Bioinformatics advances最新文献

{"title":"unsurv: clustering individualized survival curves.","authors":"Imad El Badisy","doi":"10.1093/bioadv/vbag218","DOIUrl":"10.1093/bioadv/vbag218","url":null,"abstract":"<p><strong>Motivation: </strong>Biomedical phenotyping workflows often cluster high-dimensional molecular or clinical covariates and then test survival differences post hoc. This can miss structure that is expressed primarily through the timing of risk, because two patients may have similar scalar risk at one horizon but different predicted survival trajectories. Many survival learners can output patient-specific curves <math> <mrow> <mrow> <msubsup> <mrow><mrow><mo>{</mo> <mrow> <msub> <mrow> <mrow> <mrow> <mover><mrow><mi>S</mi></mrow> <mo>^</mo></mover> </mrow> </mrow> </mrow> <mi>i</mi></msub> </mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> <mo>}</mo></mrow> </mrow> <mrow><mi>i</mi> <mo>=</mo> <mn>1</mn></mrow> <mi>n</mi></msubsup> </mrow> </mrow> </math> on a common time scale. Clustering these curves gives a survival-native post-prediction representation for exploratory phenotyping. Existing work on survival-curve clustering mainly targets predefined Kaplan-Meier curves or general functional data, whereas a practical tool for individual-level predicted survival curves remains less developed.</p><p><strong>Results: </strong>We introduce unsurv, an R package that clusters individualized survival curves on a common time grid using validity-preserving preprocessing, time-weighted integral distances, and partitioning around medoids (PAM). The output includes observed medoid prototype curves, a nearest-medoid rule for assigning new patients, model-selection summaries, resampling-based stability using the Adjusted Rand Index (ARI), and a new unsurv_compare() function for comparing partitions against baselines. The package is model-agnostic and can be paired with any survival learner that returns <math> <mrow> <mrow> <msub> <mrow> <mrow> <mrow> <mover><mrow><mi>S</mi></mrow> <mo>^</mo></mover> </mrow> </mrow> </mrow> <mi>i</mi></msub> </mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo></mrow> </math> . We illustrate the workflow on a synthetic example and on the METABRIC breast cancer cohort using survdnn-predicted survival curves, including baseline comparisons and an out-of-sample validation check showing that the curve-based cluster ordering generalizes to held-out patients.</p><p><strong>Availability and implementation: </strong>The unsurv package is available on CRAN and its full source code and documentation are provided at https://github.com/ielbadisy/unsurv under the MIT License.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag218"},"PeriodicalIF":2.6,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13524423/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148851961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating transformer-based models for structural characterization of orphan proteins.","authors":"Ercan Seçkin, Dominique Colinet, Etienne G J Danchin, Edoardo Sarti","doi":"10.1093/bioadv/vbag215","DOIUrl":"10.1093/bioadv/vbag215","url":null,"abstract":"<p><strong>Motivation: </strong>Transformer-based models (TBMs) are state-of-the-art deep learning architectures that predict protein structural features with high accuracy. Despite methodological differences, they all rely on large datasets structured in families of homologous sequences. However, 5%-30% of eukaryotic proteomes consist of orphan proteins, which are sequences without detectable similarity to known families. Although they may share structural traits with characterized proteins, their lack of homology makes them an ideal dataset for evaluating TBM generalization beyond familiar sequence space.</p><p><strong>Results: </strong>We compared predictions from several widely used TBM architectures on an expert-curated set of orphan proteins from the Meloidogyne genus, comprising some of the most destructive plant-parasitic nematodes. Multiple sequence alignment-based approaches such as AlphaFold2 performed poorly on orphan proteins, as did single-sequence or embedding-based language models ESMFold, OmegaFold, and ProtT5. This limited performance cannot be fully attributed to intrinsic disorder, as confirmed by independent non-TBM disorder predictors. While accurate tertiary structure prediction remains out of reach, secondary structure is more reliably captured: predictors share about 70% of secondary structure elements, regardless of global fold similarity, and these elements are consistently identified by dedicated secondary structure tools.