<?xml version="1.0" encoding="utf-8"?>
    <rss version="2.0">
      <channel xmlns:content="http://purl.org/rss/1.0/modules/content/">
        <title>Frontiers in Bioinformatics | Integrative Bioinformatics section | New and Recent Articles</title>
        <link>https://www.frontiersin.org/journals/bioinformatics/sections/integrative-bioinformatics</link>
        <description>RSS Feed for Integrative Bioinformatics section in the Frontiers in Bioinformatics journal | New and Recent Articles</description>
        <language>en-us</language>
        <generator>Frontiers Feed Generator,version:1</generator>
        <pubDate>2026-05-13T23:04:45.203+00:00</pubDate>
        <ttl>60</ttl>
        <item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1722578</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1722578</link>
        <title><![CDATA[Predicting anthropometric body composition variables using 3D optical imaging and machine learning]]></title>
        <pubdate>2026-05-13T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Gyaneshwar Agrahari</author><author>Kiran Bist</author><author>Monika Pandey</author><author>Jacob Kapita</author><author>Zachary James</author><author>Jackson Knox</author><author>Steven Heymsfield</author><author>Sophia Ramirez</author><author>Peter Wolenski</author><author>Nadejda Drenska</author>
        <description><![CDATA[Accurate prediction of anthropometric body composition variables, such as Appendicular Lean Mass (ALM), Body Fat Percentage (BFP), and Bone Mineral Density (BMD), is essential for early diagnosis of several chronic diseases. Currently, researchers rely on Dual-Energy X-ray Absorptiometry (DXA) scans to measure these metrics; however, DXA scans are costly and time-consuming. This work proposes an alternative to DXA scans by applying statistical and machine learning models on biomarkers (height, volume, left calf circumference, etc.) obtained from 3D optical images. The dataset consists of 847 patients and was sourced from the Pennington Biomedical Research Center. Extracting patients’ data in healthcare faces many technical challenges and legal restrictions. However, most supervised machine learning algorithms are inherently data-intensive, requiring a large amount of training data. To address this challenge, we compare the standard supervised to a semi-supervised p-Laplacian model, which leverages the limited data by incorporating the unlabeled patient information. To our knowledge, this paper is the first to demonstrate the application of a game-theoretic p-Laplacian model for regression in healthcare. Our p-Laplacian model yielded errors of ∼13% for ALM, ∼10% for BMD, and ∼20% for BFP when the training data accounted for 10 percent of all data. Among the supervised algorithms we implemented, Support Vector Regression (SVR) performed the best for ALM and BMD, yielding errors of ∼8% for both, whereas Least Squares SVR performed the best for BFP with ∼11% error when trained on 80% the data. Our findings position the p-Laplacian model as a promising tool for healthcare applications, particularly in a data-constrained environment with limited labeled data.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1795889</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1795889</link>
        <title><![CDATA[Comparative transcriptomics and computational drug discovery identify ASPM as a key oncogenic driver and therapeutic target in hepatocellular carcinoma]]></title>
        <pubdate>2026-05-07T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Pan Li</author><author>Aiye Guo</author><author>MingJing Zhao</author><author>GuangHui Chen</author>
        <description><![CDATA[IntroductionHepatocellular carcinoma (HCC) is a highly heterogeneous malignancy that necessitates the identification of robust biomarkers across diverse populations to enhance diagnostic and prognostic precision. This study aimed to identify clinically relevant biomarkers and potential therapeutic targets through integrative transcriptomic and computational analyses.MethodsA comparative transcriptomic analysis was performed on 724 HCC samples obtained from four independent cohorts (United States, South Korea, France, and Taiwan). Differential expression and survival analyses, including Kaplan–Meier estimation, were conducted to evaluate clinical significance. Functional enrichment analysis was used to explore biological roles. Structural modeling, molecular docking, 100-ns molecular dynamics (MD) simulations, MM-GBSA binding energy calculations, and in silico ADMET profiling were employed to assess ligand–target interactions.ResultsAbnormal Spindle Microtubule Assembly (ASPM) was consistently overexpressed across all cohorts and significantly associated with poor overall survival. Functional analyses indicated its involvement in mitotic spindle organization and homologous recombination–mediated DNA repair. Among screened compounds, Mol-7424 exhibited stable binding within the ASPM calponin domain, favorable binding free energy, and promising pharmacokinetic properties. Lipid bilayer simulations further supported its membrane permeability and potential cellular uptake.DiscussionThese findings highlight ASPM as a prognostic biomarker and potential therapeutic target in HCC. Mol-7424 emerges as a promising lead compound; however, its efficacy requires validation through in vitro and in vivo studies. Overall, this study underscores the utility of multi-population transcriptomics integrated with computational approaches for advancing precision oncology in HCC.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803111</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803111</link>
