SEARCH · PubMed Health

Results for “Machine learning”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6Linked to original sources

PubMed2025

Stochastic epigenetic mutation profiles as biomarkers of clinical activity in juvenile idiopathic arthritis: a multi-omic machine learning approach for gene prioritization.

BACKGROUND: Juvenile idiopathic arthritis (JIA) is a rare autoimmune disease arising from a complex interplay between genetic and environmental factors. Epigenetic modifications such as DNA methylation (DNAm) have been described as potential mediators in gene-environment interactions, contributing to immune system dysregulation. Emerging evidence suggests that DNAm profiles also predict therapeutic responses in autoimmune diseases. This study aims to identify epigenetic biomarkers and epigenetic-driven gene expression changes associated with JIA clinical activity. METHODS: We reanalyzed a publicly available dataset of 44 JIA patients, with whole-genome DNAm and gene expression from CD4 + T cells measured at two points: at anti-TNF therapy withdrawal (T0) and eight months later (Tend). At Tend, 30 patients maintained inactive disease (ID) while 14 did not (NO ID). We investigated differences between ID and NO ID patients in the epigenetic mutation load and various epigenetic clocks through linear regression models, and prioritized genomic regions with significantly higher number of epimutations in NO ID patients through machine learning. RESULTS: We found a higher mutation load in NO ID than ID patients, both at T0 and at Tend, with the differences at Tend reaching statistical significance (p = 0.02). In contrast, we found no evidence of association between epigenetic clocks and JIA clinical activity. Using a multi-omic approach, we identified a List of candidate epigenetically-driven differentially expressed genes, 80 up-regulated and 77 down-regulated, in NO ID patients. Finally, comparing our candidate gene list with the Connectivity Map database, we identified new candidate potential therapeutic targets. Key findings were validated in independent datasets: DNAm profiles from CD4 + T cells (56 JIA patients, 57 controls) and transcriptomic data from PBMCs of JIA patients with active or inactive disease, confirming dysregulation of pathways such as TNF-α signaling via NF-kB and TGF-β signaling among others. CONCLUSIONS: We described a significant association of epigenetic mutations with JIA clinical activity, indicating that epigenetic changes might precede clinical symptoms and may serve as biomarkers for early disease monitoring. Further, our results shed light on biomolecular mechanisms of JIA, supporting the development of more effective treatments.

Humans↗

PubMed2025

Identification of potential biomarkers and mechanisms for keloid disorder based on comprehensive bioinformatics analysis and machine learning algorithms.

BACKGROUND: Keloid disorder (KD) encompasses a spectrum of fibroproliferative dermal conditions, the pathogenesis remains complex and incompletely understood. This study sought to identify biomarkers and potential therapeutic targets for KD through an integrative bioinformatics approach and machine learning analysis of RNA sequencing data. METHODS: RNA sequencing was performed on skin tissue samples from 13 patients with KD and 14 healthy controls. Using weighted gene co-expression network analysis and differential expression analysis revealed differentially expressed key module genes, and the CytoHubba plugin identified candidate genes. Subsequently analyzed using least absolute shrinkage and selection operator (LASSO) and support vector machine recursive feature elimination (SVM-RFE) methods to pinpoint feature genes associated with KD. Following this, biomarkers were determined through expression level validation, enrichment analysis, and immune infiltration analysis. RESULTS: A total of 420 differentially expressed key module genes were identified, and the top 10 genes with DMNC values were selected as candidate genes. Five feature genes were selected through LASSO and SVM-RFE, with NID2, MFAP2, COL8A1, and P4HA3 showing significant expression differences between KD and control samples, along with consistent expression patterns across datasets, identified as potential biomarkers. These four biomarkers were proved to possess high diagnostic potential, and they were found to exhibit significant positive correlations with one another. Functional enrichment analysis indicated that the primary KEGG pathways associated with these biomarkers included "steroid hormone biosynthesis" and "cytokine-cytokine receptor interaction." Moreover, immune infiltration analysis revealed that the four biomarkers were negatively correlated with type 17 T helper cells and positively correlated with 15 immune cell types, including activated B cells and central memory CD4 T cells. CONCLUSION: In conclusion, NID2, MFAP2, COL8A1, and P4HA3 were identified as key biomarkers for KD, offering new avenues for more targeted and effective diagnostic and therapeutic strategies for managing this condition.