</p><p><strong>Availability: </strong>All data and analysis scripts are available at https://doi.org/10.5281/zenodo.18788931.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag215"},"PeriodicalIF":2.6,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13563664/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148957334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bioinformatics in <i>Español</i>: expanding open, diverse, and collaborative science through SEH2Bioinfo.","authors":"Carla L Padilla Franzotti, Gabriel J Olguín-Orellana, Nicolas Palopoli, Sebastián Urquiza-Zurich, Adrian Garcia-Moreno, Sara Monzón, Mónica Cabrera-Pasadas, Irene Soler Sáez, Carla Perpiñá-Clérigues, Rafael J Puche Quiñonez, Jennifer Vélez Segura, Jose Abel Lovaco-Flores, Cleidy M Osorio Mogollón, Eva M Esquinas-Román, María C Miserendino, Pradeep Eranti, Ana Castillo-Orozco, Yesid Cuesta-Astroz, R Gonzalo Parra","doi":"10.1093/bioadv/vbag182","DOIUrl":"10.1093/bioadv/vbag182","url":null,"abstract":"<p><p>Linguistic inequality remains a silent barrier in contemporary science, particularly in rapidly evolving and interdisciplinary fields such as bioinformatics. This article examines how the Spanish-speaking community has responded to this gap through the creation of the Simposio de Estudiantes Hispanohablantes de Bioinformática y Biología Computacional (SEH2Bioinfo), a collective initiative that transforms language into a tool for inclusion and collaboration. Drawing on a historical review of the use of Spanish in scientific communication and an analysis of the two editions of SEH2Bioinfo (2022 and 2025), complemented by surveys of more than 1500 participants, we explore the effects of language on confidence, participation, and academic visibility. The findings show that Spanish can serve as a space for training and mentorship without compromising rigor, while simultaneously strengthening international networks and promoting open science. The growth of SEH2Bioinfo, its collaborative approach, and its intercontinental reach establish the symposium as a model for linguistic equity in scientific practice. Its experience demonstrates that expanding linguistic diversity in bioinformatics is not only possible but essential to building a global scientific community that is fairer, more connected, and representative.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag182"},"PeriodicalIF":2.6,"publicationDate":"2026-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13455081/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148708639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ICU mortality prediction via multimodal learning on MIMIC-IV clinical data.","authors":"Alberto Falcone, Agostino Forestiero, Rocco Malivindi, Marzia Settino","doi":"10.1093/bioadv/vbag214","DOIUrl":"https://doi.org/10.1093/bioadv/vbag214","url":null,"abstract":"<p><strong>Purpose: </strong>This study proposes a multimodal framework for retrospective short-term mortality prediction in intensive care unit patients by integrating structured admission data with semantic information extracted from discharge summaries. The objective is to improve patient characterization by combining complementary clinical data that are often analyzed separately.</p><p><strong>Methods: </strong>Structured inputs, including demographics, administrative records, laboratory results, and clinical severity scores, were merged with semantic representations of clinical notes produced by a BERT-based encoder. The resulting embeddings were reduced in dimensionality and clustered with K-Means to obtain categorical semantic features reflecting thematic similarity among notes. These semantic cluster identifiers were then concatenated with tabular variables to create a unified multimodal representation, which was used to train five classical machine learning classifiers.</p><p><strong>Results: </strong>The integration of semantic features consistently improved predictive performance compared with models trained only on structured data. Among the evaluated algorithms, Random Forest achieved the best discrimination, with an area under the receiver operating characteristic curve of 0.940 in the multimodal setting.</p><p><strong>Conclusion: </strong>Combining structured clinical variables with unstructured admission notes enhances short-term mortality prediction and supports the development of interpretable multimodal decision-support tools for intensive care settings.