Frontiers in Bioinformatics

Background and objectiveHighrisk nonmuscleinvasive bladder cancer (HRNMIBC) is treated with transurethral resection and intravesical BCG instillations, yet {approx}50% recur and 20% progress to invasive disease. Although molecular subtyping, e.g., BCG-response-subtype (BRS), is associated with progression risk and may aid risk stratification, yet is costly and time-consuming. Intratumoral heterogeneity complicates accurate subtyping. To address these challenges, we developed a deep-learning model that predicts BRS from routine hematoxylin-eosin-stained images. We verified the models area-by-area predictions against tissue-level gene-expression maps. Methods and participantsHematoxylin-eosin-stained images from 231 HR-NMIBC patients with known BRS were used to develop a deep-learning model through cross-validation, then validated in 83 independent samples. The models spatial predictions were assessed using spatial transcriptomics to map gene expression to tissue locations in five HR-NMIBC tumors. Outcome measurements and statistical analysisDiscriminative ability for BRS3 vs. BRS1/2 was measured by AUC. Spatial alignment was assessed by calculating Pearson and Spearman correlation coefficients between model predictions and BRS fractions; significance was assessed through permutation analysis. Key findings and limitationsThe trained algorithm achieved AUC of 0.79 (development) and 0.71 (external) to detect BRS3 vs BRS1/2. Tile-level correlation between model output and molecular labels was significant (Pearson r = 0.33-0.44; p [≤] 0.002). Limitations include retrospective sampling and limited spatial transcriptomic cases. Conclusions and clinical implicationsOur trained algorithm showed potential to stratify HRNMIBC patients by clinically relevant BCGresponse subtypes using routine hematoxylin-eosin-stained images and showed predicted spatial heterogeneity comparable to molecular profiling. Prospective validation is required before any clinical implementation. Patient summaryStandard pathology images contain hidden details related to tumors molecular subtype. We trained an AI model to read these routine images and identify specific bladder cancer subtypes associated with poor response to BCG therapy. This approach may help reveal molecular subtype-associated information from routine pathology images, without additional laboratory procedures.

Predicting protein-protein interactions is a fundamental problem in molecular biology. Experimental approaches for identifying protein-protein interactions are time-consuming and labor-intensive, motivating the development of efficient computational alternatives, including machine learning-based methods. However, conventional machine learning methods often rely on manually engineered features that require substantial domain expertise. In this study, we propose a two-stage framework to address these limitations. In the first stage, a one-dimensional convolutional neural network autoencoder is used to automatically learn latent representations from protein sequences. The quality of these features is evaluated through reconstruction error, reflecting how accurately the model reconstructs the original sequence. In the second stage, these learned features are combined with amino acid frequency-based features to form a hybrid feature set for predicting protein-protein interactions. A systematic comparison is performed between models trained on frequency features alone and those using a hybrid representation. The comparison showed that incorporating one-dimensional convolutional neural network-derived latent features improved the models performance of predicting protein-protein interactions. The dataset was split into training, validation, and test sets. Nested cross-validation was employed, with inner loops for hyperparameter tuning and outer loops for model selection. The random forest classifier achieved the best performance, with a mean receiver operating characteristic-area under curve of 0.91 and a test F1-score of 0.87. These results highlight the effectiveness of integrating deep feature learning with ensemble methods for predicting protein-protein interactions and build upon previous work focused on this fundamental problem. Author SummaryProtein-protein interactions are fundamental in all biological processes. However, predicting these interactions is a key problem in molecular biology. Computational approaches have been tested to address this problem. We applied a mix of machine learning and deep learning to gain insight into the qualities of proteins that engage in interaction. First, we trained a deep learning model, which automatically learned the primary sequence and characters related thereto, reducing bias in the actual prediction process. We combined these features, or latent representations, with amino acid frequency features of protein sequences, and called the two together "hybrid features." Then we performed a systematic comparison of amino acid frequency features-only with hybrid features, among four different machine learning classifiers. Our results suggest that the random forest classifier performed best among all four classifiers at predicting interactions between proteins. We propose that this approach could be used to improve efficiency in testing protein-protein interactions at the bench and may have applications to other biologically relevant molecular interactions.

