Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “machine learning prediction model”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 91 records · Page 5Linked to original sources
PubMed1999

Case-based explanation of non-case-based learning methods.

We show how to generate case-based explanations for non-case-based learning methods such as artificial neural nets or decision trees. The method uses the trained model (e.g., the neural net or the decision tree) as a distance metric to determine which cases in the training set are most similar to the case that needs to be explained. This approach is well suited to medical domains, where it is important to understand predictions made by complex machine learning models, and where training and clinical practice makes users adept at case interpretation.

Artificial Intelligence↗
PubMed2026

Radiomics-based gradient boosting model on contrast-enhanced MRI for non-invasive prediction of epidermal growth factor receptor expression and therapeutic response to EGFR-targeted antibody-drug conjugates in high-grade glioma organoid models.

BACKGROUND: Epidermal growth factor (EGF) and its receptor EGF(EGFR) play crucial roles in glioblastoma (GBM) prognosis. However, non-invasive assessment of their expression remains challenging. This study aimed to determine whether radiomics features extracted from contrast-enhanced MRI could predict EGFR expression in high-grade gliomas (HGG) and to explore their associations with immune infiltration and therapeutic response of EGFR-Targeted antibody drug conjugates(EGFR-ADCs). METHODS: We extracted radiomic features from contrast-enhanced MRI of 298 GBM patients from The Cancer Imaging Archive (TCIA) and matched them with RNA-seq data from The Cancer Genome Atlas (TCGA). Feature selection was performed using minimum redundancy maximum relevance (mRMR) and recursive feature elimination (RFE). Machine learning models were built to predict EGF/EGFR expression. Radiogenomic associations were validated by immune infiltration analysis. Patient-Derived Tumor-Like Cell Clusters (PTC) were used to compare the antitumor efficacy of EGFR- ADCs and temozolomide. RESULTS: Elevated EGF/EGFR expression correlated with poor prognosis and increased infiltration of M2 macrophages, regulatory T cells, and CD4⁺ memory T cells. Pathway analysis demonstrated significant enrichment of the mechanistic target of rapamycin (mTOR) and Mitogen-Activated Protein Kinase (MAPK) signaling cascades. Radiomics-based prediction models achieved robust performance (AUC > 0.85) in stratifying EGFR expression status. In EGFR-positive tumor tissues, EGFR-ADCs exerted antitumor efficacy similar to that of temozolomide. CONCLUSIONS: EGF/EGFR expression is associated with immunosuppressive microenvironments and adverse outcomes in HGG. Radiomics may provide a non-invasive approach for estimating EGFR expression, although model performance requires external validation and EGFR-ADCs showed partial inhibitory activity within the tested range, though potency remains to be defined.These findings suggest a framework into radiogenomic stratification and targeted therapy in GBM.

Radiomics↗
PubMed2026

Proteomic and machine learning analysis predicts treatment response signatures in Myasthenia Gravis.

BACKGROUND: Myasthenia gravis (MG) is a prototypical antibody-mediated autoimmune disease with variable treatment responses with a need for biomarkers to guide therapeutic decision making. Proteomic profiling, coupled with machine learning, offers a hypothesis-free approach to identify multi-protein signatures associated with treatment response. METHODS: We analyzed sera collected at entry (baseline) from participants in a phase 3 trial randomized trial comparing thymectomy plus prednisone versus prednisone alone, along with matched controls using liquid chromatography-mass spectrometry. We derived disease-specific proteomic signatures and evaluated associations between baseline proteins and 6-month clinical outcomes using multiple machine-learning approaches with internal validation. RESULTS: Baseline serum proteomes distinguished MG from controls, with pathway enrichment implicating complement activation, immunoglobulin production, and T-cell receptor signaling. Distinct protein panels predicted 6-month clinical improvement within each treatment arm. In the thymectomy-plus-prednisone group, models captured non-linear relationships of predictive proteins in contrast with the predominant additive patterns observed in the prednisone-alone group. Predictive proteins were enriched for T-cell signaling and leukocyte trafficking functions, providing insight into treatment-specific biology. CONCLUSIONS: Baseline serum proteomics captures core disease characteristics of MG and predicts short-term clinical response in a treatment-specific manner. While our results require validation in independent cohorts, these findings could enable biomarker-guided selection of thymectomy, refine risk stratification, and furnish mechanistic readouts for future MG trials and clinical care. We aim to conduct future studies using -omic approaches to validate these baseline predictive biomarkers and pathways of treatment response in patients with MG.

