SEARCH · PubMed Health

Results for “Machine learning integration”

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 325 records · Page 18Linked to original sources

PubMed2026

mmContext: an open framework for multimodal contrastive learning of omics and text data.

SUMMARY: Multimodal approaches are increasingly leveraged for integrating omics data with textual biological knowledge. Yet there is still no accessible, standardized framework that enables systematic comparison of omics representations with different text encoders within a unified workflow. We present mmContext, a lightweight and extensible multimodal embedding framework built on top of the open-source Sentence Transformers library. The software allows researchers to train or apply models that jointly embed omics and text data using any numeric representation stored in an AnnData.obsm layer and any text encoder available in Hugging Face. mmContext supports integration of diverse biological text sources and provides pipelines for training, evaluation, and data preparation. We train and evaluate models for a RNA-Seq and text integration task, and demonstrate their utility through zero-shot classification of cell types and diseases across four independent datasets. By releasing all models, datasets, and tutorials openly, mmContext enables reproducible and accessible multimodal learning for omics-text integration. AVAILABILITY AND IMPLEMENTATION: Pretrained checkpoints and full source code for our custom MMContextEncoder are available on Hugging Face huggingface.co/jo-mengr. The Python package github.com/mengerj/mmcontext provides the model implementation and training and evaluation scripts for custom training. The releases for the publication can be accessed via zenodo: adata_hf_datasets: doi.org/10.5281/zenodo.19185217 and mmContext: doi.org/10.5281/zenodo.19185493.

Computational Biology↗

PubMed2026

Integrated metagenomic and metabolomic analysis identifies severity-specific inflammatory and metabolic signatures in post-stroke depression.

Post-stroke depression (PSD) is a common complication that significantly impacts patient prognosis. This study aimed to systematically characterize the associations among gut microbial ecology, metabolic profiles, and inflammatory responses across different severities of PSD. We conducted metagenomic sequencing, non-targeted metabolomics, and serum cytokine analysis (IL-1β, IL-6, IL-10, IL-18, TNF-α, IFN-γ, and CRP) in 91 patients with varying degrees of PSD and non-PSD controls. Bioinformatics analyzes were employed to construct multi-omics association networks and machine learning models. Results indicated that PSD patients exhibited significantly increased gut microbiota alpha-diversity, suggesting dysbiosis. Mild depression was characterized by compensatory neural signaling activation, whereas the moderate depression group exhibited abnormalities in tryptophan/indole metabolism, oxidative stress-related metabolic imbalances, and functional decompensation. Further analyzes suggested that Alistipes, Blautia_A, Evtepia gabavorous, and Lachnospira were associated with inflammatory features, GABA-related metabolic alterations, aromatic amino acid/indole metabolism, and lipid-amino acid metabolism, respectively. Under a more rigorous 10-fold cross-validation framework, the performance of different multi-omics combination models showed heterogeneity; however, some combinations still demonstrated superior discriminatory ability compared to single-omics approaches. This study provides multi-omics clues suggesting associations between different PSD severity levels and features such as increased Alistipes abundance, reduced antioxidant capacity, and altered tryptophan metabolism. It provides candidate biomarker combinations that may be useful for PSD stratification and suggests that the gut microbiome may represent a potential target for future PSD intervention. In summary, PSD may be associated with dynamic alterations along the "gut-brain-inflammation-metabolism" axis. These findings provide integrated evidence for microbial, metabolic, and inflammatory abnormalities across different PSD severity levels, but still require validation in larger samples, longitudinal cohorts, and mechanistic studies.

Humans↗

PubMed2025

Distinct immune-metabolic phenotypes underlie poor coronary collateral circulation.