Objective. Biomedical knowledge graphs (KGs) such as PrimeKG, Hetionet, UMLS, and PharmGKB are increasingly used as the substrate for downstream machine-learning, retrieval-augmented generation, drug-repurposing, and electronic health record (EHR) augmentation pipelines. The dominant assumption in published work is that integrating two or more such KGs is a tractable engineering step solved by identifier (ID) matching. This paper interrogates that assumption empirically. We quantify how much concept overlap survives realistic alignment, and we characterize the new failure modes introduced by the methods that practitioners reach for when ID matching is insufficient. Materials and Methods. We compared four widely used biomedical KGs (PrimeKG, Hetionet v1.0, the full UMLS Metathesaurus, and PharmGKB) across eleven node types using a tiered alignment pipeline: (1) direct ID matching for nodes sharing a primary vocabulary; (2) cross-ontology bridging using standard mappings (e.g., MONDO-DOID, HPO-UMLS, HPO-UMLS-MeSH for side effects, NCBI Gene-HGNC-UMLS, UBERON-FMA/SNOMEDCT_US/NCI/MeSH for anatomy); (3) ClinicalBERT cosine-similarity grouping at threshold >= 0.98 for over-segmented disease nodes, with a deterministic suffix-stripping canonicalizer; (4) exact name matching for ontology-poor types (anatomy, REACTOME pathways); and (5) embedding-based fuzzy matching with UMLS lookup (SapBERT and ClinicalBERT) for free-text microbiome concepts. We applied the pipeline to a 698-concept gut-microbiome benchmark spanning taxa, pathways, and disease labels, validated grouping decisions against the curated SSSOM mappings released by the MONDO project, and audited the ClinicalBERT consolidation against five clinical-genetics case studies drawn from the literature. Results. Per-type pairwise coverage was strikingly asymmetric. Genes/proteins and the three Gene Ontology categories aligned cleanly across PrimeKG and Hetionet (mutual coverage 94-99%), but disease overlap was sparse: only 0.7% of PrimeKG individual disease nodes mapped to Hetionet, rising to 2.0% after MONDO grouping (versus 78.7% and 18.4% from the Hetionet side). PrimeKG-to-UMLS coverage spanned 100% (effect/phenotype via HPO) down to 20.8% (REACTOME pathways), with drugs at 73.7% and anatomy at 58.8%. PrimeKG-to-PharmGKB drug coverage required up to two bridging hops (DrugBank -> UMLS -> RxNorm/ATC/MeSH). Bigger was not uniformly more complete: on a 698-concept microbiome drug benchmark, Hetionet missed 0 concepts while PrimeKG missed 16. ClinicalBERT-based grouping consolidated 22,205 raw MONDO disease nodes into 17,080 groups but introduced three reproducible failure modes documented in case studies: (i) peer over-merging: for example, all 22 osteogenesis imperfecta subtypes collapsed into a single node despite distinct severity classes; (ii) parent-child collapse: e.g. acute myeloid leukemia merged with myeloid leukemia, erasing the acute/chronic distinction that drives clinical management; and (iii) lexical false positives: neurofibromatosis and schwannomatosis grouped together despite cellular-pathology differences. Discussion. Identifier matching alone is a weak baseline for biomedical KG integration. Cross-ontology bridges and embedding-based consolidation expand coverage but do so at the cost of clinically meaningful resolution, and the resulting failures are systematic rather than random. Reporting only aggregate coverage statistics obscures these losses, which propagate silently into downstream tasks. Conclusion. We provide reusable per-type coverage tables, a taxonomy of three integration failure modes, and concrete recommendations for downstream studies that depend on a unified biomedical KG. We argue that future KG integration work should report per-type coverage and per-cluster confidence rather than aggregate match rates.

Protein language models (pLMs) offer great potential for protein sequence analysis, yet the scarcity of labeled data often limits their effectiveness in fine-tuning. Data augmentation is a promising remedy, but systematic evaluation of augmentation strategies for protein sequences remains limited, and the conditions under which augmentation confers downstream benefits are not well understood. In this paper, we systematically investigate pLM-guided substitution-based augmentation across seven protein prediction tasks. We propose ProtAug, a framework that leverages encoder-based (ESM-2) and autoregressive (ProtGPT2) pLMs to generate augmented sequences with user-controlled variation levels. Our investigation focuses on four questions: (Q1) whether pLM-synthesized sequences preserve more original signals than simpler methods, (Q2) to what extent augmentation improves prediction performance, (Q3) how variation levels affect downstream accuracy across tasks and models, and (Q4) whether biological plausibility is a necessary condition for achieving improvement. Our experimental results show that: (1) ProtAug Esm generally preserves motifs and structural similarity better than simple substitution, often comparable to homology retrieval; (2) augmentation yields consistent but task-dependent improvements, with ProtAug Esm achieving the best or second-best performance in 5 out of 7 tasks at 10% variation; (3) low-to-moderate variation levels (2-30%) perform best overall, although high-variation augmentation can benefit certain structure-related tasks; (4) the necessity of biological plausibility is task- and variation-dependent--while semantic preservation correlates with performance at low-to-moderate variation levels, improved generalization at high variation levels suggests that regularization effects, rather than label preservation, can also drive performance gains.

