SEARCH · PubMed Health

Results for “machine learning prediction”

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 577 records · Page 32Linked to original sources

PubMedDate not supplied

Handling missing values in support vector machine classifiers.

This paper discusses the task of learning a classifier from observed data containing missing values amongst the inputs which are missing completely at random. A non-parametric perspective is adopted by defining a modified risk taking into account the uncertainty of the predicted outputs when missing values are involved. It is shown that this approach generalizes the approach of mean imputation in the linear case and the resulting kernel machine reduces to the standard Support Vector Machine (SVM) when no input values are missing. Furthermore, the method is extended to the multivariate case of fitting additive models using componentwise kernel machines, and an efficient implementation is based on the Least Squares Support Vector Machine (LS-SVM) classifier formulation.

Algorithms↗

PubMed2026

A transcription factor regulatory atlas for activity inference and perturbation prediction.

Inferring transcription factor (TF) activity from transcriptomes and predicting transcriptome-wide responses to TF perturbations remain challenging, in part because available TF-mRNA resources often face a trade-off between precision and coverage and typically lack signed regulatory information. Here, we present TFActProfiler, a TF-mRNA resource and computational framework that learns signed, quantitative TF-mRNA regulatory coefficients by integrating heterogeneous prior evidence (ChIP-based, motif-based, and curated TF-mRNA annotations) with large-scale bulk and single-cell RNA-seq atlases. TFActProfiler contains 2 606 176 signed TF-mRNA interactions and improves TF activity inference in TF knockdown benchmarks relative to widely used regulon resources while retaining broad TF and target coverage. In addition, because the same learned regulatory coefficients can be used to model downstream transcriptional effects, TFActProfiler enables prediction of transcriptome-wide gene expression responses to TF knockdown without training on task-matched perturbation data. When perturbation datasets are available, TFActProfiler can be further refined to achieve performance comparable to state-of-the-art machine-learning baselines. By providing a direction-aware representation of TF-mRNA regulation for both activity inference and perturbation-response modeling, TFActProfiler supports systematic dissection of gene regulatory programs across diverse cellular contexts.

Transcription Factors↗

PubMed2025

Leveraging protein language models for cross-variant CRISPR/Cas9 sgRNA activity prediction.

MOTIVATION: Accurate prediction of single-guide RNA (sgRNA) activity is crucial for optimizing the CRISPR/Cas9 gene-editing system, as it directly influences the efficiency and accuracy of genome modifications. However, existing prediction methods mainly rely on large-scale experimental data of a single Cas9 variant to construct Cas9 protein (variants)-specific sgRNA activity prediction models, which limits their generalization ability and prediction performance across different Cas9 protein (variants), as well as their scalability to the continuously discovered new variants. RESULTS: In this study, we proposed PLM-CRISPR, a novel deep learning-based model that leverages protein language models to capture Cas9 protein (variants) representations for cross-variant sgRNA activity prediction. PLM-CRISPR uses tailored feature extraction modules for both sgRNA and protein sequences, incorporating a cross-variant training strategy and a dynamic feature fusion mechanism to effectively model their interactions. Extensive experiments demonstrate that PLM-CRISPR outperforms existing methods across datasets spanning seven Cas9 protein (variants) in three real-world scenarios, demonstrating its superior performance in handling data-scarce situations, including cases with few or no samples for novel variants. Comparative analyses with traditional machine learning and deep learning models further confirm the effectiveness of PLM-CRISPR. Additionally, motif analysis reveals that PLM-CRISPR accurately identifies high-activity sgRNA sequence patterns across diverse Cas9 protein (variants). Overall, PLM-CRISPR provides a robust, scalable, and generalizable solution for sgRNA activity prediction across diverse Cas9 protein (variants). AVAILABILITY AND IMPLEMENTATION: The source code can be obtained from https://github.com/CSUBioGroup/PLM-CRISPR.

CRISPR-Cas Systems↗

PubMed2023

Active learning of enhancer and silencer regulatory grammar in photoreceptors.