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag214"},"PeriodicalIF":2.6,"publicationDate":"2026-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13600438/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-omic spectral clustering with the flag mean.","authors":"Charlie M Carpenter, Ziwei Tian, Jonathan K Harris, Michael Kirby, Traci Lyons, Christopher Peterson, Brandie D Wagner, Debashis Ghosh","doi":"10.1093/bioadv/vbag217","DOIUrl":"https://doi.org/10.1093/bioadv/vbag217","url":null,"abstract":"<p><strong>Motivation: </strong>One common goal in multi-omics studies is to identify subgroups within the study's cohort. Many methods create subgroups through unsupervised clustering, and, to our knowledge, all of these methods attempt to infer a common subspace or latent cluster across multiple views. We argue that this is a strict assumption that may not truly exist in many datasets.</p><p><strong>Results: </strong>We use the classic spectral clustering algorithm in conjunction with the flag manifold. Together, this allows for differing cluster structures across the omics profiles, leading to a novel approach for more flexible subtyping in multi-omics studies. We study a dataset on ventilator associated pneumonia in children. These data contain airway microbiome and transcriptome. Through simulation studies, we demonstrate that the flag mean of separate clustering subspaces can accurately capture the span of a joint clustering space. It is also robust to varying noise structures and number of features across omics profiles. Our proposed method also outperforms popular multi-omics clustering methods in the presence of differing group sizes. This proposed method is general enough to apply to other multi-omics studies as well as any multi-view study that uses spectral clustering.</p><p><strong>Availability and implementation: </strong>Code for these methods are written in R and are freely available through GitHub at https://github.com/Ghoshlab/MMOC or through CRAN at https://cran.r-project.org/web/packages/MMOC/index.html.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag217"},"PeriodicalIF":2.6,"publicationDate":"2026-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13692795/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149277793","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Refining sequence-to-activity models by increasing model resolution.","authors":"Nuria Alina Chandra, Yan Hu, Jason D Buenrostro, Sara Mostafavi, Alexander Sasse","doi":"10.1093/bioadv/vbag122","DOIUrl":"10.1093/bioadv/vbag122","url":null,"abstract":"<p><p>Decoding the cis-regulatory syntax that controls gene expression is essential for improving our understanding of cell differentiation and disease. Deep learning based sequence-to-activity (S2A) models learn to identify regulatory motifs and their syntax through modeling chromatin accessibility. Previously, we developed AI-TAC, a S2A model that predicts chromatin accessibility across various immune cell types in multi-task fashion, effectively decoding the regulatory syntax underlying immune cell differentiation. While ATAC-seq is commonly used to measure regional accessibility, it also provides high-resolution profiles, the distribution of Tn5 insertion sites, that offer additional insights into the precise location and strength of TF binding sites. Here we present bpAI-TAC, a base-pair resolution ATAC-seq model, and demonstrate that modeling ATAC-seq profiles alongside accessibility consistently improves predictions of differential chromatin accessibility across cell types. Moreover, we find that multi-task learning across related immune cell types consistently outperforms single-task models. To understand what additional information bpAI-TAC learns from ATAC-seq profiles, we systematically compare sequence attributions from models trained with and without ATAC-seq profiles. We identify novel motifs with strong effect sizes that emerge when profile data is included. Our findings suggest that modeling ATAC-seq at base-pair resolution enables the model to learn a more nuanced and sensitive representation of the cis-regulatory syntax driving immune cell-specific chromatin landscapes.