Dimensionality reduction for visualization is a fundamental step in single-cell RNA sequencing (scRNA-seq) analysis due to the extremely high dimensionality of gene expression profiles. However, widely used nonlinear embedding techniques such as UMAP and t-SNE can introduce substantial distortions when projecting data into two-dimensional space, potentially altering global organization, local neighborhoods, and distance relationships in ways that may mislead downstream biological interpretation. In this study, we investigate the applicability of Clustering-Based Manifold Approximation and Projection (CBMAP) for the visualization of scRNA-seq data and systematically examine how clustering strategies influence the quality of the resulting embeddings. CBMAP was integrated with several clustering algorithms commonly used in single-cell analysis, including k-means, Leiden, HDBSCAN, Secuer, HGC, and FlowSOM. The resulting embeddings were evaluated using quantitative metrics that measure global, local, and distance-level structure preservation and were compared with widely used dimensionality reduction methods such as UMAP, t-SNE, and PaCMAP across multiple benchmark datasets. Our results demonstrate that the clustering stage plays a critical role in determining the structural fidelity of CBMAP embeddings. Clustering algorithms specifically designed for single-cell transcriptomic data, particularly Secuer, produced more consistent preservation of global relationships between cell populations. Across multiple datasets, CBMAP more faithfully preserved global structural organization and inter-population distance relationships than the compared methods, although local neighborhood preservation was generally weaker than in techniques optimized for local structure. Importantly, CBMAP embeddings retained biologically meaningful relationships in trajectory benchmark datasets. When combined with RNA velocity analysis, CBMAP successfully preserved cyclic progenitor states and branching differentiation trajectories, demonstrating compatibility with trajectory-aware visualization. These findings indicate that CBMAP provides a structure-faithful visualization framework for scRNA-seq data and that clustering selection plays a central role in determining embedding quality.

Glycans coat the surface of all cells, and every glycan is recognised by specific glycan-binding pro-teins (GBPs). There are no general tools that can accurately estimate the binding strength between glycan and GBP from the amino acid sequence of the GBP and the molecular structure of the glycan, represented as SMILES string. We describe models for predicting such binding strengths developed as a part of a Capstone Course at the University of Alberta. The models are trained on a dataset that combines BindingDB, a published database of small-molecule protein interactions, and data from glycan arrays measured by Consortium of Functional Glycomics (CFG). In this hybrid dataset of protein-ligand interactions the ligands are both glycans from CFG and small molecules from BindingDB; similarly, proteins include GBP and proteins from BindingDB. Three models are presented (i) ProMax which fuses ESM-2, MolFormer, and MolCLR features; (ii) APEX which constrains learning to a predetermined form, a physical model of binding; (iii) UltraMax adds inter-atomic distances for the ligands. To address the dataset's severe long-tail distribution, the models employ tail-aware losses for rare high-binding instances. Trained and evaluated on approximately one million protein--ligand pairs using hold-out splits for unseen molecules, the three models provide a unified framework for quantitative glycan-protein binding prediction. We observed that learning glycan-protein binding is harder than the similar task of learning small-molecule-protein interactions. Simple mirror-inversion tests led us to postulate that insufficient use of chiral features is an important source of difficulty in learning these interactions.