Cis-regulatory elements (CREs) direct gene expression in health and disease, and models that can accurately predict their activities from DNA sequences are crucial for biomedicine. Deep learning represents one emerging strategy to model the regulatory grammar that relates CRE sequence to function. However, these models require training data on a scale that exceeds the number of CREs in the genome. We address this problem using active machine learning to iteratively train models on multiple rounds of synthetic DNA sequences assayed in live mammalian retinas. During each round of training the model actively selects sequence perturbations to assay, thereby efficiently generating informative training data. We iteratively trained a model that predicts the activities of sequences containing binding motifs for the photoreceptor transcription factor Cone-rod homeobox (CRX) using an order of magnitude less training data than current approaches. The model's internal confidence estimates of its predictions are reliable guides for designing sequences with high activity. The model correctly identified critical sequence differences between active and inactive sequences with nearly identical transcription factor binding sites, and revealed order and spacing preferences for combinations of motifs. Our results establish active learning as an effective method to train accurate deep learning models of cis-regulatory function after exhausting naturally occurring training examples in the genome.

Journal Article↗

PubMed2003

Predicting risk of coronary artery disease from DNA microarray-based genotyping using neural networks and other statistical analysis tool.

This paper presents a novel approach for complex disease prediction that we have developed, exemplified by a study on risk of coronary artery disease (CAD). This multi-disciplinary approach straddles fields of microarray technology and genetics, neural networks (NN), data mining and machine learning, as well as traditional statistical analysis techniques, namely principal components analysis (PCA) and factor analysis (FA). A description of the biological background of the study is given, followed by a detailed description of how the problem has been modeled for analyses by neural networks and FA. A committee learning approach for NN has been used to improve generalization rates. We show that our NN approach is able to yield promising prediction results despite using only the most fundamental network structures. More interestingly, through the statistical analysis process, genes of similar biological functions have been clustered. In addition, a gene marker involved in breaking down lipids has been found to be the most correlated to CAD.

Algorithms↗

PubMed2006

Spritz: a server for the prediction of intrinsically disordered regions in protein sequences using kernel machines.

Intrinsically disordered proteins have long stretches of their polypeptide chain, which do not adopt a single native structure composed of stable secondary and tertiary structure in the absence of binding partners. The prediction of intrinsically disordered regions in proteins from sequence is increasingly becoming of interest, as the presence of many such regions in the complete genome sequences are discovered and important functional roles are associated with them. We have developed a machine learning approach based on two support vector machines (SVM) to discriminate disordered regions from sequence. The SVM are trained and benchmarked on two sets, representing long and short disordered regions. A preliminary version of Spritz was shown to perform consistently well at the recent biannual CASP-6 experiment [Critical Assessment of Techniques for Protein Structure Prediction (CASP), 2004]. The fully developed Spritz method is freely available as a web server at http://distill.ucd.ie/spritz/ and http://protein.cribi.unipd.it/spritz/.

Artificial Intelligence↗

PubMed2005

Analysis of mass spectral serum profiles for biomarker selection.

MOTIVATION: Mass spectrometric profiles of peptides and proteins obtained by current technologies are characterized by complex spectra, high dimensionality and substantial noise. These characteristics generate challenges in the discovery of proteins and protein-profiles that distinguish disease states, e.g. cancer patients from healthy individuals. We present low-level methods for the processing of mass spectral data and a machine learning method that combines support vector machines, with particle swarm optimization for biomarker selection. RESULTS: The proposed method identified mass points that achieved high prediction accuracy in distinguishing liver cancer patients from healthy individuals in SELDI-QqTOF profiles of serum. AVAILABILITY: MATLAB scripts to implement the methods described in this paper are available from the HWR's lab website http://lombardi.georgetown.edu/labpage

Algorithms↗

PubMed2005

Artificial neural networks for diagnosis and survival prediction in colon cancer.

ANNs are nonlinear regression computational devices that have been used for over 45 years in classification and survival prediction in several biomedical systems, including colon cancer. Described in this article is the theory behind the three-layer free forward artificial neural networks with backpropagation error, which is widely used in biomedical fields, and a methodological approach to its application for cancer research, as exemplified by colon cancer. Review of the literature shows that applications of these networks have improved the accuracy of colon cancer classification and survival prediction when compared to other statistical or clinicopathological methods. Accuracy, however, must be exercised when designing, using and publishing biomedical results employing machine-learning devices such as ANNs in worldwide literature in order to enhance confidence in the quality and reliability of reported data.