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag122"},"PeriodicalIF":2.6,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13427759/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148655023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DiDyNet: a robust framework for differential dynamic network inference from longitudinal multi-omics data.","authors":"Zhe Liu, Kesong Wu, Taesung Park","doi":"10.1093/bioadv/vbag192","DOIUrl":"https://doi.org/10.1093/bioadv/vbag192","url":null,"abstract":"<p><strong>Motivation: </strong>Understanding disease dynamics from longitudinal multi-omics is hindered by traditional approaches that focus on univariate trajectories and static networks while ignoring temporal evolution. We developed DiDyNet, a framework for identifying phenotype-specific temporal molecular networks by defining dynamic coupling as coordinated molecular trajectories. DiDyNet operates through four steps: (i) two-dimensional variance-based filtering to prioritize dynamic features; (ii) quantification of subject-specific coordination using Dynamic Time Warping to accommodate asynchrony; (iii) statistical testing for differential dynamic couplings; and (iv) linear mixed model-based post-hoc refinement to distinguish genuine coordinated dynamics from stochastic noise.</p><p><strong>Results: </strong>Simulation studies showed that DiDyNet significantly outperformed static summary statistics, including the mean, median, and difference, which cannot capture dynamic signals. Dynamic Time Warping-based quantification also demonstrated greater robustness than Euclidean distance, correlation-based distance, and constrained alignment methods under temporal misalignment and signal sparsity. Application to an insulin resistance cohort identified a coordinated cross-omics network linking systemic inflammation with intracellular stress responses.</p><p><strong>Availability: </strong>Source code is freely available at https://github.com/bioinfoliu/DiDyNet.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag192"},"PeriodicalIF":2.6,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13493046/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MLHeatmap: an interactive application for transcriptomic marker-panel discovery.","authors":"Eun Young Lee, Jihye Park, Seoyeon Youn, Kyungho Choi, Sungryul Yu, Keunsoo Kang","doi":"10.1093/bioadv/vbag213","DOIUrl":"https://doi.org/10.1093/bioadv/vbag213","url":null,"abstract":"<p><strong>Motivation: </strong>Transcriptomic biomarker studies often require separate tools for normalization, classification, feature ranking, and visualization. MLHeatmap was developed to integrate these steps into a cross-platform, browser-based workflow and to support compact marker-panel discovery directly from count matrices.</p><p><strong>Results: </strong>MLHeatmap accepts a count matrix and sample-group labels and performs gene mapping, normalization, multiclass classification with nested cross-validation, feature attribution, differential expression analysis, and interactive heatmap visualization, with support for multiple classifiers and panel-selection methods in the same workflow. As a case study, we applied MLHeatmap to colorectal cancer consensus molecular subtype (CMS) classification using 511 primary tumors from The Cancer Genome Atlas (TCGA) with published Colorectal Cancer Subtyping Consortium (CRCSC) labels. Random Forest with forward selection achieved 89.8% out-of-fold accuracy and a macro-averaged area under the receiver operating characteristic curve (macro AUC) of 0.973, yielding a 13-gene compact panel with a held-out AUC of 0.960. External validation of the compact panel gave 74.8% accuracy and a macro AUC of 0.917 in GSE39582, an independent Affymetrix GPL570 microarray cohort evaluated in 500 tumors with confident CMScaller-derived CMS assignments, and 69.18% accuracy and a macro AUC of 0.909 in CMCBSN, an independent Korean RNA-seq cohort with 159 confidently labeled tumors. Across the five classifiers and four panel-selection methods evaluated, compact-panel AUCs ranged from 0.895 to 0.970, with tree-ensemble classifiers reaching the highest values.</p><p><strong>Availability and implementation: </strong>MLHeatmap is freely available at https://github.com/kangk1204/MLHeatmap and can be installed and run locally on Windows 11, macOS, and Ubuntu.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag213"},"PeriodicalIF":2.6,"publicationDate":"2026-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13472729/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148765678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"visPedigree: a comprehensive R package for tidying, analyzing, and visualizing breeding pedigrees.","authors":"Sheng Luan, Jie Kong, Zhenglong Xia, Ziyi Kang, Guangfeng Qiang, Kun Luo, Juan Sui","doi":"10.1093/bioadv/vbag210","DOIUrl":"https://doi.org/10.1093/bioadv/vbag210","url":null,"abstract":"<p><strong>Motivation: </strong>Pedigrees support relatedness control, mate allocation, inbreeding monitoring, and diversity assessment, but large breeding populations are often curated, analyzed, and visualized using separate tools and data structures. This fragmentation complicates routine analysis and reporting, particularly in high-fecundity systems with extensive full-sib structure, irregular pedigree depth, and repeated cohort-based evaluation.