Clear cell renal cell carcinoma (ccRCC) is characterized by transcriptional reprogramming driven by hypoxia signaling, metabolic rewiring, and immune modulation. While gene-level analyses have defined key features of ccRCC biology, they do not capture isoform-level variation arising from alternative splicing. Differential transcript usage (DTU) represents an additional regulatory layer that may influence protein function, pathway activity, and clinical outcomes, yet its role in ccRCC biology and prognosis remains incompletely understood. We assessed differential expression in 127 ccRCC tumors and 33 normal-adjacent tissues from the Dartmouth Cancer Center cohort, with external validation in 94 CPTAC tumors, adjusting for cell-type proportions. DTU was identified using DRIMSeq/stageR, followed by limmavoom modeling with clinical and tumor microenvironment covariates. Transcript-based consensus clustering defined tumor subgroups, and Cox proportional hazards modeling integrated transcript-level features with clinical variables. In tumor versus normal comparisons, 1,170 transcripts exhibited significant differential usage, mapping to canonical ccRCC pathways with distinct patterns across functional and non-functional transcript classes. Consensus clustering based on transcript us-age identified two subgroups with distinct angiogenic profiles and significant survival differences. Cluster-level analysis revealed DTU in genes involved in cytoskeletal organization (ACTB), immune processes (B2M), extracellular matrix organization (FN1, APLP2), and iron metabolism (FTH1) with protein domain alterations, including the loss of actin-associated domains in ACTB and immunoglobulin-like domains in B2M. Prognostic modeling identified twelve transcripts consistently retained across bootstraps, improving risk stratification over clinical variables alone. External validation confirmed overlapping prognostic transcripts, including FGFR1 and NUCB1. Isoform-level features may serve as biomarkers and therapeutic targets in ccRCC. Statement of significanceTranscript-level analysis uncovers potential regulatory pathways in ccRCC missed by gene-level approaches, revealing isoform-specific alterations that define survival sub-groups and offer potential biomarkers and therapeutic targets.

BackgroundThe Critical Assessment of Functional Annotation (CAFA) is a community effort held to understand the field of computational protein function prediction. Every three years, since 2010, the organizers initiate an experiment to collect function predictions on a large set of proteins and then evaluate the performance of predicting methods on a subset of proteins that have accumulated experimental annotations between the submission deadline and the evaluation time. CAFA provides an independent and rigorous assessment of the current state of the art, thus leveling the playing field, highlighting successes, revealing bottlenecks, and offering a forum for the exchange of ideas in protein science. Here, we report the results of the fourth CAFA experiment (CAFA4). ResultsCAFA4 featured the participation of 148 methods from 70 research groups on a total of 46,205 unique proteins over a 5-year annotation accumulation phase, the longest in any CAFA. In a comparison across CAFA2-CAFA4 methods, the prediction of Gene Ontology (GO) terms has clearly improved across all three GO aspects and traditional evaluation settings. While not achieving the first rank, several CAFA2 and CAFA3 methods featured in the top ten methods in many evaluations, suggesting that earlier methods still hold relevance. The performance is weaker in the newly introduced "partial knowledge" evaluation category (proteins with experimental annotations before submission deadline that gained additional annotations in the same GO aspect during the annotation accumulation phase), highlighting the need for a new class of methods. The rankings of the methods were stable over the years in traditional evaluation settings, but less so in the new partial knowledge evaluation. Overall, the field continues to progress with some influx of new participants. Sustained efforts will be necessary to substantially advance it.

Most computational studies of the breast cancer immune microenvironment rely on a single deconvolution algorithm. Since different tools analyze different parameters, integrated information across a dataset can be missed. Here, we employed two complementary approaches, xCell and a CIBERSORT-style ssGSEA using LM22 signatures, on 1,099 PAM50-classified primary tumors from the TCGA-BRCA cohort. PD-L1 was highest in basal-like tumors (Kruskal-Wallis p < 0.001) and correlated with CD8+ T cells ({rho} = 0.65) and M1 macrophages ({rho} = 0.67) in that subtype, which fits the standard model of IFN-{gamma}-driven adaptive upregulation. In HER2-enriched cancers, PD-L1 tracked with both effector and regulatory populations simultaneously, while luminal tumors were largely immune-quiet. The most consequential finding involved {gamma}{delta} T cells and M2 macrophages: xCell showed a non-significant correlation ({rho} = 0.048, p = 0.11), whereas CIBERSORT-ssGSEA, using curated {gamma}{delta} gene signatures, produced a significant correlation ({rho}= 0.565, p < 2.2 x 10-{superscript 1}) that held across all five subtypes. A multivariate model explained 49% of PD-L1 variance (adjusted R{superscript 2} = 0.49), and Cox regression incorporating immune features gave a concordance of 0.60. These results suggest a baseline {gamma}{delta}-M2 immunosuppressive circuit in breast cancer, particularly in ER+ disease, that could be useful to set a therapeutic target.