Adult↗
PubMed2025

Proteomic signature of dementia risk in type 2 diabetes.

INTRODUCTION: Type 2 diabetes (T2D) significantly increases dementia risk, yet the molecular mechanisms underlying this association remain unclear. OBJECTIVES: This study aimed to identify protein signatures that distinguish dementia risk in T2D patients, develop a proteomic prediction model, and elucidate biological pathways connecting T2D and dementia. METHODS: We analyzed 2,920 plasma proteins from 52,958 participants (including 3,292 with T2D) in the UK Biobank Pharma Proteomics Project with a median follow-up of 14.6 years. Cox regression models with interaction terms identified T2D-specific protein associations with dementia risk. Machine learning models were developed to predict dementia in T2D patients. Pathway analysis and weighted gene co-expression network analysis identified biological mechanisms linking T2D and dementia. RESULTS: We identified 471 proteins with significant interaction effects between T2D and dementia risk. In non-T2D individuals, elevated levels of neuronal pentraxin receptor (NPTXR, HR = 0.74, 95 %CI:0.66-0.83) and carbonic anhydrase 14 (CA14, HR = 0.67, 95 %CI:0.60-0.75) were exclusively associated with decreased dementia risk. Conversely, in T2D patients, elevated rho guanine nucleotide exchange factor 12 (ARHGEF12, HR = 1.45, 95 %CI:1.10-1.91) was specifically associated with increased dementia risk. A 51-protein model accurately predicted 15-year dementia risk in T2D patients (AUC = 0.835, C-index = 0.829), outperforming conventional clinical risk scores and maintaining high accuracy for Alzheimer's disease and vascular dementia. Pathway analysis revealed enrichment of IL6-JAK-STAT3 signaling in T2D-related dementia, while dysregulation of fatty acid metabolism was specific to T2D-associated Alzheimer's disease. CONCLUSIONS: This large-scale proteomic analysis identifies specific molecular signatures that differentiate dementia risk in diabetic and non-diabetic populations, with potential applications for early risk stratification and targeted interventions. The identified pathways provide novel insights into the pathophysiological processes connecting T2D and dementia and suggest potential therapeutic targets.

Humans↗
PubMed2026

Predicting training outcomes for developmental dyslexia from EEG data.

Developmental dyslexia (DD) is characterised by lower-than-average reading abilities and is diagnosed in approximately 10% of individuals. The societal barriers may limit professional fulfilment and psychological wellbeing of individuals with DD, calling for the development of effective interventions to counteract them. As DD is associated with challenges in both phonological and visuo-attentional domains, different longitudinal training approaches were developed to strengthen them. However, they require a considerable amount of personal, social and economic resources and the outcomes may vary depending on individual differences in behavioural and neurophysiological functionality. Hence, predicting training outcomes might help in developing personalised treatment protocols and optimising the use of resources. In the present work we applied machine learning to resting-state EEG to predict longitudinal training outcomes in adults with DD enrolled in a randomized clinical trial. In particular, one group received a visuo-attentional training combined with transcranial alternating current stimulation (tACS), another group received visuo-attentional training with sham/placebo stimulation, and the third group received a phonological training with sham/placebo stimulation. The improvement in text reading speed was associated with spectral power in low-beta and individual frequencies in the alpha (IAF) and beta (IBF) bands, while the improvement in pseudoword reading was associated with IBF. The findings highlight the potential of capturing neural markers of treatment responsiveness in DD. Future studies should focus on the generalisability of predictive models to real-world settings, while investigating whether specific EEG markers predict responsiveness to distinct remediation protocols, thus supporting the development of personalised interventions.

Humans↗
PubMed2026

PGS-GS: a framework integrating polygenic scores and genomic selection in animal breeding.