Colonic Neoplasms↗

PubMed2006

Identifying cysteines and histidines in transition-metal-binding sites using support vector machines and neural networks.

Accurate predictions of metal-binding sites in proteins by using sequence as the only source of information can significantly help in the prediction of protein structure and function, genome annotation, and in the experimental determination of protein structure. Here, we introduce a method for identifying histidines and cysteines that participate in binding of several transition metals and iron complexes. The method predicts histidines as being in either of two states (free or metal bound) and cysteines in either of three states (free, metal bound, or in disulfide bridges). The method uses only sequence information by utilizing position-specific evolutionary profiles as well as more global descriptors such as protein length and amino acid composition. Our solution is based on a two-stage machine-learning approach. The first stage consists of a support vector machine trained to locally classify the binding state of single histidines and cysteines. The second stage consists of a bidirectional recurrent neural network trained to refine local predictions by taking into account dependencies among residues within the same protein. A simple finite state automaton is employed as a postprocessing in the second stage in order to enforce an even number of disulfide-bonded cysteines. We predict histidines and cysteines in transition-metal-binding sites at 73% precision and 61% recall. We observe significant differences in performance depending on the ligand (histidine or cysteine) and on the metal bound. We also predict cysteines participating in disulfide bridges at 86% precision and 87% recall. Results are compared to those that would be obtained by using expert information as represented by PROSITE motifs and, for disulfide bonds, to state-of-the-art methods.

Amino Acid Sequence↗

PubMed2025

Harnessing deep learning for proteome-scale detection of amyloid signaling motifs.

MOTIVATION: Amyloid signaling sequences adopt the cross-β fold that is capable of self-replication in the templating process. Propagation of the amyloid fold from the receptor to the effector protein is used for signal transduction in the immune response pathways in animals, fungi, and bacteria. So far, a dozen of families of amyloid signaling motifs (ASMs) have been classified. Unfortunately, due to the wide variety of ASMs it is difficult to identify them in large protein databases available, which limits the possibility of conducting experimental studies. To date, various deep learning (DL) models have been applied across a range of protein-related tasks, including domain family classification and the prediction of protein structure and protein-protein interactions. RESULTS: In this study, we develop tailor-made bidirectional LSTM and BERT-based architectures to model ASM, and compare their performance against a state-of-the-art machine learning grammatical model. Our research is focused on developing a discriminative model of generalized ASMs, capable of detecting ASMs in large datasets. The DL-based models are trained on a diverse set of motif families and a global negative set, and used to identify ASMs from remotely related families. We analyze how both models represent the data and demonstrate that the DL-based approaches effectively detect ASMs, including novel motifs, even at the genome scale. AVAILABILITY AND IMPLEMENTATION: The models are provided as a Python package, asmscan-bilstm, and a Docker image at https://github.com/chrispysz/asmscan-proteinbert-run. The source code can be accessed at https://github.com/jakub-galazka/asmscan-bilstm and https://github.com/chrispysz/asmscan-proteinbert. Data and results are at https://github.com/wdyrka-pwr/ASMscan.

Deep Learning↗

PubMed2024

Blood-based DNA methylation and exposure risk scores predict PTSD with high accuracy in military and civilian cohorts.