BACKGROUND: Coronary collateral circulation (CCC) significantly impacts myocardial perfusion and clinical outcomes in coronary artery disease patients, yet the underlying molecular heterogeneity remains inadequately characterized. OBJECTIVE: To identify distinct molecular phenotypes in patients with poor CCC, validate these phenotypes using clinical parameters, and evaluate their prognostic implications. METHODS: This study enrolled 149 patients (80 with good CCC and 69 with poor CCC) for high-throughput proteomic profiling. Unsupervised consensus clustering identified molecular subtypes within poor CCC patients, followed by differential expression analysis and KEGG pathway enrichment. Boruta feature selection was implemented, and multiple machine learning algorithms were tested on clinical data, with XGBoost optimization (accuracy 80.0%, F1-score 80.31%) and SHAP value interpretation. External validation was performed using the MIMIC database. Kaplan-Meier analysis and Cox regression models assessed major adverse cardiovascular events (MACE). RESULTS: Two distinct phenotypes emerged among poor CCC patients: Cluster 1 (n = 39, Complement-Driven Vascular Remodeling [CDVR]) and Cluster 2 (n = 30, Immuno-Thrombotic Myocardial Dysfunction [ITMD]). An XGBoost model incorporating fasting glucose, eosinophil percentage, and HbA1c achieved excellent discrimination (AUC > 0.91). External validation confirmed the phenotype-specific clinical patterns. Notably, Cluster 2 demonstrated significantly higher MACE incidence compared to Cluster 1 (Log-rank p < 0.05), with KEGG analysis revealing significant upregulation of platelet activation, diabetic cardiomyopathy, and metabolic pathways in the ITMD phenotype. CONCLUSION: Poor CCC encompasses distinct immune-metabolic phenotypes that can be accurately classified using integrated proteomic-clinical modeling. This classification enables more precise risk stratification and may guide personalized therapeutic strategies for coronary artery disease patients with inadequate collateralization.

Humans↗

PubMed2026

Can host genetics transform the sustainable control of tropical theileriosis? Insights from the Tick-Theileria interface.

Tropical theileriosis, caused by the tick-transmitted apicomplexan parasite Theileria annulata, remains a major constraint on cattle production across North Africa, the Mediterranean basin, the Middle East and South Asia. Current control depends on acaricides, the theilericidal drug buparvaquone and live attenuated schizont vaccines, but acaricide resistance, buparvaquone-resistance mutations and the logistical demands of vaccination are eroding the sustainability of these tools. Host genetics offers a complementary and durable alternative. Indigenous Bos indicus breeds are consistently more resistant to ticks and tolerate T. annulata infection better than exotic Bos taurus cattle, and this advantage has a measurable heritable component. Unlike previous reviews, which treat tick resistance, T. annulata immunobiology and livestock genomic selection as separate subjects, we integrate all three and assess host genetics specifically against the failure modes of current control. We review the tick, parasite and host interface, the evidence for natural resistance, and the genetic and immunological mechanisms involved, including signal-regulatory protein, bovine major histocompatibility complex class II and inflammatory pathway genes. We then assess whether genomic selection, multi-omics, machine learning and gene editing can translate these mechanisms into resistant cattle, and we weigh the biological, economic and infrastructural barriers to implementation. The evidence indicates that host genetics will not replace existing control but could reduce reliance on acaricides and chemotherapy. That contribution remains prospective rather than demonstrated: no resistance marker for T. annulata has yet been validated, prediction accuracies are moderate and transfer poorly between breeds, and no endemic production system has implemented selection for resistance.

Animals↗

PubMed2006

Gene selection in cancer classification using sparse logistic regression with Bayesian regularization.

MOTIVATION: Gene selection algorithms for cancer classification, based on the expression of a small number of biomarker genes, have been the subject of considerable research in recent years. Shevade and Keerthi propose a gene selection algorithm based on sparse logistic regression (SLogReg) incorporating a Laplace prior to promote sparsity in the model parameters, and provide a simple but efficient training procedure. The degree of sparsity obtained is determined by the value of a regularization parameter, which must be carefully tuned in order to optimize performance. This normally involves a model selection stage, based on a computationally intensive search for the minimizer of the cross-validation error. In this paper, we demonstrate that a simple Bayesian approach can be taken to eliminate this regularization parameter entirely, by integrating it out analytically using an uninformative Jeffrey's prior. The improved algorithm (BLogReg) is then typically two or three orders of magnitude faster than the original algorithm, as there is no longer a need for a model selection step. The BLogReg algorithm is also free from selection bias in performance estimation, a common pitfall in the application of machine learning algorithms in cancer classification. RESULTS: The SLogReg, BLogReg and Relevance Vector Machine (RVM) gene selection algorithms are evaluated over the well-studied colon cancer and leukaemia benchmark datasets. The leave-one-out estimates of the probability of test error and cross-entropy of the BLogReg and SLogReg algorithms are very similar, however the BlogReg algorithm is found to be considerably faster than the original SLogReg algorithm. Using nested cross-validation to avoid selection bias, performance estimation for SLogReg on the leukaemia dataset takes almost 48 h, whereas the corresponding result for BLogReg is obtained in only 1 min 24 s, making BLogReg by far the more practical algorithm. BLogReg also demonstrates better estimates of conditional probability than the RVM, which are of great importance in medical applications, with similar computational expense. AVAILABILITY: A MATLAB implementation of the sparse logistic regression algorithm with Bayesian regularization (BLogReg) is available from http://theoval.cmp.uea.ac.uk/~gcc/cbl/blogreg/

