2026-07-07v2

• Real-time mass defect-driven prediction of glycopeptide precursors enables enrichment-free serum glycoproteomics

  Zhang, B.; Chau, T. H.; Bienes, K. M.; Arakawa, H.; Kaji, H.; Kawahara, R.; Ashwood, C.; Matsui, Y.; Thaysen-Andersen, M.

  Abstract

  Glycopeptide enrichment remains a cornerstone in glycoproteomics, but bias and reproducibility issues continue to hinder biological insight and clinical translation. Using curated glycoproteomics datasets and machine learning, we trained a glycopeptide classifier to promptly recognize N-glycopeptide precursor ions in peptide mixtures through mass defect signatures. Integration of the classifier into a data-dependent acquisition framework facilitated efficient and unbiased real-time prediction of N-glycopeptides directly from serum opening avenues for enrichment-free glycoproteomics.

  bioinformatics2026-07-07v2

• PEPstrMOD2: Next-generation tertiary structure prediction of chemically modified and non-natural peptides

  Jain, S.; Mehta, N. K.; Raina, S.; Kumar, P.; Varun, V.; Raghava, G. P. S.

  Abstract

  While most existing methods are limited to predicting the tertiary structures of proteins containing only canonical residues, the PEPstrMOD server (developed in 2015) pioneered structure prediction for chemically modified and non-natural peptides. Despite its widespread use, the original framework was restricted to peptides of 7 to 25 residues and relied on older backbone-prediction algorithms. To address these limitations, we present PEPstrMOD2, which introduces three major advancements over its predecessor. First, it replaces the original in-house coordinate generation with state-of-the-art deep learning (DL) algorithms, leveraging AlphaFold2 and ESMFold for highly accurate initial structure prediction. Secondly, it greatly expands the accessible chemical space through incorporation of new, AMBER force-field compatible library of 257 post-translational modifications (PTMs), 428 non-canonical amino acids (NCAAs), and 243 terminal modifications. Lastly, through the application of native scalability of AlphaFold2 (AF2) and ESMFold (EF), PEPstrMOD2 eliminates the original restrictions of the length, enabling the structural modeling of longer, complex therapeutic peptides and small proteins. We evaluated the performance of PEPstrMOD2 against state-of-the-art methods across three distinct peptide datasets. For the AfCyc dataset consisting of 80 cyclic peptides, PEPstrMOD2 obtained a competitive average atom-level Root Mean Square Deviation (RMSD) of 2.05 angstroms, compared to 1.13 angstroms by AlphaFold3 (AF3) and 1.82 angstroms by AfCycDesign. Remarkably, for the modified peptide ModPep433 dataset, PEPstrMOD2 outperformed AF3, achieving the lower average RMSD score of 4.49 angstroms against 4.67 angstroms of AF3. Furthermore, in the case of the ModPep16 benchmark, PEPstrMOD2 achieved 2.50 angstroms average RMSD value, which is two times more accurate than that of the original PEPstrMOD (5.84 angstroms). In summary, PEPstrMOD2 provides a powerful, high-throughput, and highly accurate platform to facilitate peptide-based drug development and structural biology research. While the original PEPstrMOD was restricted to a web server interface, PEPstrMOD2 is available as both an intuitive webserver and a standalone command-line tool via GitHub, featuring Docker support for easy deployment and reproducible, large-scale modeling pipelines (https://webs.iiitd.edu.in/raghava/pepstrmod/).

  bioinformatics2026-07-07v2

• LINKER-Pred: A Deep Learning Method and Web-Server for the Prediction of Disordered Flexible Linkers in Proteins

  Meng, D.; Garcia Alvarez, H. M.; Glavina, J.; Leonetti, C. O.; Pollastri, G.; Chemes, L. B.