        <title><![CDATA[Delineating novel diagnostic biomarkers and therapeutic targets for oral submucosal fibrosis: an integrative multi-omics and machine learning approach]]></title>
        <pubdate>2026-05-04T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Chinmay Nitin Mokal</author><author>Piyush Agrawal</author>
        <description><![CDATA[BackgroundOral submucosal fibrosis (OSF) is a chronic and progressive disorder, caused by chewing areca nuts, affecting the oral cavity and oropharynx. OSF is characterized by severe symptoms like severe burning sensation, restricted mouth opening, etc. Given the multifactorial and poorly understood molecular basis of the disease, there is a need for novel biomarkers and therapeutic targets.MethodWe downloaded 3 RNA-seq, two microarray, one epigenomic, and one single-cell RNA-seq datasets from the gene expression omnibus database. Differentially expressed genes (DEGs) were characterized using DESeq2. Several analyses, including gene enrichment, immune cell infiltration, protein-protein interaction, and more, were performed. Machine learning models were developed using all DEGs and top5 selected features with leave one out cross validation technique. Independent validations were performed using two microarray datasets with appropriate statistical measures. Epigenetic analysis revealed hyper- and hypomethylated genes based on delta-beta values, and an integrative analysis of the transcriptome and methylome was performed to obtain key biomarkers. Single-cell analysis was performed to identify key cell types showing higher DEG expression.ResultDESeq2 analysis identified 29 upregulated and 15 downregulated DEGs. Upregulated DEGs show enrichment for the inflammatory, metabolic, and signaling processes, whereas downregulated DEGs were largely associated with metabolic processes. Immune cell enrichment analysis using CIBERSORTx shows higher enrichment of “T cells,” “mast cells,” and “macrophages” in OSF patients. We validated our findings in two independent microarray datasets and observed a similar gene expression pattern of DEGs. Machine learning performed using top5 features where Random Forest model achieved AUROC of 0.99 and AUPRC of 0.99. Further, ROC analysis and AUC plot show that DEGs can discriminate OSF patients from the normal population with high AUROC. Integrative analysis of methylation and transcriptomic data identified 11 genes as potential diagnostic biomarkers and therapeutic targets. Finally, single-cell analysis elucidates the higher expression of DEGs in “keratinocyte”, “epithelial cells” and “dendritic cells”.ConclusionIntegrative analysis identified 11 gene signatures as potential early diagnostic biomarkers and therapeutic targets for the OSF. Furthermore, the study hints towards mechanistic insight into potential mechanism leading to oral cancer. All the codes and ML models are provided at our GitHub repository https://github.com/agrawalpiyush-srm/OSF.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1811916</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1811916</link>
        <title><![CDATA[On the optimization of copy number variations representation in pangenome graphs]]></title>
        <pubdate>2026-05-01T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Mirko Coggi</author><author>Lorenzo Basile</author><author>Beatrice Branchini</author><author>Gabriele Amodeo</author><author>Guido Walter Di Donato</author><author>Marco D. Santambrogio</author>
        <description><![CDATA[Graph-based pangenome references often misrepresent Copy Number Variations (CNVs) and Variable Number Tandem Repeats (VNTRs) as alternative acyclic paths, which hinders downstream analyses, degrades alignment behavior, and reduces interpretability in graph visualizations. For these reasons, we introduce PANPHORTE, a topology-optimization methodology that detects repeat-driven misrepresentations within superbubbles and rewrites them into structures that more faithfully reflect the underlying biology. Given a pangenome graph annotated with haplotype paths, PANPHORTE identifies repetitive elements inside superbubbles, isolates shared repeat sequences across distinct subpaths, and refactors the graph by splitting nodes and introducing explicit cycles, encoding CNVs and VNTRs without loss of information. We provide a C++ command-line implementation of the proposed specifications, and a complementary pipeline that applies PANPHORTE followed by GFAffix to further reduce redundancy in regions not affected by repeat-induced artifacts. We evaluate PANPHORTE on synthetic and real pangenome graphs, showing reductions in memory footprint of up to 71.69%, improvements in exact read matches of up to 34.4%, and substantially clearer visual identification of repeated loci.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1841924</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1841924</link>
        <title><![CDATA[Correction: An integrated automated deep learning framework for annotating tumor-infiltrating lymphocytes in lung adenocarcinoma pathology]]></title>
        <pubdate>2026-04-30T00:00:00Z</pubdate>
        <category>Correction</category>
        <author>Xia Li</author><author>Kang-Lai Wei</author><author>Zhao-Quan Huang</author><author>Zi-Yan Huang</author>