Algorithms↗

PubMed2001

ABC stenosis morphology classification and outcome of coronary angioplasty: reassessment with computing techniques.

BACKGROUND: The American College of Cardiology/American Heart Association (ACC/AHA) stenosis morphology classification (MC) stratifies coronary lesions for probability of success and complications after coronary angioplasty (PTCA). Modern computing techniques were used to evaluate the individual predictive value of MC in random PTCA cases. METHODS AND RESULTS: MC was attributed to the target lesions by consensus of 2 observers. The predictive value regarding procedural success (PS) and major adverse cardiac events (MACE) of MC was analyzed by conventional logistic regression analyses and by inductive machine learning models. The study was adequately powered for the methods applied with 325 target lesions of 250 cases. Overall, PS decreased and MACE increased from type A to type C lesions. Regression analysis identified no single factor as predictive. Logistic regression showed an error rate of 42%. Machine learning techniques achieved an individual predictive error of only 10%, which could be further reduced to 2% by addition of parameters. For PS, MC parameters showed a high ranking for building the model. For MACE, variables of the medical history showed more impact. CONCLUSIONS: MC per se cannot individually predict PS or MACE. However, when all MC parameters are integrated together with additional lesion-specific and history variables, a high individual predictive value can be achieved. This technique may be clinically helpful for risk stratification in the catheterization laboratory and improvement of classification systems in interventional cardiology.

Algorithms↗

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

Global microbial DNA signatures of temperature and nutrient limitation across ecosystems.

Microbial genomes continuously adapt to environmental conditions, but identifying universal signatures of adaptation remains challenging. Here we show that environmental temperature can be accurately predicted across ecosystems from DNA composition alone (R2 = 0.75), using tetranucleotide frequencies from 1,235 marine and soil metagenomes and a machine learning approach. This predictive signal was also apparent within individual taxa, consistent with a fundamental temperature-associated signature. By contrast, GC content exhibited opposite correlations with temperature in soil (positive) and marine (negative) environments. This phenomenon was probably driven by differences in nutrient availability, as GC content increases with nutrients while nutrients decrease with temperature in marine samples. By integrating these observations, we identified specific tetranucleotides, with 50% GC, that displayed consistent and robust temperature correlations across environments and may have contributed to the stability of predictions. This work highlights metagenome-wide DNA-temperature associations, relevant for understanding microbial community responses to global changes.

Journal Article↗

PubMed2002

Neurotoxic lesions of the rat perirhinal cortex fail to disrupt the acquisition or performance of tests of allocentric spatial memory.

Rats with neurotoxic lesions of the perirhinal cortex (n = 9) were compared with sham controls (n = 14) on a working memory task in the radial arm maze. Rats were trained under varying levels of proactive interference and with different retention intervals. Finally, performance was assessed when the maze was switched to a novel room. None of these manipulations differentially impaired rats with perirhinal lesions. Rats were next trained on delayed matching-to-place in the water maze. Even with retention delays of 30 min, there was no evidence of a deficit. Although interactions between the perirhinal cortex and hippocampus may be important for integrating object-place information, the perirhinal cortex is often not necessary for tasks that selectively tax allocentric spatial memory.

Animals↗

PubMed2025

Integration of Genetic Information to Improve Brain Age Gap Estimation Models in the UK Biobank.