</p><p><strong>Results: </strong>We present visPedigree, a software package for large breeding pedigrees built around a new tidyped object model. It integrates pedigree standardization, structural validation, candidate-centered tracing, pedigree statistics, inbreeding and partial inbreeding analysis, founder- and ancestor-based diversity summaries, effective population size estimation, relationship matrices, and scalable visualization. The package also provides Shannon/Hill-based diversity measures, the pedhalflife() temporal diagnostic, and compact visualization of full-sib-dominated subsets. In a giant freshwater prawn pedigree, it produced an interpretable ancestry overview and quantified recent diversity erosion. Simulated benchmarks showed that representative workflows completed within seconds for pedigrees containing up to one million individuals on a standard laptop.</p><p><strong>Availability and implementation: </strong>visPedigree is implemented in the R language and is freely available at https://cran.r-project.org/package=visPedigree, with documentation at https://luansheng.github.io/visPedigree/.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag210"},"PeriodicalIF":2.6,"publicationDate":"2026-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13479741/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A leakage-aware genomic prediction pipeline for meropenem resistance in <i>Klebsiella pneumoniae</i> using transformer-based resistome representation learning.","authors":"İlkay Sibel Kervancı","doi":"10.1093/bioadv/vbag193","DOIUrl":"10.1093/bioadv/vbag193","url":null,"abstract":"<p><strong>Motivation: </strong>Antimicrobial resistance (AMR) in <i>Klebsiella pneumoniae</i>, particularly to carbapenems such as meropenem, is a major global health problem. Machine learning is increasingly used to predict resistance from genomic markers; however, many models fail to capture high-level gene-gene interactions and may exhibit inflated performance due to lineage-biased prediction. Existing genomic prediction models largely rely on flat feature representations that fail to capture epistatic gene interactions, and commonly suffer from inflated performance estimates due to phylogenetic data leakage. To address these limitations simultaneously, a leakage-aware hybrid TabTransformer-CatBoost pipeline was developed, combining self-attention-based resistome representation learning with gradient boosting classification under clade-aware data partitioning. A self-attention encoder converts sparse gene presence-absence profiles into contextualized latent embeddings, which are subsequently classified using gradient boosting to capture lineage-aware AMR patterns.</p><p><strong>Results: </strong>The proposed architecture outperformed classical baselines including Logistic Regression, Random Forest, XGBoost, and optimized CatBoost models. Internal accuracy reached 92.59% for the Chained Hybrid configuration (area under the receiver operating characteristic curve, AUROC = 0.8670, F1 = 0.8537). Performance gains primarily originated from the embedding stage, as confirmed by ablation analysis. External validation across independent multinational cohorts (<i>n</i> = 305) demonstrated generalizability (AUROC = 0.8105; F1 = 0.7552). Permutation testing produced near-zero Matthews Correlation Coefficient (MCC) = 0.0091, indicating predictions reflect genuine biological signal rather than noise. These results establish attention-based genomic embedding with gradient boosting as a scalable, interpretable, and leakage-aware framework for clinical AMR prediction.</p><p><strong>Availability and implementation: </strong>The source code for the TabTransformer-CatBoost framework, including preprocessing pipelines and pre-trained embeddings, is available at https://github.com/SibelKervanci/kp-meropenem-tabtransformer.</p>","PeriodicalId":72368,"journal":{"name":"Bioinformatics advances","volume":"6 1","pages":"vbag193"},"PeriodicalIF":2.6,"publicationDate":"2026-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13532757/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148876635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}