Humans↗

PubMed2026

Systematic review of machine learning approaches for predicting sickle cell crisis and mortality risk at the climate-health nexus.

BACKGROUND: Sickle cell anemia (SCA) is a severe genetic blood disorder characterized by recurrent vaso-occlusive crises and increased mortality, with the greatest burden occurring in low- and middle-income countries. Climatic and environmental conditions, including temperature variability, humidity, rainfall, air pollution, and seasonal changes, have been associated with disease exacerbation. However, the extent to which these factors have been incorporated into predictive models remains unclear. This study systematically reviews the application of machine learning (ML) models for predicting SCA crises and mortality in relation to climate and environmental factors. METHODOLOGY: The PRISMA guidelines were used, and 34 peer-reviewed studies published between 2005 and 2026 were analyzed to identify the climate variables, ML approaches employed, and predictive performance. The reviewed studies applied a range of ML techniques, including artificial neural networks, random forests, support vector machines, decision trees, logistic regression, and deep learning models. Temperature, humidity, rainfall, wind speed, air quality indicators, and seasonal patterns were the most frequently examined environmental variables. RESULTS: The findings indicate that most existing models rely predominantly on clinical and demographic data, with limited integration of climate information and inadequate representation of high-burden regions, especially Sub-Saharan Africa. Studies incorporating environmental variables reported improved predictive performance and highlighted the potential of climate-informed early warning systems for SCA management. CONCLUSION: The review recommends development of interdisciplinary, climate-aware ML frameworks, expansion of longitudinal environmental datasets, and increased research in underrepresented regions to support climate-resilient and patient-centered SCA care.

Humans↗

PubMed2026

Development and external validation of an explainable machine learning model for predicting chronic kidney disease progression in the Korean population.

BACKGROUND: Current risk stratification models, such as the Kidney Failure Risk Equation (KFRE), exhibit variable performance across ethnic groups and fail to capture dynamic clinical trajectories. This study aimed to develop and validate a Korean-specific machine learning (ML) model for predicting chronic kidney disease (CKD) progression using an ensemble approach. METHODS: We used electronic health records from Seoul National University Hospital for model development (n = 28,209) and the Korean Genome and Epidemiology Study (KoGES) CKD cohort for external validation (n = 3,960). The primary outcome was a composite of ≥40% decline in estimated glomerular filtration rate (eGFR) or progression to end-stage renal disease within 2 years. A soft-voting ensemble of four ML algorithms (XGBoost, LightGBM, CatBoost, and Random Forest) was developed. RESULTS: The ensemble model demonstrated robust discrimination in internal validation (area under the receiver operating characteristic curve [AUROC], 0.939; 95% confidence interval [CI], 0.934-0.944), significantly exceeding the KFRE (AUROC, 0.879-0.884). External validation in the KoGES cohort showed comparable discrimination (AUROC, 0.859; 95% CI, 0.798-0.914) versus KFRE (four-variable AUROC, 0.882; 95% CI, 0.818-0.935). Shapley Additive exPlanations (SHAP) analysis identified baseline eGFR, serum creatinine, eGFR slope, albumin, and hemoglobin as key prognostic features, supporting a complementary framework using KFRE for community screening and the ML model for hospital-based risk stratification. CONCLUSION: The ensemble ML model accurately predicts short-term CKD progression in Korean patients. By incorporating longitudinal features and ensemble learning, it provides a precise alternative to Western-derived equations, particularly in tertiary care settings.

Chronic kidney failure↗

PubMed2026

Pathomics-based machine learning models for predicting METTL5 expression and prognosis in lung adenocarcinoma.