Genomic prediction has become a central paradigm in biology, enabling quantitative inference of genetic contributions to complex traits across humans, animals, and plants. Although genomic research in human genetics and animal breeding shares a highly homologous methodological foundation, significant barriers persist in their analytical paradigms and application scenarios. This study aims to promote cross-disciplinary integration by introducing human-derived polygenic scores (PGS) algorithms into animal genomic selection (GS) and proposing a PGS-GS framework with a preliminary weighting-based implementation. We systematically benchmarked the predictive performance and computational efficiency of 20 algorithms, including classical linear models, machine learning, PGS, and PGS-GS using both array and whole-genome sequencing (WGS) data across four major agricultural species: beef cattle, sheep, pigs, and chickens. Our results demonstrate that PGS and PGS-GS algorithms achieve predictive accuracy competitive with genomic best linear unbiased prediction (GBLUP) while offering markedly higher computational efficiency. Moreover, incorporating PGS-derived prior information into weighted linear and non-linear models outperformed conventional weighted GBLUP. The results provide empirical evidence to inform algorithm selection and highlight the potential of integrating human-derived PGS methodologies into animal genomic prediction frameworks.

Animals↗
PubMed2026

Diagnostic performance of machine learning models for malignant and non-malignant pleural effusion: Systematic review and meta-analysis.

BACKGROUND: Accurately distinguishing malignant pleural effusion (MPE) from non-malignant pleural effusion is clinically important, but the generalisability and methodological quality of machine-learning (ML) models remain uncertain. METHODS: We searched eight databases to 23 April 2026. Diagnostic performance was pooled using random-effects and Reitsma bivariate models, and study quality was assessed using PROBAST+AI. RESULTS: Forty-two studies were included; 17 contributed to the AUC meta-analysis and 14 to the bivariate analysis. The pooled AUC was 0.90 (95 % CI 0.85-0.94; 95 % prediction interval 0.62-0.98), with sensitivity of 0.80 (95 % CI 0.77-0.83) and specificity of 0.87 (95 % CI 0.79-0.92). Only nine studies reported external, temporal or independent validation. Externally validated studies had a lower pooled AUC than studies without external validation (0.83 vs 0.92), with lower specificity observed in the two externally validated studies contributing sensitivity and specificity data. All 42 development assessments had high overall quality concerns, and all 42 model evaluations were judged at high risk of bias. CONCLUSIONS: ML models showed good apparent accuracy for distinguishing MPE from non-MPE, but the evidence was limited by substantial heterogeneity, high risk of bias and scarce external validation. The pooled estimates reflect the average performance of different selected models rather than the expected accuracy of a single clinical test. ML models should be regarded as adjuncts to existing diagnostic pathways until they are confirmed by rigorous multicentre prospective external validation and clinical-impact studies.

Humans↗
PubMed2026

An encyclopedia of human enhancer-gene regulatory interactions.

Identifying transcriptional enhancers and their target genes is essential for understanding gene regulation and the effect of human genetic variation on disease1-6. Here we create and evaluate a resource of more than 92 million enhancer-gene regulatory interactions across 1,458 biosamples covering 369 cell types and tissues, by integrating predictive models, chromatin states, three-dimensional contacts and large-scale genetic perturbations generated by the ENCODE Consortium7. We first create a systematic benchmarking pipeline to compare predictive models, assembling a dataset of 10,356 element-gene pairs measured in CRISPR perturbation experiments, more than 30,000 fine-mapped expression quantitative trait loci and 569 fine-mapped genome-wide association study (GWAS) variants linked to a probable causal gene. Using this framework, we develop ENCODE-rE2G, a predictive model achieving state-of-the-art performance across several prediction tasks, demonstrating that iterative perturbations and supervised machine learning can build increasingly accurate predictive models of enhancer regulation. Using ENCODE-rE2G, we build an encyclopedia of enhancer-gene regulatory interactions in the human genome, revealing global properties of enhancer networks, identifying differences in regulatory complexity across genes and improving analyses linking noncoding variants to target genes and cell types for common complex diseases. By interpreting the model, we find that beyond enhancer activity and three-dimensional enhancer-promoter contacts, additional features that guide enhancer-promoter communication include promoter class and enhancer-enhancer synergy. These genome-wide maps of enhancer-gene regulatory interactions, benchmarking software, predictive models and insights about enhancer function provide a valuable resource for future studies of gene regulation and human genetics.

Humans↗
PubMed2026

NMR metabolomics and glycomics for cancer detection in patients with non-specific symptoms: a prospective observational cohort study.