Neurodegeneration occurs when the body's central nervous system becomes impaired as a person ages, which can happen at an accelerated pace. Neurodegeneration impairs quality of life, affecting essential functions, including memory and the ability to self-care. Genetics play an important role in neurodegeneration and longevity. Brain age gap estimation (BrainAGE) is a biomarker that quantifies the difference between a machine learning model-predicted biological age of the brain and the true chronological age for healthy subjects; however, a large portion of the variance remains unaccounted for in these models, attributed to individual differences. This study focuses on predicting the BrainAGE more accurately, aided by genetic information associated with neurodegeneration. To achieve this, a BrainAGE model was developed based on MRI measures, and then the associated genes were determined with a Genome-Wide Association Study. Subsequently, genetic information was incorporated into the models. The incorporation of genetic information yielded improvements in the model performances by 7% to 12%, showing that the incorporation of genetic information can notably reduce unexplained variance. This work helps to define new ways of determining persons susceptible to neurological aging decline and reveals genes for targeted precision medicine therapies.

Humans↗

PubMed2004

Recognizing names in biomedical texts: a machine learning approach.

MOTIVATION: With an overwhelming amount of textual information in molecular biology and biomedicine, there is a need for effective and efficient literature mining and knowledge discovery that can help biologists to gather and make use of the knowledge encoded in text documents. In order to make organized and structured information available, automatically recognizing biomedical entity names becomes critical and is important for information retrieval, information extraction and automated knowledge acquisition. RESULTS: In this paper, we present a named entity recognition system in the biomedical domain, called PowerBioNE. In order to deal with the special phenomena of naming conventions in the biomedical domain, we propose various evidential features: (1) word formation pattern; (2) morphological pattern, such as prefix and suffix; (3) part-of-speech; (4) head noun trigger; (5) special verb trigger and (6) name alias feature. All the features are integrated effectively and efficiently through a hidden Markov model (HMM) and a HMM-based named entity recognizer. In addition, a k-Nearest Neighbor (k-NN) algorithm is proposed to resolve the data sparseness problem in our system. Finally, we present a pattern-based post-processing to automatically extract rules from the training data to deal with the cascaded entity name phenomenon. From our best knowledge, PowerBioNE is the first system which deals with the cascaded entity name phenomenon. Evaluation shows that our system achieves the F-measure of 66.6 and 62.2 on the 23 classes of GENIA V3.0 and V1.1, respectively. In particular, our system achieves the F-measure of 75.8 on the "protein" class of GENIA V3.0. For comparison, our system outperforms the best published result by 7.8 on GENIA V1.1, without help of any dictionaries. It also shows that our HMM and the k-NN algorithm outperform other models, such as back-off HMM, linear interpolated HMM, support vector machines, C4.5, C4.5 rules and RIPPER, by effectively capturing the local context dependency and resolving the data sparseness problem. Moreover, evaluation on GENIA V3.0 shows that the post-processing for the cascaded entity name phenomenon improves the F-measure by 3.9. Finally, error analysis shows that about half of the errors are caused by the strict annotation scheme and the annotation inconsistency in the GENIA corpus. This suggests that our system achieves an acceptable F-measure of 83.6 on the 23 classes of GENIA V3.0 and in particular 86.2 on the "protein" class, without help of any dictionaries. We think that a F-measure of 90 on the 23 classes of GENIA V3.0 and in particular 92 on the "protein" class, can be achieved through refining of the annotation scheme in the GENIA corpus, such as flexible annotation scheme and annotation consistency, and inclusion of a reasonable biomedical dictionary. AVAILABILITY: A demo system is available at http://textmining.i2r.a-star.edu.sg/NLS/demo.htm. Technology license is available upon the bilateral agreement.

Abstracting and Indexing↗

PubMed2006

A novel structure-based encoding for machine-learning applied to the inference of SH3 domain specificity.