BACKGROUND: METTL5, an N6-methyladenosine (m6A) RNA methyltransferase, has been implicated in tumor progression, but its prognostic value and non-invasive prediction in lung adenocarcinoma (LUAD) remain unclear. This study aimed to develop a pathomics-based machine learning model to predict METTL5 expression from histopathological images and evaluate its prognostic significance in LUAD. METHODS: A total of 327 LUAD patients from The Cancer Genome Atlas (TCGA) with matched hematoxylin and eosin (H&E) slides, transcriptomic, and clinical data were included and randomly divided into training and validation sets (7:3). Quantitative histopathological features were extracted using PyRadiomics. Feature selection was performed via maximum relevance minimum redundancy (mRMR) and recursive feature elimination (RFE), followed by construction of a Gradient Boosting Machine (GBM) model. A pathomics score (PS) was generated to assess prognostic relevance. Survival analyses, gene set variation analysis (GSVA), tumor mutational burden (TMB), immune infiltration analysis, and in vitro functional assays were conducted. RESULTS: METTL5 overexpression was independently associated with poor overall survival [hazard ratio (HR) =1.637, P=0.007]. The model achieved good predictive performance [area under the curve (AUC) =0.847 in the training set and 0.752 in the validation set]. High PS was significantly associated with worse survival and remained an independent prognostic factor (HR =1.563, P=0.03). Elevated PS correlated with altered metabolic pathways, increased TMB, and immune microenvironment changes. METTL5 knockdown reduced proliferation, migration, invasion, and epithelial-mesenchymal transition (EMT) in A549 cells. CONCLUSIONS: The pathomics-based model accurately predicts METTL5 expression and provides prognostic stratification in LUAD, supporting its potential as a practical imaging-derived biomarker.

Methyltransferase-like 5↗

PubMed2025

Design of nanobody targeting SARS-CoV-2 spike glycoprotein using CDR-grafting assisted by molecular simulation and machine learning.

The design of proteins capable effectively binding to specific protein targets is crucial for developing therapies, diagnostics, and vaccine candidates for viral infections. Here, we introduce a complementarity-determining region (CDR) grafting approach for designing nanobodies (Nbs) that target specific epitopes, with the aid of computer simulation and machine learning. As a proof-of-concept, we designed, evaluated, and characterized a high-affinity Nb against the spike protein of SARS-CoV-2, the causative agent of the COVID-19 pandemic. The designed Nb, referred to as Nb Ab.2, was synthesized and displayed high-affinity for both the purified receptor-binding domain protein and to the virus-like particle, demonstrating affinities of 9 nM and 60 nM, respectively, as measured with microscale thermophoresis. Circular dichroism showed the designed protein's structural integrity and its proper folding, whereas molecular dynamics simulations provided insights into the internal dynamics of Nb Ab.2. This study shows that our computational pipeline can be used to efficiently design high-affinity Nbs with diagnostic and prophylactic potential, which can be tailored to tackle different viral targets.

Spike Glycoprotein, Coronavirus↗

PubMed2026

Machine learning identifies ac4C-related prognostic signature and TUBA1C as therapeutic target in COAD.

To explore the role of N4-acetylcytidine (ac4C)-related genes (acRGs) in colon adenocarcinoma (COAD) and identify reliable prognostic biomarkers and potential therapeutic targets. Multi-source transcriptomic datasets (TCGA-COAD, GSE39582, GSE17536) and single-cell RNA-seq data were analyzed. Ten machine learning algorithms were integrated to construct an acRG-based prognostic signature (acRGBS). Immune microenvironment (TME) and genomic profiling were performed, with in vitro functional experiments validating TUBA1C's role. acRGBS, comprising four hub genes (SARAF, CDC42SE2, TSPYL2, TUBA1C), effectively stratified COAD patients into high- and low-risk groups with distinct survival outcomes and was an independent prognostic factor. High-risk patients exhibited increased genomic instability and immunosuppressive TME, while low-risk patients had favorable immunotherapy response. TUBA1C was overexpressed in COAD cells, and its knockdown inhibited proliferation/migration and induced apoptosis. The acRGBS is a robust prognostic tool for COAD, and TUBA1C serves as a candidate therapeutic target, providing new insights for personalized COAD management.

Humans↗

PubMed2026

DBP-CanPred: a machine learning model for predicting cancer-causing mutations in DNA-binding proteins.