Protein language models (PLMs) are shown to be powerful predictors of protein structure and function but their internal mechanisms remain poorly understood. Recent mechanistic interpretability methods have decomposed PLM representations into interpretable features, but they have not combined methods on a single biologically meaningful task. This paper tests whether an InterPLM sparse autoencoder and ProtoMech cross-layer transcoder can discover features in ESM-2 (6 layers, 8M) that can mainly discriminate between Class A {beta}-lactamase and Class B {beta}-lactamase with class C and D used as more challenging comparisons. The main goal is to find distinct features for Class A {beta}-lactamase that are not shared by other classes. We find that both methods find distinct features for Class A {beta}-lactamase, but the cross-layer transcoders show that the concepts for Class A {beta}-lactamase seems to be distributed among nodes such as in layer 4 and 6 rather than one node. We also showcase a validation framework to prevent overclaiming the role of a node, and we use it to show that several strong nodes fail in some stages of the framework meaning that they cannot be the sole node that defines Class A {beta}-lactamase.

IntroductionThe translation of biomarkers into binary clinical decisions requires the determination of precise cut-off points. This study validates the TholdStormDX v0.0.1 tool, a mathematical engine that employs Dual Annealing, 2- and 4-parameter logistic fitting, and vectorized Monte Carlo simulations for panel optimization under Boolean OR logic. MethodsThe tool was evaluated using datasets from four diagnostic domains (Pulmonary Nodules, Hepatocellular Carcinoma [HCC], Cervical Cancer, and Breast Cancer), along with a prognosis-oriented analytical context (Breast Cancer). Validation followed a strict workflow: characterization and selection of the best individual and combined thresholds in the Training (Train) and Validation (Val) sets, using the Test set in a completely independent manner, solely to assess the models performance and generalizability. ResultsThe tool enabled precise derivation of cut-off points for both individual biomarkers and multivariable combinations. Evaluation on the Test set objectively demonstrated in which scenarios a single biomarker outperforms a complex panel, promoting clinical parsimony. For example, in Breast Cancer diagnosis, an individual predictor outperformed the optimized panel (Sensitivity: 0.953 / Specificity: 0.952 in Test); conversely, in Hepatocellular Carcinoma, the multivariable combination showed superior performance compared to the single marker (Sens: 0.707 / Spe: 0.718 in Test). Additionally, the self-auditing system effectively flagged metric degradation when noisy variables were included, preventing potential issues. ConclusionTholdStormDX v0.0.1 proves to be a robust and transparent bioinformatics platform for deriving clinical thresholds. Its main contribution lies in mitigating local minima and promoting clinical parsimony, enabling researchers to objectively identify when a single biomarker is sufficient and when a panel provides real added value. Furthermore, it transforms the problem of biological noise into a safety feature: by systematically warning about algorithmic instability, it prevents overfitting and ensures the clinical viability of medical decisions. AvailabilityThe software is free and distributed under the GNU GPLv3 license. TholdStormDX v0.0.1 is written in Python, and its source code is available at the following GitHub address: https://github.com/roberto117343/TholdStormDX. Contactroberto117343@gmail.com

Structure prediction models have moved from single proteins to assemblies that include diverse biomolecules and their modifications. AlphaFold3 (AF3) and related models extended structural modelling via an all-atom framework, opening many new potential applications in structural biology. We evaluate how well the new capabilities of AF3 translate into application tasks in diverse areas: prediction of ubiquitinated protein structures, T-cell receptor (TCR)-epitope recognition, antibody-antigen complexes, protein-RNA and protein-lipid interactions. We find that, while AF3 can perform well in favourable settings, this performance is uneven across applications. In RNA-target predictions, the model confidence fails to separate genuine from decoy interaction partners and in several tasks accuracy depends on the presence of related complexes in the training set. Taken together, our assessment is more cautious than for AF2, whose gains in modelling monomers and complexes were clear and broadly generalisable. AF3s extension to new biomolecule types shows less consistent performance and generalisation. AF3 can be a powerful tool for hypothesis generation and prioritisation, but its predictions and use of confidence metrics will depend strongly on the specific application area and must be interpreted with respect to training-set overlap. We expect that the benchmarks provided here will serve for testing of future developments in the structure prediction field.