        <description></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1784287</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1784287</link>
        <title><![CDATA[Comprehensive genomics, systems, and structural assessment for novel target identification in penicillin-resistant Streptococcus pneumoniae]]></title>
        <pubdate>2026-04-29T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Avani Panickar</author><author>Suvitha Anbarasu</author><author>Anand Manoharan</author><author>Sudha Ramaiah</author><author>Anand Anbarasu</author>
        <description><![CDATA[IntroductionThe increasing prevalence of penicillin-resistant Streptococcus pneumoniae (PRSP) has compromised the efficacy of conventional β-lactam therapies, and the inefficiency of penicillin-binding proteins (PBPs) as reliable drug targets further underscores the urgent need to explore novel alternatives. The current study employs an in silico strategy that integrates genomics, genome-wide association studies (GWASs), network analyses, and membrane protein simulations to systematically identify and prioritize new antimicrobial targets.MethodologyA total of 665 PRSP genomes from Indian clinical isolates collected between 1996 and 2022 were analyzed. High-quality genome assemblies were annotated and used for pangenome construction and GWASs to identify gene clusters associated with penicillin resistance. Candidate genes were further prioritized through essentiality screening, functional annotation, subcellular localization prediction, evolutionary conservation analysis, druggability assessment, and structural modeling.ResultsIntegrated analysis identified OppC2, an essential oligopeptide permease of the ABC transporter family, as a highly favorable drug target. Network and functional enrichment analyses linked OppC2 to transport-associated pathways relevant to pneumococcal survival and adaptation. Structural modeling revealed a high-confidence protein model with a druggable binding pocket, while molecular dynamics simulations confirmed the stability of the structure in a physiological membrane environment.ConclusionThis comprehensive approach enabled the identification of conserved, essential, and accessible drug targets within PRSP populations, providing an adaptable framework to guide next-generation antimicrobial target identification beyond traditional PBPs.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803572</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803572</link>
        <title><![CDATA[Public health risk stratification using hybrid machine learning: a reproducible analysis of performance, stability, and risk attribution]]></title>
        <pubdate>2026-04-23T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Alejandro Cabrera-Andrade</author><author>Ana Karina Zambrano</author><author>Joselin García-Ortiz</author><author>William Villegas-Ch</author>
        <description><![CDATA[Risk stratification in public health involves organizing heterogeneous health-related signals into consistent representations that support population-level analysis. In large-scale datasets, such as National Health and Nutrition Examination Survey (NHANES) and Behavioral Risk Factor Surveillance System (BRFSS), the integration of clinical, biometric, behavioral, and self-reported variables introduces structural variability that challenges conventional modeling approaches. This study proposes a hybrid learning framework that combines linear and nonlinear components to analyze induced risk representations derived from multidimensional health data. The model is evaluated using NHANES 2017–2018, BRFSS 2019, and an Integrated Public Health Dataset constructed through semantic harmonization of both sources. The experimental design is based on a controlled formulation in which a continuous risk index is constructed from the available variables and discretized into ordinal classes using quantiles, enabling systematic analysis of how models approximate structured partitions of the input space rather than predicting independent clinical outcomes. The results show that the hybrid scheme maintains consistent macro F1 and macro-ROC-AUC values across all scenarios with low fold-to-fold variability, reflecting the regularity of the induced class structure rather than predictive generalization. Attribution analysis reveals that the organization of the risk representation varies according to the nature of the data, with concentrated patterns in clinical signals, distributed contributions in behavioral variables, and intermediate structures in the integrated dataset. These findings demonstrate that hybrid schemes provide a stable and interpretable framework for analyzing the structural organization of risk in heterogeneous public health data.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803237</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1803237</link>
        <title><![CDATA[How benchmarking of bioinformatics tools is essential for informed workflow selection: a case study on SARS-CoV-2 subgenomic RNA detection]]></title>
        <pubdate>2026-04-22T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Gabriele Leoni</author><author>Mauro Petrillo</author><author>Man-Hung Eric Tang</author><author>Soren Alexandersen</author>