INTRODUCTION: The fundamental cellular processes, including transcriptional regulation, chromatin organization, and genome maintenance, are regulated by DNA-binding proteins (DBPs). Mutations in DBPs can alter protein-DNA interactions, leading to tumor development. However, identifying such driver mutations remains a major challenge due to limitations of experimental approaches. METHODS: We have trained a machine learning model, DBP-CanPred, to identify driver mutations in DBPs. We used the sequence-derived evolutionary features, as well as structure-based features such as mutation-perturbed structural descriptors. RESULTS: We evaluated DBP-CanPred using a curated test set, achieving an AU-ROC of 0.86 and a balanced accuracy of 0.79. Further analysis based on substitution-type showed consistent performance across different categories, especially higher performance on charged residues. In addition, we applied the model on an independent dataset and identified potential driver mutations with high confidence scores. DISCUSSION: The study contributes to understanding mutation patterns in DNA-binding proteins and supports variant interpretation in cancer research.

DNA-binding proteins↗

PubMed1993

Knowledge-based generation of machine learning experiments: learning with DNA crystallography data.

Though it has been possible in the past to learn to predict DNA hydration patterns from crystallographic data, there is ambiguity in the choice of training data (both in terms of the relevant set of cases and the features needed to represent them), which limits the usefulness of standard learning techniques. Thus, we have developed a knowledge-based system to generate machine learning experiments for inducing DNA hydration pattern classifiers. The system takes as input (1) a set of classified training examples described by a large set of attributes and (2) information about a set of learning experiments that have already been run. It outputs a new learning experiment, namely a (not necessarily proper) subset of the input examples represented by a new set of features. Domain specific and domain independent knowledge is used to suggest subsets of training examples from suspected subpopulations, transform attributes in the training data or generate new ones, and choose interesting ways to substitute one experiment's set of attributes with another. Automatic hydration pattern predictors are of both theoretical and practical interest to DNA crystallographers, because they can speed up a labor intensive process, and because the extracted rules add to the knowledge of what determines DNA hydration.

Artificial Intelligence↗

PubMed2026

Aberrant mucin expression and keratinization distinguishing severe from mild asthma revealed by interpretable machine learning.

Type 2 (T2) immune cells dominate the airways of patients with mild-moderate asthma (MMA) with a more complex type 1 (T1)-T2 mixed immune response evident in treatment-refractory severe asthma (SA). We hypothesized that comparing the transcriptomes of the airway epithelium of patients with SA and MMA would reveal molecular signatures associated with more severe disease in the context of a complex immune response. Using our interpretable machine learning tool, SLIDE, meaningful latent factors (context-specific gene co-expression networks) were revealed that distinguished SA from MMA. Unexpectedly, an aberrant high expression of normally host-protective, membrane-tethered, and IFN-inducible mucins, MUC1 and MUC4, was identified in SA. Gene networks in the significant latent factors discriminating SA from MMA corresponded to enrichment of a keratinization program in SA airways. Keratinization was marked by increased expression of the stress keratin KRT16, signifying squamous metaplasia suggesting adaptive reprogramming of the airway epithelium in response to chronic stress. These mucins and KRT16 were inversely associated with lung function in 2 separate asthma cohorts. Imaging of endobronchial biopsies revealed significantly higher KRT16 protein expression in SA compared with MMA that strongly correlated with MUC1 protein expression. Our study identifies dysregulated host-protective and maladaptive repair responses in SA distinguishing from MMA.

Humans↗

PubMed2026

Development and Validation of Machine Learning Models for Predicting Early Cognitive Decline Using Home Sensor-Derived Behavioral Data: Sensors in-Home for Elder Wellbeing (SINEW) Cohort Study.