BACKGROUND: Early cancer diagnosis in patients with non-specific symptoms is limited by the lack of discriminatory tests. Within the Oxfordshire Suspected CANcer (SCAN) pathway, exploratory biomarker work showed that serum 1H NMR-based metabolomics can identify cancer with high accuracy. SCAN2 evaluated whether integrating metabolomics with glycomics provides complementary molecular information and improves discrimination in a clinically complex, real-world population. METHODS: Serum from 369 SCAN patients (59 cancers) was analysed using AXINON® System-derived NMR metabolomics and HPLC-MS glycomics. Machine-learning models were trained to predict cancer status, with performance assessed by receiver operating characteristic (ROC) analysis of pooled cross-validated predictions. To place cancer risk in a broader clinical context, a second classifier modelling alternative non-cancer diagnosis was incorporated, and mean predicted probabilities from both models were jointly projected into a two-dimensional space, maintaining strict separation of training and test data. FINDINGS: In the full cohort, integration of glycomics with metabolomics achieved an AUC of 0.814 (95% CI 0.808-0.820). In a refined sub-cohort excluding major comorbidities and selected cancer types (32 cancers, 277 non-cancers), performance improved to an AUC of 0.884 (95% CI 0.879-0.890). Discriminatory features included cancer-associated biantennary fucosylated glycans alongside amino acid metabolites (glutamate, histidine) and lipoprotein-related measures. A classifier distinguishing metastatic from non-metastatic disease (n = 29 vs. 30) achieved an AUC of 0.80. Joint probability analysis in the full cohort preserved cancer-associated signatures across comorbidity burden, with projection-based classification achieving an accuracy of 89.2% (95% CI 85.7-92.6). INTERPRETATION: These findings validate the SCAN1 metabolomic signature in a more clinically complex cohort and indicate that integrating glycomics with metabolomics provides complementary biological information for cancer discrimination. Joint probability analysis provides an interpretable framework for cancer risk stratification within multimorbid diagnostic pathways, supporting the clinical potential of scalable multi-omics blood testing. FUNDING: EPSRC, EU Horizon 2020, Wellcome/MLSTF, Novo Nordisk Foundation.

Humans↗
PubMed2007

Feature-map vectors: a new class of informative descriptors for computational drug discovery.

In order to develop robust machine-learning or statistical models for predicting biological activity, descriptors that capture the essence of the protein-ligand interaction are required. In the absence of structural information from X-ray or NMR experiments, deriving informative descriptors can be difficult. We have developed feature-map vectors (FMVs), a new class of descriptors based on chemical features, to address this challenge. FMVs, which are derived from the conformational models of a few actives, are low dimensional, problem specific, and highly interpretable. By using shape-based alignments and scoring with chemical features, FMVs can combine information about a molecule's shape and the pharmacophores it can match. In five validation studies, bag classifiers built using FMVs have shown high enrichments for identifying actives for five diverse targets: CDK2, 5-HT(3), DHFR, thrombin, and ACE. The interpretability of these descriptors has been demonstrated for CDK2 and 5-HT(3), where the method automatically discovers the standard literature pharmacophore.

Algorithms↗
PubMed2000

A probabilistic learning approach to whole-genome operon prediction.

We present a computational approach to predicting operons in the genomes of prokaryotic organisms. Our approach uses machine learning methods to induce predictive models for this task from a rich variety of data types including sequence data, gene expression data, and functional annotations associated with genes. We use multiple learned models that individually predict promoters, terminators and operons themselves. A key part of our approach is a dynamic programming method that uses our predictions to map every known and putative gene in a given genome into its most probable operon. We evaluate our approach using data from the E. coli K-12 genome.

Gene Expression Profiling↗
PubMed2025

Multi-criteria decision making and its application to in silico discovery of vaccine candidates for Toxoplasma gondii.