        <description><![CDATA[IntroductionSelecting appropriate bioinformatics tools is critical for accurate and reproducible analysis, particularly in support of genomic surveillance and molecular biomarker monitoring. The importance of these analyses is underscored by the need for effective public health responses to emerging diseases like SARS-CoV-2.MethodsBy using the detection of SARS-CoV-2 subgenomic RNAs (sgRNAs) as a case study, we show the importance of systematic benchmarking in selecting optimal workflows. We generated 25 synthetic Illumina datasets simulating both shotgun and amplicon sequencing strategies, along with a real-world wastewater dataset. Using these datasets, we assessed the influence of key variables including mutation profiles, read lengths, aligner choice, and primer design for targeted sequencing.ResultsOur results revealed substantial performance variability: common tools developed to identify sgRNAs struggled with shotgun data and were sensitive to mutations depending on the chosen aligner, while amplicon sequencing improved detection sensitivity, with aligners and primer design choices still significantly impacting outcomes.DiscussionOur results highlight the need for benchmarking steps and analyses to inform workflow selection. Without such evaluations, researchers risk drawing inaccurate conclusions from suboptimal workflows. This case study underscores the value of context-aware tool selection and encourages standardised benchmarking practices to ensure reproducibility and reliability in bioinformatics analysis, particularly in evidence-based decision-making environments such as public health and policymaking.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1800237</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1800237</link>
        <title><![CDATA[Feature representation for explainable CRISPR off-target prediction and base editing efficiency]]></title>
        <pubdate>2026-04-20T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Faiza Hasin</author><author>Michele Minervini</author><author>Corrado Mencar</author><author>Giuseppe Ventrella</author><author>Arianna Consiglio</author><author>Alessandro Orro</author><author>Tommaso Selmi</author>
        <description><![CDATA[IntroductionThe interaction between guide RNAs (gRNAs) and target DNA sequences is a critical factor in the effectiveness of CRISPR/Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated protein 9) gene editing. Predicting these interactions accurately necessitates models that offer biological knowledge in addition to high accuracy. This study analyzes the impact of feature representation on accuracy and interpretability in off-target prediction.MethodsWe address two CRISPR applications: gene knockout (KO) and base editing (BE) using distinct benchmark datasets. For the KO problem, we utilized CHANGE-seq and GUIDE-seq to evaluate paired sequence representations, while the Hanna screening dataset has been used for BE. We approached the prediction problem both as a classification and regression task using XGBoost models.ResultsIn the case of KO, there is not a single universally optimal encoding. For both classification and regression, One-Hot and its variants (OH, OH5C) achieve the best results on GUIDE-seq (AUPR = 0.661, Pearson = 0.756), while the Bulges representation performs best on CHANGE-seq (AUPR = 0.612, Pearson = 0.602). In the case of BE, One-hot encoding consistently outperforms K-mer representation for predictive accuracy both as regression and classification (AUPR = 0.723, Pearson = 0.746).DiscussionOur analysis demonstrates comparable predictive performance across both gene knockout and base editing tasks, confirming the robustness of the framework in distinct editing domains. Interpretability analysis using SHapley Additive exPlanations (SHAP) reveals that despite different mechanisms, the Protospacer Adjacent Motif (PAM)-proximal region remains a critical feature for prediction for both editing mechanisms.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1806975</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1806975</link>
        <title><![CDATA[Geometric multidimensional representation of omic signatures]]></title>
        <pubdate>2026-04-17T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Higor Almeida Cordeiro Nogueira</author><author>Enrique Medina-Acosta</author>
        <description><![CDATA[IntroductionMulti-omic signatures are widely used in biomarker discovery, precision oncology, and systems biology, yet they are typically treated as vectors or composite scores that collapse intrinsically multidimensional biological organization into one-dimensional summaries. As a result, their internal structure, contextual dependencies, and functional coherence remain largely inaccessible.MethodsHere, we introduce a geometric framework that reconceptualizes omic signatures as multidimensional informational entities whose biological meaning arises from structural organization rather than molecular membership alone. Each signature is embedded in a shared latent space integrating regulatory, phenotypic, microenvironmental, immune, and clinical constraints, and represented as a convex polytope. This representation preserves internal organization and enables intrinsic geometric measurements—including barycenter distance, volume, anisotropy, and asymmetry—that quantify concordance, divergence, and latent complexity. We applied this framework to 24,796 metabolic regulatory circuitries reconstructed across 32 TCGA cancer types, encoded as paired regulatory and metabolic signatures in an 18-dimensional latent space.ResultsGeometric analysis shows that discordance predominates: most circuitries occupy strong or extreme discordance regimes and display high-dimensional, frequently asymmetric geometries, whereas fully concordant circuitries are rare and structurally constrained. These geometric phenotypes stratify metabolic pathways and superfamilies in reproducible, non-uniform patterns that are not readily captured by conventional vector- or network-based representations.DiscussionBy transforming omic signatures into measurable geometric objects, this framework provides a principled approach for the comparison and de-redundancy of multi-omic biomarkers, providing a scalable method for analyzing complex regulatory systems across cancer and beyond. All geometric representations and derived descriptors are available through the SigPolytope Shiny application (https://sigpolytope.shinyapps.io/geometricatlas/).]