BACKGROUND: As the global population continues to age, the prevalence of geriatric conditions, including dementia and frailty, is also increasing. Early identification of individuals at an elevated risk of these conditions, such as those presenting with mild cognitive impairment (MCI) or prefrailty, can provide a critical window for prompt intervention aimed at preventing or reversing disease progression. To promote such early identification, there is a burgeoning interest in the use of digital sensor technology and predictive modeling. OBJECTIVE: This study aimed to use a continuous, home-based monitoring sensor system for older adults to distinguish those exhibiting normal aging from those with MCI, early dementia, prefrailty, or frailty, and to predict their transition from normal aging to one of these conditions. METHODS: This longitudinal cohort study will recruit 200 community-dwelling adults aged ≥65 years with normal cognition or MCI at baseline. A multi-sensor system will be installed in participants' homes, including passive infrared motion sensors, door contact sensors, bed sensors, medication box sensors, wearable activity bands, and Bluetooth proximity beacons. These devices will continuously capture spatiotemporal activity patterns, mobility indicators, sleep behaviors, and medication-taking routines. Annual assessments will include standardized cognitive tests (eg, Montreal Cognitive Assessment, Mini-Mental State Examination, Rey Auditory-Verbal Learning Test, digit span, Color Trails Test, semantic fluency, Stroop), frailty measures (modified Fried phenotype, gait speed, grip strength), mental health scales, sleep quality, and psychosocial indicators. Sensor-derived features-such as gait variability, activity regularity, sleep fragmentation, and medication adherence patterns-will be integrated with clinical data to develop supervised machine learning models. Planned approaches include logistic regression, random forests, gradient boosting, and deep learning. Model performance will be evaluated using cross-validation and independent test sets. Primary metrics will include area under the receiver operating characteristic curve, sensitivity, specificity, precision, recall, and F1-score. Models will be benchmarked against gold-standard clinical diagnoses and validated using temporal subsets of the dataset. RESULTS: Enrollment for this study started in November 2019 and will continue until March 2030. As of June 2025, we have enrolled 138 participants. Full data analysis has yet to begin. CONCLUSIONS: We aim to develop a reliable and effective sensor system for in-home use that will facilitate the early detection of cognitive and physical decline. In so doing, it will add to our current understanding of digital biomarkers. It is common for older adults to seek clinical intervention only when their cognitive impairment has already reached an advanced stage. The implementation of readily deployable sensor systems within community settings presents us with opportunities for prompt intervention, which holds the potential for delaying or reversing disease progression and allowing for a greater number of functional and meaningful years.

Humans↗

PubMed2026

MegaPlantTF: a machine learning framework for comprehensive identification and classification of plant transcription factors.

MOTIVATION: Understanding the role of transcription factors (TFs) in plants is essential for the study of gene regulation and various biological processes. However, both TF detection and classification remain challenging due to the great diversity and complexity of these proteins. Conventional approaches, such as BLAST, often suffer from high computational complexity and limited performance on less common TF families. RESULTS: We introduce MegaPlantTF, the first comprehensive machine learning and deep learning framework for the prediction (TF versus non-TF) and classification (family-level) of plant TFs. Our method employs k-mer-based protein representations and a two-stage architecture combining a deep feed-forward neural network with a stacking ensemble classifier. To ensure robust performance assessment, we report micro-, macro-, and weighted-average performance metrics, providing a holistic evaluation of both frequent and underrepresented TF families. Additionally, we employ threshold-based evaluation to calibrate confidence in TF detection. The results show that MegaPlantTF achieves strong accuracy and precision, particularly with a k-mer size of 3 and a classification threshold of 0.5, and maintains stable performance even under stringent thresholds. In addition to the standard cross-validation tests, a use case study on Sorghum bicolor confirms that our method performs strongly in the genome-wide analysis, making it highly suitable for large-scale TF identification and classification tasks. MegaPlantTF represents a novel contribution by integrating k-mer encoding, binary family-specific classifiers, and a two-stage stacking ensemble into a unified, reproducible framework for large-scale plant TF identification and classification. AVAILABILITY AND IMPLEMENTATION: MegaPlantTF is freely accessible through a public web server available at https://bioinformatics.um6p.ma/MegaPlantTF. The complete source code, including pretrained models and example datasets, is available at https://github.com/Bioinformatics-UM6P/MegaPlantTF.

Transcription Factors↗

PubMed2026

seq2ribo: structure-aware integration of machine learning and simulation to predict ribosome location profiles from RNA sequences.