Vaccine discovery against eukaryotic parasites is not trivial and few exist. Reverse vaccinology is an in silico vaccine discovery approach, designed to identify vaccine candidates from the thousands of protein sequences encoded by a target genome. Previously, we produced the Vacceed bioinformatics pipeline for identification of parasite membrane and excreted/secreted proteins that were likely be exposed to the hosts immune system. More recently, we improved upon machine learning as the final decision-making process to identify parasite proteins that induce a protective response in an animal model. Subsequently, we combined Vacceed with metrics on B and T cell epitope types to produce a new in silico discovery workflow. In this study we extend this in silico workflow to the developability of proteins as vaccines by the incorporation of metrics on the physicochemical properties of proteins. To demonstrate this process, every Toxoplasma gondii protein was ranked in its capacity to provide exposure to the immune system (Vacceed exposure score), presence of epitopes and solubility characteristics by several multicriteria decision making (MCDM) tools (such as TOPSIS, VIKOR and MABAC). A consensus rank was subsequently generated from the results of these tools using a variety of aggregate ranking methods. Levels of uncertainty in the aggregate protein rankings was assessed by conformal interval prediction in association with a machine learning model. Several of the top ranked proteins identified by this approach were novel, uncharacterized membrane transporters or proteins associated with RNA metabolism. In conclusion, MCDM automated the decision making using well known algorithms while conformal prediction intervals varied significantly across the 8000+ proteins of T. gondii. Highly ranked proteins (e.g. the top 100) typically generated low prediction intervals, providing high levels of confidence in their ranks.

Toxoplasma↗
PubMed2006

Predicting the effect of missense mutations on protein function: analysis with Bayesian networks.

BACKGROUND: A number of methods that use both protein structural and evolutionary information are available to predict the functional consequences of missense mutations. However, many of these methods break down if either one of the two types of data are missing. Furthermore, there is a lack of rigorous assessment of how important the different factors are to prediction. RESULTS: Here we use Bayesian networks to predict whether or not a missense mutation will affect the function of the protein. Bayesian networks provide a concise representation for inferring models from data, and are known to generalise well to new data. More importantly, they can handle the noisy, incomplete and uncertain nature of biological data. Our Bayesian network achieved comparable performance with previous machine learning methods. The predictive performance of learned model structures was no better than a naïve Bayes classifier. However, analysis of the posterior distribution of model structures allows biologically meaningful interpretation of relationships between the input variables. CONCLUSION: The ability of the Bayesian network to make predictions when only structural or evolutionary data was observed allowed us to conclude that structural information is a significantly better predictor of the functional consequences of a missense mutation than evolutionary information, for the dataset used. Analysis of the posterior distribution of model structures revealed that the top three strongest connections with the class node all involved structural nodes. With this in mind, we derived a simplified Bayesian network that used just these three structural descriptors, with comparable performance to that of an all node network.

Algorithms↗
PubMed2025

Whole-genome phenotype prediction with machine learning: open problems in bacterial genomics.

MOTIVATION: How can we identify causal genetic mechanisms governing bacterial traits? Initial efforts entrusting machine learning models to handle the task of predicting phenotype from genotype yield high accuracy scores. However, attempts to extract meaningful interpretations from the predictive models are found to be corrupted by falsely identified 'causal' features. Relying solely on pattern recognition and correlations is unreliable, significantly so in bacterial genomics settings where high-dimensionality and spurious associations are the norm. Though it is not yet clear whether we can overcome this hurdle, significant efforts are being made towards discovering potential high-risk bacterial genetic variants. In view of this, we set up open problems surrounding phenotype prediction from bacterial whole-genome datasets and extending those approaches to learning causal effects, and discuss challenges that impact the reliability of a machine's decision-making when faced with datasets of this nature. RESULTS: We identify major sources of non-injectivity in the formulation of the genotype-to-phenotype mapping function-linkage-disequilibrium, limited sampling, information loss in representations, unmeasured confounders and observational noise-and analyse their implications for machine learning applications. Using a collection of 4,140 Staphylococcus aureus isolates, we illustrate challenges surrounding the defined open problems. AVAILABILITY AND IMPLEMENTATION: Raw sequencing data are available from the European Nucleotide Archive (ENA) under project accessions ERP001012, PRJEB3174, PRJEB2655, PRJEB2756, and PRJEB2944. Assemblies and annotations were generated with the Sanger bacterial pipeline (https://github.com/sanger-pathogens/vr-codebase) and unitigs extracted using DBGWAS (https://gitlab.com/leoisl/dbgwas).

Machine Learning↗
PubMed1994

A shape-based machine learning tool for drug design.