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1810835</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1810835</link>
        <title><![CDATA[Self-organizing maps for allele specific expression data reconstruction and identification of anomalous genomic regions]]></title>
        <pubdate>2026-04-14T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Roberto Pagliarini</author><author>Francesco Nascimben</author><author>Alberto Policriti</author>
        <description><![CDATA[Allele Specific Expression data quantifies expression variation between the two haplotypes of a diploid individual distinguished by heterozygous sites. Current methodologies of genome-wide sequencing produce large amounts of missing data that may affect statistical inference and bias the outcome of experiments. Machine learning tools could be employed to explore the data and to estimate missing signatures. We present a two-phase procedure based on Self-Organizing Maps (SOMs), an unsupervised clustering technique, to recover missing allele specific expression data from RNA-seq experiments. Specifically, a SOM trained on a complete population P is used to assign a so-called corrupted individual p̂ to its most fitting cluster c; then, a completion rule based on allele frequencies within the subpopulation of Pc⊆P defined by c is employed to reconstruct p̂. To evaluate our approach, we first apply it to purely artificial datasets, in order to have full control over all experimental conditions. After that, we consider a real population of Vitis vinifera, which we also extend by applying a computational framework to generate synthetic individuals from allele expression data. We then introduce two local feature relevance indices in order to assess the influence of specific alleles on the topological placement of corrupted individuals in the SOM structure. Our results, showing promising accuracy in the prediction of missing alleles, suggest that the developed approach could be very useful for recovering incomplete samples in a dataset instead of discarding them, mainly in situations where experiments are challenging.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1842628</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1842628</link>
        <title><![CDATA[Correction: Human-associated NDM-5-producing multidrug-resistant Escherichia coli detected in retail beef and pork in Hungary, 2021]]></title>
        <pubdate>2026-04-13T00:00:00Z</pubdate>
        <category>Correction</category>
        <author>Frontiers Production Office </author>
        <description></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1719700</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1719700</link>
        <title><![CDATA[Boolean network and meshless simulations for the comparison of transport and reaction mechanisms arising in one-short tri-exponential and uniform infusion electrochemotherapeutic treatments]]></title>
        <pubdate>2026-03-25T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Fabián Mauricio Vélez Salazar</author><author>Iván David Patiño Arcila</author><author>Ismael E. Rivera Madrid</author><author>Marlon Rincón Fulla</author>
        <description><![CDATA[Drug administration via the bloodstream involves some transport and reaction mechanisms (RTMs), such as extravasation, perfusion along blood vessels, transmembrane and interstitial transport, protein dissociation and association, and lymphatic drainage. These RTMs can be influenced by the type of pharmacokinetic (PK) profile used for drug delivery in the circulatory system, as well as by the bloodstream velocity λinl. In electroporated tissues, the electric field magnitude (E) can also affect the RTMs because it brings about vessel vasoconstriction, cell membrane and vessel wall electro-permeabilization, and changes in tissue porosity. In the present work, in-house computational tools are employed to examine how the combination of E and λinl influences the RTM′s existence, interaction, and rates arising in electrochemotherapy for two different PKs: One-short tri-exponential (TPK), where the drug concentration decreases exponentially after a one-short infusion, and one uniform (UPK), where the drug concentration is kept constant during the whole treatment. First, the ratios between extracellular, free intracellular, and bound intracellular concentrations are obtained from numerical simulations with a meshless code previously developed, calibrated, and validated. Subsequently, the interaction between the RTMs is investigated by means of a Boolean model presented here that is based on the comparison of the spatio-temporal evolution of the concentration ratios. Several combinations of E (0 kV/m; 46 kV/m; 70 kV/m), λinl (1×10−4m/s; 1×10−3m/s; 1×10−2m/s), and PK (TPK and UPK) are tested. The in silico findings indicate that RTM′s existence, interaction, and rates can vary between the two PKs (TPK and UPK) for a specific permutation of E and λinl. Nevertheless, common features are identified between these pharmacokinetic profiles. In general, the lower E, the more uniform the transmembrane transport in the radial and axial direction; the decrease of λinl also improves the radial homogeneity of this transport mechanism but negatively influences the axial uniformity. The uniformity of the mechanisms of association and dissociation is only altered monotonously by E (Vélez Salazar and Patiño Arcila, 2025).]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1793862</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1793862</link>