MOTIVATION: Ribosome dynamics are vital in the process of protein expression. Current methods rely on ribosome profiling (Ribo-seq), RNA-seq profiles, and full genomic context. This restricts their use in de novo sequence design, like messenger RNA (mRNA) vaccines. Simulation-only approaches like the Totally Asymmetric Simple Exclusion Process (TASEP) oversimplify translation by focusing solely on codon elongation times. RESULTS: We present seq2ribo, a hybrid simulation and machine learning framework that predicts ribosome A-site locations using only an mRNA sequence as input. Our method first employs a novel structure-aware TASEP (sTASEP), which models translation using a comprehensive set of fitted parameters that include codon wait times and structural features, such as local angles, base-pairing, and discrete positional buckets. The ribosome locations generated by sTASEP are then processed by a polisher model, which learns to refine the simulated ribosome distributions. seq2ribo provides high-fidelity predictions of ribosome locations across diverse cell types (iPSC, HEK293, LCL, and RPE-1), significantly outperforming baselines. seq2ribo is the first method to achieve meaningful positional correlation with observed ribosome profiles from sequence alone, reaching transcript-level Pearson correlations up to 0.920 and within-transcript shape correlations up to 0.186, where all baselines yield near-zero values on these metrics. seq2ribo also reduces elementwise error by up to 37.7% relative to the sequence-only Translatomer baseline. By adding a task-specific head, seq2ribo achieves Pearson correlations up to 0.732 with experimental translation efficiency (TE) across several cell lines, and up to 0.903 with measured protein expression. By operating from sequence alone, seq2ribo provides a new tool for synthetic biology, enabling the rational design and optimization of mRNA sequences without the need for expression-level data or genomic context. AVAILABILITY: seq2ribo is available at https://github.com/Kingsford-Group/seq2ribo.

Machine Learning↗

PubMed2025

Inflammatory pathways and immune dysregulation in pediatric postoperative septic shock: A study integrating transcriptomics, machine learning and molecular docking.

This study elucidates the molecular and immune regulatory mechanisms of pediatric postoperative septic shock. Transcriptomic data were obtained from the Gene Expression Omnibus database. Differentially expressed genes were identified using the limma package, and gene co-expression modules were constructed using Weighted Gene Co-expression Network Analysis. Functional enrichment was performed via gene set enrichment analysis, Gene Ontology, and Kyoto Encyclopedia of Genes and Genomes analyses. Immune cell infiltration was assessed using ESTIMATE and CIBERSORT. Mendelian randomization was applied to explore causal relationships between gene expression and septic shock. Feature genes were selected using machine learning algorithms, and a diagnostic nomogram model was constructed. Finally, molecular docking analysis was performed to screen and evaluate the binding affinity of traditional Chinese medicine monomers to core target proteins. A total of 1331 differentially expressed genes were identified, and the turquoise module was strongly correlated with septic shock. Enrichment analysis revealed significant activation of IL-6/JAK/STAT3, TNF-α/NF-κB, and PI3K/Akt/mTOR pathways. Immune infiltration analysis indicated suppressed immune scores and imbalances in neutrophils, macrophages, T cells, and B cells. Mendelian randomization confirmed causal associations for 6 genes, including PIM3. The predictive model based on feature genes demonstrated high diagnostic performance. Molecular docking suggested that quercetin and astramembrannin I could stably bind PIM3. This study systematically identified core genes, dysregulated immune pathways, and candidate small-molecule interventions in pediatric septic shock, providing novel insights for early diagnosis and targeted therapy.

Humans↗

PubMed2026

Seeing and Feeling DNA Methylation: Single-Molecule Biophysics Meets Machine Learning.

DNA methylation at 5-methylcytosine (5mC) is crucial for embryonic development and cellular function, while aberrant patterns strongly drive disease onset and progression. Its reversible nature offers substantial therapeutic potential, emphasizing the need for precise, context-specific genome wide 5mC mapping. Conventional techniques such as bisulfite sequencing and ensemble biosensor assays are hindered by DNA degradation, amplification bias, high cost, and inability to resolve single-molecule structural and mechanical effects of methylation. This review examines advances in single-molecule biophysical methods (nanopore sensing, smFRET, optical/magnetic tweezers, and AFM) that provide direct, label-free/minimally invasive 5mC detection, along with quantitative insights into DNA conformation, mechanics, and protein-DNA interactions. These techniques complement traditional methylome mapping by linking genomic localization to molecular mechanisms. Emerging machine-learning approaches are revolutionizing analysis, particularly in nanopore sensing, while promising applications in smFRET, tweezers, and AFM address throughput and reproducibility challenges. Their convergence promises scalable, high-resolution epigenetic profiling, advancing precision epigenomics toward clinical application.

DNA Methylation↗

PubMed2026

Exploring an Intermediate Colorectal Cancer Screening Test Based on Stool Proteomics and Machine Learning for Optimizing the Selection of Patients for Colonoscopy Identified From FIT.