Clinical metagenomics uses sequencing for culture-independent identification of pathogens directly from clinical specimens. While a number of protocols claim to be pathogen agnostic, sensitivity for RNA viruses is likely lower than for bacteria or fungi, as it requires additional processing steps including conversion to cDNA. Sequence-independent, single-primer amplification (SISPA) was first described in 1991, yet how it preferentially enriches viral molecules has never been described. Here we propose that single-primer amplification exploits the PCR suppression effect, which selectively amplifies longer viral molecules over shorter host-derived cDNA fragments on the basis of size. This model predicts that any upstream processing step that disrupts fragment length will prevent this enrichment occurring. To test this, we systematically compared two adapter introduction strategies - during cDNA synthesis and via tagmentation - followed by single primer amplification, using the ZeptoMetrix Respiratory Panel 2.1 containing 16 RNA and 3 DNA virus strains. SISPA-based approaches recovered all of the viral genomes in the control, whereas using tagmentation to amplify cDNA recovered none. We then spiked the controls into extracted clinical samples and found that SISPA-based methods performed best in all background settings, however in high-background settings no viral genomes were recovered by any approach. Finally, using a modified SMART-9N protocol, we demonstrated that single-primer PCR is critical to overall performance, indicating that direct tagmentation of cDNA and dual-primer PCR should be avoided in protocols for clinical metagenomics where high sensitivity for RNA viruses is critical. These findings demonstrate that library preparation strategy fundamentally determines RNA virus sensitivity and offer mechanistic insights for protocol optimisation with direct relevance to clinical metagenomics.

Integrating histological images with gene expression data offers a promising approach for linking tissue morphologies to molecular signatures and improving disease subtyping. However, such integration remains challenging due to the high dimensionality of these datasets, cross-modal heterogeneity, and limited interpretability. Multivariate methods such as Sparse Canonical Correlation Analysis (Sparse CCA), Joint Nonnegative Matrix Factorisation (Joint NMF), and Angle-based Joint and Individual Variation Explained (AJIVE), have been used to address these challenges by reducing dimensionality while identifying features associated with latent factors, thereby enhancing biological interpretability. Despite increasing application in imaging-omics research, systematic comparisons of their methodological properties remain limited. Consequently, users often lack guidance on how to appropriately select these methods in practice, and these approaches are frequently treated as interchangeable despite differing modelling assumptions. Here, we use paired H&E images and gene expression data from breast cancer as a representative case study to examine the methodological characteristics, interpretability, and complementary properties of these integration approaches. Our results show that each method captures distinct yet complementary aspects of the underlying information. Although the biological findings are derived from the TCGA-BRCA datasets, the methodological insights identified here extend more broadly to imaging-omics integration studies. Overall, this comparative review highlights the strengths and limitations of each approach and outlines considerations for future methodological development.

Colorectal cancer (CRC) is challenging to track because its molecular changes are very complex as the disease progresses, creating significant challenges for robust biomarker discovery. In this study, we developed a machine learning framework by integrating monotonic progression and the StepMiner approach. We conducted external validation to identify reproducible, consistent transcriptomic biomarkers associated with CRC progression. Gene expression datasets were analyzed across four disease states from publicly available GEO: normal colon, adenoma, primary colorectal cancer, and metastasis. First, we identified genes with monotonic expression, then used the StepMiner approach to identify genes that act as switches between stages. A balanced 74-gene signature was used for machine-learning classification with a Random Forest. External validation showed strong performance in tissue-based datasets. However, tissue-derived signatures and plasma and blood-based datasets showed poor performance, highlighting biological differences between transcriptomic profiles. Cross-filtering between tissue-derived genes and blood expression datasets was performed, which resulted in the selection of 62 blood-compatible gene signatures. Leakage-free retraining on GSE164191 achieved a mean AUC of 0.868 with balanced precision. Functional enrichment analysis showed that these genes are highly active in cancer growth. Specifically, genes CBX3, S100A11, PDK4, NCOR1, and SOX4 demonstrated stable and reliable performance across the validation fold. Overall, our study presents a progression-aware transcriptomic framework for CRC biomarker discovery and demonstrates the importance of external validation. Additionally, we evaluate whether tissue-derived signatures can predict blood profiles. This proposed approach may help the future development of tissue-based diagnostics and minimally liquid-biopsy strategies for CRC. To ensure reproducibility, our proposed workflow was automated as a Nextflow pipeline. The tissue-derived model was deployed as an application utilizing Angular, ASP.NET Core, and Plumber (R).