        <title><![CDATA[Human-associated NDM-5-producing multidrug-resistant Escherichia coli detected in retail beef and pork in Hungary, 2021]]></title>
        <pubdate>2026-03-24T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Mirena Ivanova</author><author>Joana Mourão</author><author>Judit Szarvas</author><author>Elif S. Tosun</author><author>Niamh Lacy-Roberts</author><author>Natasia Rebekka Thornval</author><author>Zita Záborcki</author><author>Szilárd Jánosi</author><author>Raquel Garcia-Fierro</author><author>Pierre-Alexandre Beloeil</author><author>Ernesto Liebana</author><author>Beatriz Guerra</author><author>Rene S. Hendriksen</author><author>Jette S. Kjeldgaard</author>
        <description><![CDATA[BackgroundCarbapenem-resistant Enterobacterales pose a significant public health threat, particularly when detected in food-producing animals and retail meat. Although carbapenems are not used in European Union animal production, sporadic cases of carbapenemase-producing Escherichia coli have emerged across multiple European countries since 2019. The detection of human-associated carbapenemase genes in meat raises concerns about potential transmission to humans through the food chain.MethodsIn this study, we characterize three multidrug-resistant (MDR) E. coli isolates harboring blaNDM-5 recovered from retail beef and pork in Hungary in 2021. E. coli isolates were subjected to phenotypic antimicrobial susceptibility testing using broth microdilution, conjugation experiments, and genotypic characterization through whole-genome sequencing using Illumina and Oxford Nanopore platforms. Hybrid assemblies enabled comprehensive comparative genomic and plasmid analyses.ResultsAll three isolates belonged to the human-associated uropathogenic clone ST405 (O102:H6) and were clonally related with a maximum of two single nucleotide polymorphisms. They exhibited identical genomic profiles conferring resistance to carbapenems, cephalosporins, fluoroquinolones, tetracycline, and azithromycin. Comparative genomic analysis revealed close genetic relationships with human clinical isolates from Australia and the United Kingdom, suggesting international dissemination. The blaNDM-5 gene was located on conjugative IncFII-IncFIB hybrid plasmids (approximately 132 kb) closely related to clinical plasmids from human isolates in the United States, differing only by the absence of a blaCTX-M-15-ISEcp1 transposition unit.ConclusionThe detection of human-associated blaNDM-5-carrying E. coli ST405 in retail meat represents a serious food safety concern, highlighting potential transmission routes to humans and emphasizing the need for enhanced surveillance and epidemiological investigations.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1792643</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1792643</link>
        <title><![CDATA[Forest-EMCBE: an evolutionary ensemble learning algorithm for multiclass diagnosis of bacterial pneumonia using the CBC dataset]]></title>
        <pubdate>2026-03-18T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Yimin Shen</author><author>Xiaotian Xu</author><author>Xiaoxi Hao</author><author>Cuimin Sun</author><author>Wei Lan</author>
        <description><![CDATA[IntroductionRapid diagnosis of bacterial pneumonia is crucial for clinical diagnosis and treatment, but traditional methods are time-consuming. The wide application of machine learning techniques in medical diagnosis provides an effective way to solve this problem. However, the complexity of medical datasets and the problem of class imbalance poses serious challenges to classical machine learning algorithms.MethodsAiming at the multiclass imbalanced problem in complete blood count (CBC) datasets, this study proposes a novel ensemble learning algorithm, Forest of Evolutionary Multi-Classifiers Based on Bagging with Error-Correcting Output Coding (Forest-EMCBE). The algorithm integrates Multi-Objective Genetic Algorithm, Error-Correcting Output Codes (ECOC), and balanced sampling strategy, which enhances the generalization ability of the classifiers through a three-layer integrated structure.ResultsTo validate the effectiveness of the proposed method, we trained the diagnostic model on a CBC dataset, which contains 1,457 samples and 4 different classes of bacterial pneumonia results, and compared it with 11 state-of-the-art algorithms. The experimental results demonstrate the superior performance of the Forest-EMCBE algorithm on the CBC dataset, outperforming all other compared algorithms.DiscussionBased on the Shapley value-based feature importance analysis method, this study dissects the contributions of key features to the prediction outcomes and further elucidates the differential impacts of features such as age, gender, and neutrophil percentage on predicting infections by different bacterial species.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1784011</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1784011</link>