Building predictive models for iterative drug design in the absence of a known target protein structure is an important challenge. We present a novel technique, Compass, that removes a major obstacle to accurate prediction by automatically selecting conformations and alignments of molecules without the benefit of a characterized active site. The technique combines explicit representation of molecular shape with neural network learning methods to produce highly predictive models, even across chemically distinct classes of molecules. We apply the method to predicting human perception of musk odor and show how the resulting models can provide graphical guidance for chemical modifications.

Algorithms↗
PubMed1997

An evaluation of machine-learning methods for predicting pneumonia mortality.

This paper describes the application of eight statistical and machine-learning methods to derive computer models for predicting mortality of hospital patients with pneumonia from their findings at initial presentation. The eight models were each constructed based on 9847 patient cases and they were each evaluated on 4352 additional cases. The primary evaluation metric was the error in predicted survival as a function of the fraction of patients predicted to survive. This metric is useful in assessing a model's potential to assist a clinician in deciding whether to treat a given patient in the hospital or at home. We examined the error rates of the models when predicting that a given fraction of patients will survive. We examined survival fractions between 0.1 and 0.6. Over this range, each model's predictive error rate was within 1% of the error rate of every other model. When predicting that approximately 30% of the patients will survive, all the models have an error rate of less than 1.5%. The models are distinguished more by the number of variables and parameters that they contain than by their error rates; these differences suggest which models may be the most amenable to future implementation as paper-based guidelines.

Artificial Intelligence↗
PubMed2025

Integrating Imaging-Derived Clinical Endotypes with Plasma Proteomics and External Polygenic Risk Scores Enhances Coronary Microvascular Disease Risk Prediction.

Coronary microvascular disease (CMVD) is an underdiagnosed but significant contributor to the burden of ischemic heart disease, characterized by angina and myocardial infarction. The development of risk prediction models such as polygenic risk scores (PRS) for CMVD has been limited by a lack of large-scale genome-wide association studies (GWAS). However, there is significant overlap between CMVD and enrollment criteria for coronary artery disease (CAD) GWAS. In this study, we developed CMVD PRS models by selecting variants identified in a CMVD GWAS and applying weights from an external CAD GWAS, using CMVD-associated loci as proxies for the genetic risk. We integrated plasma proteomics, clinical measures from perfusion PET imaging, and PRS to evaluate their contributions to CMVD risk prediction in comprehensive machine and deep learning models. We then developed a novel unsupervised endotyping framework for CMVD from perfusion PET-derived myocardial blood flow data, revealing distinct patient subgroups beyond traditional case-control definitions. This imaging-based stratification substantially improved classification performance alongside plasma proteomics and PRS, achieving AUROCs between 0.65 and 0.73 per class, significantly outperforming binary classifiers and existing clinical models, highlighting the potential of this stratification approach to enable more precise and personalized diagnosis by capturing the underlying heterogeneity of CMVD. This work represents the first application of imaging-based endotyping and the integration of genetic and proteomic data for CMVD risk prediction, establishing a framework for multimodal modeling in complex diseases.

Cardiovascular Disease↗
PubMed2000

Machine learning for survival analysis: a case study on recurrence of prostate cancer.

Machine learning techniques have recently received considerable attention, especially when used for the construction of prediction models from data. Despite their potential advantages over standard statistical methods, like their ability to model non-linear relationships and construct symbolic and interpretable models, their applications to survival analysis are at best rare, primarily because of the difficulty to appropriately handle censored data. In this paper we propose a schema that enables the use of classification methods--including machine learning classifiers--for survival analysis. To appropriately consider the follow-up time and censoring, we propose a technique that, for the patients for which the event did not occur and have short follow-up times, estimates their probability of event and assigns them a distribution of outcome accordingly. Since most machine learning techniques do not deal with outcome distributions, the schema is implemented using weighted examples. To show the utility of the proposed technique, we investigate a particular problem of building prognostic models for prostate cancer recurrence, where the sole prediction of the probability of event (and not its probability dependency on time) is of interest. A case study on preoperative and postoperative prostate cancer recurrence prediction shows that by incorporating this weighting technique the machine learning tools stand beside modern statistical methods and may, by inducing symbolic recurrence models, provide further insight to relationships within the modeled data.

Artificial Intelligence↗