Andes orthohantavirus (ANDV), the primary etiological agent of hantavirus pulmonary syndrome (HPS) in South America, is uniquely capable of limited human-to-human transmission, posing a significant challenge for outbreak control. Recent events, including the 2018-2019 Epuyen outbreak and the 2026 MV Hondius incident, underscore the need for rapid, lineage-specific molecular diagnostics. In this study, we present an artificial intelligence (AI)-driven framework for the design of diagnostic primers targeting the S genomic segment of the Epuyen lineage. Using an evolutionary algorithm integrated with thermodynamic evaluation via Primer3Plus, candidate primers were optimized to maximize classification accuracy while satisfying stringent biochemical constraints. The resulting primer set enables amplification of lineage-specific regions suitable for molecular characterization and surveillance. In silico validation demonstrates that the proposed primers achieve perfect discrimination between 2026 outbreak sequences and other ANDV variants. Furthermore, in silico comparison with standard protocol-based primers reveals substantially reduced sensitivity and specificity in the latter, highlighting the limitations of static diagnostic designs when applied to evolving viral populations. Overall, this work demonstrates that AI-assisted primer design provides a robust and adaptable strategy to improve viral detection, enhance outbreak tracking, and support timely public health interventions. Integrating computational optimization into diagnostic development is essential for strengthening preparedness against emerging zoonotic threats.

1.Over the last two decades, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) has become the standard method for identifying bacteria and has found a wide range of applications, especially in clinical microbiology. The methods high taxonomic resolution, minimal sample preparation, and complete, ready-to-use commercial systems, which include instrumentation, experimental protocols, spectral databases, and identification analysis software, were key factors in the success of MALDI-ToF MS as the standard for identifying microorganisms in routine diagnostic laboratories. However, despite the availability of these commercial solutions, there is also a growing need for efficient, cost-effective, vendor-neutral databases and analysis tools. These tools would enable the compilation of user-defined mass spectral databases and the testing of new analysis methods and algorithms, particularly in an academic context. To this end, MicrobeMS software has been developed to cover all stages of MALDI-ToF MS-based identification analysis. MicrobeMS is an easy-to-use desktop application for analyzing mass spectra from microorganisms and performing tasks related to spectrum database compilation. It includes routines for direct data import and export, biomarker peak searches, management of spectrum metadata, testing of spectrum quality, supervised and unsupervised identification analysis and intuitive result display. MicrobeMS is implemented in MATLAB and is freely available as MATLAB pcode for Windows and Linux, as well as a standalone application. Over the last fifteen years, the software has undergone continuous development and is now used routinely in various settings at the Centre for Biological Threats and Special Pathogens (ZBS) at the Robert Koch Institute (RKI) in Berlin, Germany, for example in supporting spectrum database compilation, to identify special or rare pathogenic bacteria by advanced identification analysis concepts, or to test in silico MALDI-ToF MS databases derived from microbial genomes. In this software publication the versatility and capabilities of MicrobeMS are demonstrated using a test data set from highly pathogenic bacteria (HPB) which has been obtained as part of a published European Union (EU)-funded External Quality Assurance Exercise (EQAE). MicrobeMS and HPB test data can both be downloaded from https://wiki.microbe-ms.com/. The goal of this software publication is twofold: to raise awareness of MicrobeMS within the scientific community and to encourage the testing of the software and custom-developed MALDI-ToF MS databases of the RKI, which are published at the ZENODO data repository (https://doi.org/10.5281/zenodo.7702374).