MOTIVATION: Unravelling the rules underlying protein-protein and protein-ligand interactions is a crucial step in understanding cell machinery. Peptide recognition modules (PRMs) are globular protein domains which focus their binding targets on short protein sequences and play a key role in the frame of protein-protein interactions. High-throughput techniques permit the whole proteome scanning of each domain, but they are characterized by a high incidence of false positives. In this context, there is a pressing need for the development of in silico experiments to validate experimental results and of computational tools for the inference of domain-peptide interactions. RESULTS: We focused on the SH3 domain family and developed a machine-learning approach for inferring interaction specificity. SH3 domains are well-studied PRMs which typically bind proline-rich short sequences characterized by the PxxP consensus. The binding information is known to be held in the conformation of the domain surface and in the short sequence of the peptide. Our method relies on interaction data from high-throughput techniques and benefits from the integration of sequence and structure data of the interacting partners. Here, we propose a novel encoding technique aimed at representing binding information on the basis of the domain-peptide contact residues in complexes of known structure. Remarkably, the new encoding requires few variables to represent an interaction, thus avoiding the 'curse of dimension'. Our results display an accuracy >90% in detecting new binders of known SH3 domains, thus outperforming neural models on standard binary encodings, profile methods and recent statistical predictors. The method, moreover, shows a generalization capability, inferring specificity of unknown SH3 domains displaying some degree of similarity with the known data.

Algorithms↗

PubMed2002

Real-time computing without stable states: a new framework for neural computation based on perturbations.

A key challenge for neural modeling is to explain how a continuous stream of multimodal input from a rapidly changing environment can be processed by stereotypical recurrent circuits of integrate-and-fire neurons in real time. We propose a new computational model for real-time computing on time-varying input that provides an alternative to paradigms based on Turing machines or attractor neural networks. It does not require a task-dependent construction of neural circuits. Instead, it is based on principles of high-dimensional dynamical systems in combination with statistical learning theory and can be implemented on generic evolved or found recurrent circuitry. It is shown that the inherent transient dynamics of the high-dimensional dynamical system formed by a sufficiently large and heterogeneous neural circuit may serve as universal analog fading memory. Readout neurons can learn to extract in real time from the current state of such recurrent neural circuit information about current and past inputs that may be needed for diverse tasks. Stable internal states are not required for giving a stable output, since transient internal states can be transformed by readout neurons into stable target outputs due to the high dimensionality of the dynamical system. Our approach is based on a rigorous computational model, the liquid state machine, that, unlike Turing machines, does not require sequential transitions between well-defined discrete internal states. It is supported, as the Turing machine is, by rigorous mathematical results that predict universal computational power under idealized conditions, but for the biologically more realistic scenario of real-time processing of time-varying inputs. Our approach provides new perspectives for the interpretation of neural coding, the design of experiments and data analysis in neurophysiology, and the solution of problems in robotics and neurotechnology.

Action Potentials↗

PubMed1996

A generalized hidden Markov model for the recognition of human genes in DNA.

We present a statistical model of genes in DNA. A Generalized Hidden Markov Model (GHMM) provides the framework for describing the grammar of a legal parse of a DNA sequence (Stormo & Haussler 1994). Probabilities are assigned to transitions between states in the GHMM and to the generation of each nucleotide base given a particular state. Machine learning techniques are applied to optimize these probabilities using a standardized training set. Given a new candidate sequence, the best parse is deduced from the model using a dynamic programming algorithm to identify the path through the model with maximum probability. The GHMM is flexible and modular, so new sensors and additional states can be inserted easily. In addition, it provides simple solutions for integrating cardinality constraints, reading frame constraints, "indels", and homology searching. The description and results of an implementation of such a gene-finding model, called Genie, is presented. The exon sensor is a codon frequency model conditioned on windowed nucleotide frequency and the preceding codon. Two neural networks are used, as in (Brunak, Engelbrecht, & Knudsen 1991), for splice site prediction. We show that this simple model performs quite well. For a cross-validated standard test set of 304 genes [ftp:@www-hgc.lbl.gov/pub/genesets] in human DNA, our gene-finding system identified up to 85% of protein-coding bases correctly with a specificity of 80%. 58% of exons were exactly identified with a specificity of 51%. Genie is shown to perform favorably compared with several other gene-finding systems.

Chromosomes, Human↗

PubMed1999

Decision tree-based formation of consensus protein secondary structure prediction.

MOTIVATION: Prediction of protein secondary structure provides information that is useful for other prediction methods like fold recognition and ab initio 3D prediction. A consensus prediction constructed from the output of several methods should yield more reliable results than each of the individual methods. METHOD: We present an approach that reveals subtle but systematic differences in the output of different secondary structure prediction methods allowing the derivation of coherent consensus predictions. The method uses a machine learning technique that builds decision trees from existing data. RESULTS: The first results of our analysis show that consensus prediction of protein secondary structure may be improved both quantitatively and qualitatively.