  Abstract

  Disordered Flexible Linkers (DFLs) are unstructured regions that play critical roles in inter-domain communication and multivalent protein interactions. Despite their biological significance, the accurate identification of DFLs remains challenging due to limited experimental annotations and sparsity of dedicated prediction tools. Here we introduce LINKER-Pred, a publicly available web server featuring two convolutional neural network-based predictors trained on a novel large-scale dataset of linkers connecting folded domains (DLD dataset) and DisProt linkers. LINKER-Pred2 combines ProtTrans and MSA-Transformer embeddings within an ensemble CNN framework, achieving state-of-the-art performance on CAID2 and CAID3 benchmarks. LINKER-Pred-Lite excludes MSA-based features, improving speed while maintaining competitive predictive accuracy. LINKER-Pred predictors offer robust residue-level DFL predictions directly from sequence, providing a scalable solution for DFL annotation across proteomes. The LINKER-Pred web server and associated resources are freely available at https://linkerpred.chemeslab.org/, offering the research community an accessible tool for studying protein disorder and modularity.

  bioinformatics2026-07-07v2

• FAMUS: A Few-Shot Learning Framework for Large-Scale Protein Annotation

  Shur, G.; Burstein, D.

  Abstract

  Predicting gene function is a pivotal and challenging step in genomic and metagenomic data analysis. Current automatic annotation tools typically rely on the single most similar sequence from the query database and struggle to robustly set hit thresholds for annotation. The sparsity of proteins per annotation makes it challenging to confidently assign gene function for underrepresented families. Here, we present a contrastive learning framework for functional annotation. FAMUS (Functional Annotation Method Using Supervised contrastive learning) compares query sequences to a full array of profile Hidden Markov Models and transforms the similarity scores into a condensed vector space that minimizes the distance of proteins from the same family. The similarity scores of a query to all profiles are used for its representation instead of considering only the top-ranking hit. Unannotated sequences are incorporated as negative examples during training, enabling robust detection of proteins that fall outside the scope of the reference database without requiring a user-defined threshold. Using this approach, FAMUS outperformed KEGGs native KofamScan for KEGG Orthology annotation and InterPros InterProScan for PANTHER family annotation. We thus created four protein annotation models using protein families from the KEGG Orthology, InterPro family, OrthoDB, and EggNOG databases. All four models are available as a conda package and via our user-friendly web server, allowing users to annotate large-scale datasets. FAMUS is the first comprehensive and modular annotation framework based on contrastive learning. It supports both pre-defined and user-specific databases for tailored annotation, and can be easily integrated into any genomic and metagenomic analysis pipeline to facilitate accurate, large-scale functional annotation.

  bioinformatics2026-07-07v2

• BRAID: RT-PCR-calibrated conformal intervals for splicing ΔPSI

  Park, J.; Kang, K.

  Abstract

  Differential splicing workflows usually report a {Delta}PSI point estimate and a statistical score, but these outputs do not directly state whether the RNA-seq estimate is close enough to an orthogonal validation measurement. We developed BRAID as a post-processing calibration step for splicing analyses. BRAID estimates RNA-seq {Delta}PSI from rMATS inclusion and skipping counts, retains the upstream caller evidence, and adds a 95% interval whose width is calibrated from empirical RNA-seq-to-RT-PCR residuals using split conformal prediction. The packaged differential-splicing calibrator uses a residual half-width of q = 0.341, estimated from 162 RT-PCR-validated skipped-exon events. We evaluated BRAID on three RT-PCR validation datasets covering TRA2 knockdown, mouse cerebellum versus liver, and a prostate epithelial-to-mesenchymal comparison. On the pooled common set of 139 cassette-exon events, BRAID reached 0.971 RT-PCR coverage, whereas MAJIQ, betAS, and rMATS-derived intervals reached 0.518, 0.734, and 0.633, respectively. BRAID also had the lowest pooled interval score, 0.720, compared with 2.040 for MAJIQ, 1.414 for betAS, and 1.625 for rMATS. Applying the same residual calibration to other caller outputs brought MAJIQ, betAS, rMATS, and SUPPA2 {Delta}PSI estimates close to nominal RT-PCR coverage, indicating that the gain came from interval calibration rather than from a caller-specific point estimate. In a TRA2 positive-negative validation panel, using q as a hard rMATS effect-size cutoff reduced recall, whereas using q as an interval half-width improved RT-PCR coverage. Applied to a public DM1 skeletal-muscle rMATS table, BRAID reduced 967 large-effect significant events to 68 high-confidence interval-supported events and retained known DM1 and muscle-splicing signals. BRAID provides a practical calibrated reliability layer for RNA-seq splicing studies where downstream follow-up depends on the precision of reported {Delta}PSI estimates.

  bioinformatics2026-07-07v1

• Molecular Clock Dating of Ancient Environmental DNA Reveals Damage Beyond Deamination

  Lemmon-Kishi, M.; Pipes, L.; De Sanctis, B.; Nielsen, R.