The prediction of phage-host interactions is key for several applications in biotechnology, medicine, and microbial ecology. Wide studies in machine learning tools have allowed the exploration of these interactions across multiple taxonomic levels. A systematic review and meta-analysis were conducted on 570 records retrieved from PubMed, Scopus, and Web of Science. Eleven studies were selected for the meta-analysis, encompassing 61 datasets. Precision across taxonomic levels (Domain, Phylum, Class, Order, Family, Genus, Species) was evaluated for several prediction tools. Statistical tests, including the Shapiro-Wilk and ANOVA tests, were used. A mixed-effects meta-regression model was used to examine the impact of taxonomic subgroups on the prediction of the proportion of Correctly Predicted PHIs. The results indicated significant variability in the performance of prediction tools across taxonomic levels. Domain-level predictions exhibited near-perfect Proportion of Correctly Predicted PHIs (0.99), whereas finer resolutions (Family and Order) showed considerable variability, with average precision values of 0.682 and 0.775, respectively. The mixed-effects meta-regression analysis revealed that Family and Species taxonomic subgroups were associated with significant reductions in the prediction Proportion of Correctly Predicted PHIs with effect sizes of -0.1464 and -0.1944, respectively. Residual heterogeneity was negligible, indicating that the moderators adequately explained the variability in prediction precision. This study highlights the importance of selecting the appropriate prediction tool based on the desired taxonomic resolution. The findings emphasize the need for further refinement of prediction algorithms, particularly at the Family and Species levels, where tools exhibit the most variability. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=136 SRC="FIGDIR/small/721508v1_ufig1.gif" ALT="Figure 1"> View larger version (39K): org.highwire.dtl.DTLVardef@4105bforg.highwire.dtl.DTLVardef@e07c46org.highwire.dtl.DTLVardef@1ff139corg.highwire.dtl.DTLVardef@1608690_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphical Abstract.C_FLOATNO Overview of the systematic review and meta-analysis framework evaluating ML-based phage-host interaction prediction tools across taxonomic levels. C_FIG

BackgroundOxford Nanopore Technologies (ONT) sequencing is increasingly used for whole-genome sequencing (WGS) across a wide range of applications. However, the platform has evolved rapidly through updates to flow cell chemistry and basecalling algorithms, altering the characteristics of the resulting sequencing data. Read simulators provide synthetic datasets with known ground truth, enabling controlled development and evaluation of methods. However, many existing simulators were developed for earlier versions of ONT sequencing or use generic long-read assumptions, and their realism for contemporary ONT data is unclear. ResultsWe benchmarked six ONT-compatible read simulators (Badread, LongISLND, lrsim, NanoSim, PBSIM3 and SimLoRD) using a microbial genome reference and ONT R10.4.1 reads as the empirical standard. Each tool was configured to maximise realism, including training on empirical reads when supported. We compared simulated and real datasets with respect to read length, read accuracy, FASTQ quality scores and sequence error profiles. No simulator reproduced all metrics of the real data well. PBSIM3 most closely reproduced read length, read accuracy and FASTQ quality scores, making it a strong simulator for broad read-level realism. However, it did not capture important features of the real error profile, including context-dependent substitution rates and homopolymer-length errors. Badread and LongISLND better reproduced some aspects of the error profile, but showed other departures from the real data. ConclusionPBSIM3 is a good general-purpose choice for many ONT WGS simulation tasks because it reproduced several key read-level properties well. However, Badread or LongISLND may be preferable for applications where error structure is more important. No evaluated tool was realistic across all tested metrics, highlighting a gap for improved long-read simulators.

The emergence of unidentified pathogens, or "Disease X," poses a significant threat to global health, necessitating the development of proactive surveillance strategies for the wildlife and human virosphere. Since novel viruses often lack universal genetic markers or known homologs, this study evaluates four reference-independent computational pipelines: coverage-based, k-mer-based, nucleotide clustering, and Large Language Model (LLM)-based designed to detect spreading organisms by comparing distinct metagenomic datasets. Using a real-world pandemic dataset of human nasopharyngeal RNA-seq runs and a semi-synthetic dataset enriched with divergent Egovirales sequences, we measured the sensitivity, selectivity, and computational efficiency of each approach. The coverage-based method proved most robust, consistently achieving 100% genome coverage of SARS-CoV-2 and maintaining high selectivity even at low viral concentrations, though it required extensive computational resources (20 days of CPU time for 2B reads). In contrast, the k-mer-based approach offered a tenfold reduction in execution time and high selectivity but was sensitive to data depletion, failing to detect targets at very low abundances. The clustering-based pipeline performed effectively at moderate concentrations but suffered from sequence fragmentation in sparse data, while the LLM-based method (using ViraLM), despite its efficiency, exhibited critically low selectivity due to current latent space partitioning limitations. These results demonstrate that while k-mer and LLM-based tools provide rapid screening capabilities, the coverage-based approach remains the most reliable for sensitive pathogen discovery. Ultimately, these reference-independent workflows are essential for illuminating metagenomic "dark matter" and establishing early warning systems for emerging infectious diseases