Algorithms↗

PubMed2025

Proteomics-enabled learning machine algorithms enhance the prediction of cardiovascular diseases in patients with type 2 diabetes mellitus.

BACKGROUND AND AIMS: Estimating the risk of cardiovascular disease (CVD) complications in type 2 diabetes mellitus (T2DM) patients is critical in the medical decision-making process. This study aimed to use a machine learning technique combined with proteomics to develop personalized models for predicting CVD in patients with T2DM. METHODS AND RESULTS: In total, 874 patients with T2DM and 2,920 Olink proteins obtained from the UK Biobank were used in this study. Proteins were screened using Cox regression and LASSO regression. A basic model containing clinical features and a full model combining proteome and clinical features were constructed using the random survival forest algorithm. The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the predictive performance of the models and compare them with other CVD predictive models. Compared with the basic model, the full model performed better in predicting CVD, with time-dependent AUCs of 0.81 (3 years), 0.74 (5 years) and 0.74 (10 years) (0.77, 0.69 and 0.67). We calculated the risk scores of the Framingham, ASCVD and Score2-Diabetes models. The results revealed that the prediction performance of the full model was also better than that of the abovementioned models. In terms of differentiation accuracy, the results of the net reclassification improvement index and integrated discrimination improvement index showed that the full model can identify high-risk individuals more accurately (accuracy rate: 79% vs. 69%). CONCLUSIONS: Proteomics can be used to predict cardiovascular complications in diabetic patients. It is also necessary to consider the applicability of the model due to the limitations of the sample size and the constraints of proteomics in clinical applications.

Humans↗

PubMed2026

Investigating cross-organism prediction of prokaryotic essential proteins using unsupervised language model and ensemble strategy.

Cross-organism prediction of essential proteins is a critical task for drug discovery and microbial engineering, yet the generalizability of existing machine learning models across diverse species remains a significant challenge. In this study, we propose DeepPEP, a large language model-based framework designed to reliably transfer essential protein annotations between distantly related organisms. Utilizing 66 curated prokaryotic datasets, we systematically evaluated DeepPEP's cross-organism performance under various conditions. Initial pairwise predictions revealed a correlation between performance and evolutionary distance; however, further investigation demonstrated that integrating training data from multiple organisms yields superior predictive power. In a benchmark scenario designed to simulate real-world applications, DeepPEP outperformed the state-of-the-art tool Geptop 2.0, showcasing a robust ability to identify species-specific essential proteins. Finally, a case study on novel genomes confirmed the model's practical effectiveness. Our results suggest that DeepPEP is a powerful strategy for prokaryotic essential protein prediction, and the rigorous evaluation framework established in this study provides a new benchmark for the field.

Large Language Models↗

PubMed2026

Integrating single-cell transcriptomics to construct an oncogene-driven prognostic model and elucidate metabolic-immune crosstalk in hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths, its progression and treatment heterogeneity are mainly influenced by driver gene and tumor micro-environment (TME) interactions. Nevertheless, the mechanisms of this process at the single-cell level remain unclear. This study integrated TCGA and multi-center single-cell transcriptome data to identify a 575 genes HCC-specific core set, developing a single-cell "oncogene scoring" system to quantify individual carcinogenic activity. This score is significantly elevated in malignant and proliferative T cells and is closely associated with metabolic reprogramming, aberrant cell‒cell communication, and immunosuppressive phenotypes. Based on these characteristics, we constructed a machine learning-based Random Survival Forest (RSF) prognostic model validated in multiple independent cohorts, which classifies patients into distinct risk subtypes. The high-risk group exhibits genomic instability, increased tumor stemness, and immune evasion, while the low-risk group was more sensitive to drugs such as sorafenib. This study highlights the potential pathways by which high oncogenic activity is associated with HCC progression, suggesting a profound link with single-cell metabolic‒immune crosstalk. The constructed RSF model offers a promising computational framework for risk stratification and provides hypothesis-generating insights that may inform future personalized treatment strategies for HCC patients.

Hepatocellular carcinoma↗