  Abstract

  Ancient environmental DNA (aeDNA) from permafrost, lake, cave, and marine sediments provides a rich source of genetic data that captures broad perspectives of past biodiversity. Accurate dating is crucial for discovering ecologically relevant patterns from aeDNA, and molecular clock dating would allow for sample ages to be estimated from the recovered genetic material itself instead of the geological components. However, the fragmented and damaged nature of short-read ancient DNA (aDNA) from multiple taxonomic sources poses significant challenges and has limited this dating approach for aeDNA. Here we developed ratePlacer, a phylogeny-based method for analyzing aeDNA that can combine information from many short reads in a sample while accounting for DNA damage to provide maximum likelihood estimates of sample ages. Simulations demonstrate that ratePlacer accurately dates samples even under the fragmented, damaged conditions characteristic of aeDNA and outperforms Bayesian tip-dating approaches for taxonomically mixed samples commonly found in aeDNA. Yet age estimates from re-dating Kap Kobenhavn varied across taxa, highlighting the difficulty of molecular clock dating in aeDNA. This dating also revealed elevated G[-&gt;]T and C[-&gt;]A mismatches consistent with oxidative damage. These patterns reveal aDNA damage beyond deamination and that remains understudied, suggesting that aeDNA should be carefully evaluated in genomic and evolutionary analyses. The new dating method, ratePlacer, extends molecular clock dating of aDNA from single-specimen to pooled environmental DNA data, where traditional methods struggle.

  bioinformatics2026-07-07v1

• ThermoFusion: A Multimodal Deep Learning Framework for Generalizable Prediction of Enzyme Thermostability

  Wei, Y.; Eberini, I.; Meyer, F.

  Abstract

  Protein thermostability is a critical property for both industrial and biomedical enzyme applications, yet experimental evaluation of mutation-induced stability changes remains laborious and costly. Here, we present ThermoFusion, a hybrid deep learning framework that integrates 3D protein structure embeddings from ThermoMPNN with sequence-based embeddings from the pretrained protein language model ESM2 to predict the effects of single-point mutations on protein stability ({Delta}{Delta}G). ThermoFusion exhibits robust generalization, maintaining high predictive accuracy across out of distribution sequences with low identity to the training set -- a scenario where many other machine learning models, including ThermoMPNN and state-of-the-art tools, perform poorly due to reliance on memorization. Benchmarking on a curated enzyme dataset comprising of 105 enzymes and 3144 mutations shows that ThermoFusion reliably identifies stabilizing mutations while accurately predicting stability for enzymes beyond its training set. These results establish ThermoFusion as a powerful tool for rational enzyme design beyond its training set.

  bioinformatics2026-07-07v1

• PACMOS: an R package for Projection And Classification of Multi-Omic Samples

  Kalson, L.; Sexton-Oates, A.; Drevet, G.; Fernandez-Cuesta, L.; Foll, M.; Alcala, N.

  Abstract

  Motivation: Integrated multi-omic analyses have transformed our understanding of cancer biology, giving rise to data-driven molecular classifications that capture disease heterogeneity beyond conventional histopathology. Among these approaches, multi-omic factor analysis (MOFA), a multimodal extension of principal component analysis, has been widely used to identify sources of molecular variation across omic layers and classify samples into molecular groups. However, classifying query samples according to an existing MOFA-based classification remains challenging, as there is no validated computational method for projecting samples into pretrained MOFA latent factor spaces. Results: We present PACMOS, an R package that provides a generalizable approach to project query samples into pretrained MOFA latent factor spaces. We validate PACMOS using two cancer datasets with published MOFA-based classifications - lung neuroendocrine neoplasms and pleural mesothelioma - showing that PACMOS preserves the existing MOFA latent factor space while allowing to classify query samples. Availability and implementation: PACMOS is an open-source R package available on the IARC bioinformatics GitHub organization (submitted to Bioconductor) at https://github.com/IARCbioinfo/PACMOS and DOI in Zenodo: https://doi.org/10.5281/zenodo.20933824, along with installation instructions and a vignette with an application. Supplementary information: Supplementary data are available in separate files.

  bioinformatics2026-07-07v1