The fecal immunochemical test (FIT) for detecting fecal occult blood, used alone or in combination with other stool biomarkers, has been demonstrated to be effective in the context of colorectal cancer (CRC) screening programs. However, FIT yields a significant proportion of false positives leading to unnecessary colonoscopies. In this study, we have investigated whether leftover FIT stool samples could be repurposed for proteomics analysis as a triage step for patients before recommending colonoscopy. High-throughput mass spectrometry analyses on a set of 141 FIT-positive samples (50 controls with no lesion, 45 with advanced adenomas and 46 with CRC) in combination with machine learning tools were used. Results showed that with a specificity ≥90%, a large proportion of the false FIT positives could be identified thus providing an efficient strategy for reducing unnecessary colonoscopies. Furthermore, CRC cases were also precisely predicted to be true positives, thus providing an approach for prioritizing patients for colonoscopy. In conclusion, this study demonstrates the feasibility of using proteomics for analysis of leftover FIT stool samples as an intermediate step to triage patients selected for colonoscopy in CRC screening programs.

Humans↗

PubMed2026

Integrating machine learning and GWAS for variant prioritization in the INCIPE cohort highlights ABC transporter genes in chronic kidney disease.

INTRODUCTION: Chronic kidney disease (CKD) is a major public health challenge, affecting approximately 674 million people worldwide and representing one of the fastest-growing causes of mortality. Since CKD is frequently asymptomatic in its early stages, the identification of novel genetic biomarkers may improve early detection and risk stratification. Genome-Wide Association Studies (GWAS) have identified numerous genetic loci associated with CKD and related traits; however, their performance is often limited in small and imbalanced cohorts, where reduced statistical power increases both false-positive and false-negative findings. Machine learning (ML) approaches can complement conventional GWAS by prioritizing biologically relevant genetic signals from high-dimensional genomic data. METHODS: In this study, we implemented a nested ensemble (NCBC) model composed of an undersampler and a CatBoostClassifier (CBC) to prioritize candidate genetic variants associated with CKD in the INCIPE cohort. Prioritized variants were functionally annotated and evaluated through enrichment analyses, GTEx gene expression profiling, and protein-protein interaction network analyses. Genes identified by the CKDGen Consortium were analysed as an external reference set and used to validate the biological relevance of the prioritized results. RESULTS: The NCBC model outperformed conventional ML classifiers, achieving a ROC AUC score of 87.77%, compared to 50%-53% for the other evaluated models. Among the prioritized genes, 56.25% showed protein-protein interactions with genes previously reported by the CKDGen Consortium, whereas only 1.9% of randomly generated gene sets showed interactions. DISCUSSION: Our study demonstrates that the NCBC model improves the prioritization of biologically plausible candidate variants in a small and imbalanced CKD cohort. Functional analyses suggested ABC transporter-related genes, including ABCA13, ABCA4, and ABCC4 genes, as promising candidate for future validation, with ABCA4 showing substantial expression in kidney tissues. Overall, these findings support the integration of ML with GWAS to prioritize candidate genes and investigate the genetic architecture of complex diseases.

SNP prioritization↗

PubMed2025

Rapid assessment of clinical severity for salmonellosis cases via protein family domain analysis and machine learning.

Salmonella is a common pathogen, infecting more than a million people yearly. Rapid assessment of clinical case severity is essential for improving patient outcomes and optimizing healthcare resources. Advancements in genome sequencing technologies have enabled the analysis of bacterial genomes from many clinical cases, opening up new opportunities for precise and timely diagnosis. This study proposes a genome-based framework for identifying critical Salmonella cases before the onset of critical symptoms and facilitating early medical intervention. By leveraging protein family (Pfam) domains as the representation for genomic data, the complex genetic profiles of Salmonella cases are simplified into interpretable features. The severity levels of cases were investigated through rigorous data analysis, resulting in a set of 70 Pfam domains that could be potentially used as biomarkers. Machine Learning was employed to assess the predictive power of the curated Pfam biomarkers, achieving high accuracy (~93%) in sorting cases into critical, moderate, and mild categories. The results demonstrate the efficacy of the proposed approach. This framework highlights the potential of using bacterial genomic data in clinical decision-making, opening the window for timely personalized interventions for Salmonella infection management.

Domains of unknown function (DUFs)↗