        <title><![CDATA[A multi-epitope pan-betacoronavirus vaccine construct predicted to induce broad-spectrum and durable immune responses: an immunoinformatics approach]]></title>
        <pubdate>2026-03-18T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Anabella Margareth Arapa</author><author>Trina Ekawati Tallei</author><author>Rinaldi Idroes</author><author> Fatimawali</author><author>Elly Juliana Suoth</author><author>Maghfirah Savitri</author><author>Ahmad Akroman Adam</author><author>Beivy Jonathan Kolondam</author><author>Chika Yamada</author><author>Rosy Iara Maciel de Azambuja Ribeiro</author><author>Amama Rani</author><author>Moon Nyeo Park</author><author>Youdiil Ophinni</author><author>Bonglee Kim</author>
        <description><![CDATA[IntroductionRecurrent zoonotic spillovers and continuous antigenic evolution among betacoronaviruses, including SARS-CoV, MERS-CoV, and SARS-CoV-2, highlight the urgent need for a broad-spectrum vaccine capable of eliciting cross-protective immunity. Conventional vaccines, although effective against specific strains, may be limited by antigenic mismatch and waning immunity. This study aimed to design a multi-epitope pan-betacoronavirus vaccine targeting conserved regions within the receptor-binding domain (RBD) using an integrated immunoinformatics and reverse vaccinology framework.MethodsCytotoxic T-lymphocyte (CTL), helper T-lymphocyte (HTL), and linear B-cell epitopes were predicted and screened for antigenicity, allergenicity, toxicity, and non-homology to host proteins. Selected epitopes were assembled into a 285–amino acid multi-epitope construct using optimized linkers (AAY, GPGPG, EAAAK, and GGGGS) and human β-defensin 3 as an adjuvant. Structural modeling and refinement were performed to generate a three-dimensional vaccine model, followed by molecular docking with a B-cell receptor (BCR) Fab model using ClusPro. Molecular dynamics simulations were conducted to evaluate structural stability, and immune responses were assessed through computational immune simulation.ResultsThe refined vaccine construct produced a stable structural model with a C-score of −3.60. Molecular docking identified a highly ranked complex from a well-populated cluster (Cluster 1; 49 members) with a Lowest Energy score of −865.9, indicating favorable interface complementarity under the docking scoring function. Molecular dynamics simulation over 100 ns supported the structural integrity and dynamic stability of the complex, with minimal backbone deviation and sustained intermolecular interactions. Immune simulations predicted coordinated humoral and cellular responses following a simulated prime–boost regimen, including increased antibody titers, elevated IL-2 and IFN-γ levels, and sustained memory B- and T-cell populations. The selected epitope set showed an estimated global HLA population coverage of 93.28%.ConclusionThis study identifies a promising in silico multi-epitope RBD-based pan-betacoronavirus vaccine candidate with predicted broad HLA population coverage and favorable structural stability. These findings provide a computational basis for subsequent experimental validation of the construct’s immunogenicity, safety, and potential cross-protective capacity in relevant in vitro and in vivo models.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1764743</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1764743</link>
        <title><![CDATA[An integrated automated deep learning framework for annotating tumor-infiltrating lymphocytes in lung adenocarcinoma pathology]]></title>
        <pubdate>2026-03-16T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Xia Li</author><author>Kang-Lai Wei</author><author>Zhao-Quan Huang</author><author>Zi-Yan Huang</author>
        <description><![CDATA[ObjectiveQuantitative analysis of tumor-infiltrating lymphocytes (TILs) is crucial in computational pathology studies of lung adenocarcinoma. However, acquiring large-scale, fully annotated datasets remains a major obstacle for the supervised learning approaches that currently dominate high-precision modeling. To address this data bottleneck, we developed a fully automated pipeline for the precise annotation of tissue contours, tumor parenchyma, and lymphocytes in whole-slide images (WSIs).MethodsThis study utilized WSI data from The Cancer Genome Atlas (TCGA) cohort, with comprehensive manual annotations performed by two pathologists using QuPath software, with all annotations subsequently reviewed by a third senior pathologist. The resulting training dataset comprised over 20,000 annotated units. These annotated data were used to train three core modules consisting of an OpenCV-based image processing pipeline for tissue contour detection, a lightweight U2-NetP model for tumor parenchyma segmentation, and a YOLOv7 object detection framework for TILs identification within stromal regions. The pipeline was rigorously validated on both an independent internal cohort and an external hospital cohort, and its outputs were benchmarked against semi-quantitative assessments from expert pathologists.ResultsThe pipeline demonstrated robust and generalizable performance. For tissue contour detection, the OpenCV-based pipeline achieved a Dice coefficient of 90.90% on the test set. For the core learning-based tasks, the tumor parenchyma segmentation model achieved a Dice coefficient of 87.17% on the internal test set and maintained consistent accuracy on the external cohort, with Dice coefficients ranging from 0.8509 to 0.9178. In the particularly challenging task of lymphocyte detection, the YOLOv7-based model attained an F1-score of 78.84% and mAP@0.5 of 81.16% on the test set, with performance sustained on external data. Critically, the automated TILs quantifications showed excellent agreement with independent pathologist assessments (ICC >0.96). The implementation of optimized lightweight architectures enables the pipeline to serve as an accessible solution for large-scale WSIs analysis in computational pathology.ConclusionThis study has successfully developed a fully automated annotation pipeline for lung adenocarcinoma WSIs. By generating high-quality annotations of stromal TILs, this pipeline establishes a reliable data foundation for subsequent computational pathology research and facilitates the advancement of artificial intelligence applications in pathology.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1806001</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1806001</link>
        <title><![CDATA[Editorial: Integrating machine learning and AI in biological research: unraveling complexities and driving advancements]]></title>
        <pubdate>2026-03-10T00:00:00Z</pubdate>
        <category>Editorial</category>
        <author>Bindu Nanduri</author><author>Inimary Toby-Ogundeji</author>
        <description></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2026.1715377</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2026.1715377</link>
        <title><![CDATA[Evaluating transcriptomic integration for cyanobacterial constraint-based metabolic modelling]]></title>
        <pubdate>2026-02-04T00:00:00Z</pubdate>
        <category>Original Research</category>
        <author>Thomas Pugsley</author><author>Guy Hanke</author><author>Christopher D. P. Duffy</author>
        <description><![CDATA[Metabolic modelling has wide-ranging applications, including for the improved production of high-value compounds, understanding complex diseases and analysing microbial community interactions. Integrating transcriptomic data with genome-scale metabolic models is crucial for deepening our understanding of complex biological systems, as it enables the development of models tailored to specific conditions, such as particular tissues, environments, or experimental setups. Relatively little attention has been given to the validation and comparison of such integration methods in predicting intracellular fluxes. While a few validation studies offer some insights, their scope remains limited, particularly for organisms like cyanobacteria, for which little metabolic flux data are available. Cyanobacteria hold significant biotechnological potential due to their ability to synthesise a wide range of high-value compounds with minimal resource inputs. Using existing transcriptomic data, we evaluated different methodological options that can be taken when integrating transcriptomics with a genome-scale metabolic model of Synechocystis sp. PCC 6803 (iSynCJ816), when predicting autotrophic flux distributions. We find METRADE* (using single objective optimisation) to be the best-performing method in cyanobacteria owing to its ability to perform well across both metrics but emphasise the importance of configuration and scaling in achieving these outcomes.]]></description>
      </item><item>
        <guid isPermaLink="true">https://www.frontiersin.org/articles/10.3389/fbinf.2025.1721028</guid>
        <link>https://www.frontiersin.org/articles/10.3389/fbinf.2025.1721028</link>
        <title><![CDATA[Essential nucleic acid omics: a theoretical foundation for early-stage users]]></title>
        <pubdate>2026-02-04T00:00:00Z</pubdate>
        <category>Review</category>
        <author>Andrew J. Maritan</author><author>Frank J. Stewart</author>
        <description><![CDATA[Modern biology often relies on the analysis of entire sets of molecules (omics). A subset of omics uses nucleic acid sequencing to reconstruct genomes and profile gene expression. Novel findings and existing data are contextualized by databases, which have been growing exponentially due to falling sequencing costs and increased computing access. The increasing accessibility of omics has led to rapid adoption and widespread self-training via open-access tools. In this training environment new users (many of whom are students also applying computing for the first time) are confronted with Terabytes of sequence data and an ocean of topic-specific computing guides (often directed at high-level users). This flood of information creates an initial barrier of confusion and frustration, where it is challenging to identify the overarching goals of omics analyses through the details of computing. We believe this confusion is understandable but not pre-destined, as omics is–at its core–simple. This simplicity comes from its modular nature, where any analysis requires familiarity with only a few consistent steps. Here, we identify core elements of all omics analyses–data products, tools, and workflows–using microbiology applications to ground the discussion. This structure is informed by first-hand experience training early-stage omics users, where covering omics theory provides a foundation for practical implementation.]]></description>
      </item>
      </channel>
    </rss>