BACKGROUND: Incorporating genomic data into risk prediction has become an increasingly useful approach for rapid identification of individuals most at risk for complex disorders such as PTSD. Our goal was to develop and validate Methylation Risk Scores (MRS) using machine learning to distinguish individuals who have PTSD from those who do not. METHODS: Elastic Net was used to develop three risk score models using a discovery dataset (n = 1226; 314 cases, 912 controls) comprised of 5 diverse cohorts with available blood-derived DNA methylation (DNAm) measured on the Illumina Epic BeadChip. The first risk score, exposure and methylation risk score (eMRS) used cumulative and childhood trauma exposure and DNAm variables; the second, methylation-only risk score (MoRS) was based solely on DNAm data; the third, methylation-only risk scores with adjusted exposure variables (MoRSAE) utilized DNAm data adjusted for the two exposure variables. The potential of these risk scores to predict future PTSD based on pre-deployment data was also assessed. External validation of risk scores was conducted in four independent cohorts. RESULTS: The eMRS model showed the highest accuracy (92%), precision (91%), recall (87%), and f1-score (89%) in classifying PTSD using 3730 features. While still highly accurate, the MoRS (accuracy = 89%) using 3728 features and MoRSAE (accuracy = 84%) using 4150 features showed a decline in classification power. eMRS significantly predicted PTSD in one of the four independent cohorts, the BEAR cohort (beta = 0.6839, p-0.003), but not in the remaining three cohorts. Pre-deployment risk scores from all models (eMRS, beta = 1.92; MoRS, beta = 1.99 and MoRSAE, beta = 1.77) displayed a significant (p < 0.001) predictive power for post-deployment PTSD. CONCLUSION: Results, especially those from the eMRS, reinforce earlier findings that methylation and trauma are interconnected and can be leveraged to increase the correct classification of those with vs. without PTSD. Moreover, our models can potentially be a valuable tool in predicting the future risk of developing PTSD. As more data become available, including additional molecular, environmental, and psychosocial factors in these scores may enhance their accuracy in predicting the condition and, relatedly, improve their performance in independent cohorts.

DNA methylation↗

PubMed2026

Quantitative assessment of the fingerprint evidential value using machine learning.

Fingerprints as physical evidence have long supported criminal investigation and adjudication. In practice, however, fingerprint identification relies mainly on examiners' experience. Furthermore, expert opinions tend to be categorical, even though the opinions with the same conclusion could differ substantially in evidential strength. To quantitatively assess fingerprint evidential value, this study proposes a machine learning-based framework as an interpretable decision-support tool. A lightweight residual one-dimensional convolutional neural network was constructed, incorporating channel recalibration and a similarity-driven attention mechanism to learn adaptive contribution weights for different matched minutiae (minutiae for short). Controlled experiments revealed that the predicted evidential value increased with the number of minutiae and was significantly influenced by the quality of minutiae. With 10 minutiae, the mean predicted scores were 4.49, 7.00, and 9.09 for blurred, moderately blurred, and clear minutiae, respectively. Multiple regression analysis indicated that replacing a pair of blurred minutiae with a pair of clear minutiae increased the score by 0.492, whereas replacing it with a pair of moderately blurred minutiae increased the score by only 0.216. By mapping predicted scores to graded levels of evidential strength, the framework contributes to a paradigm shift from categorical expert opinions to graded ones, helping courts evaluate fingerprint evidence more scientifically.

Humans↗

PubMed2024

Blood-based DNA methylation and exposure risk scores predict PTSD with high accuracy in military and civilian cohorts.

BACKGROUND: Incorporating genomic data into risk prediction has become an increasingly popular approach for rapid identification of individuals most at risk for complex disorders such as PTSD. Our goal was to develop and validate Methylation Risk Scores (MRS) using machine learning to distinguish individuals who have PTSD from those who do not. METHODS: Elastic Net was used to develop three risk score models using a discovery dataset (n = 1226; 314 cases, 912 controls) comprised of 5 diverse cohorts with available blood-derived DNA methylation (DNAm) measured on the Illumina Epic BeadChip. The first risk score, exposure and methylation risk score (eMRS) used cumulative and childhood trauma exposure and DNAm variables; the second, methylation-only risk score (MoRS) was based solely on DNAm data; the third, methylation-only risk scores with adjusted exposure variables (MoRSAE) utilized DNAm data adjusted for the two exposure variables. The potential of these risk scores to predict future PTSD based on pre-deployment data was also assessed. External validation of risk scores was conducted in four independent cohorts. RESULTS: The eMRS model showed the highest accuracy (92%), precision (91%), recall (87%), and f1-score (89%) in classifying PTSD using 3730 features. While still highly accurate, the MoRS (accuracy = 89%) using 3728 features and MoRSAE (accuracy = 84%) using 4150 features showed a decline in classification power. eMRS significantly predicted PTSD in one of the four independent cohorts, the BEAR cohort (beta = 0.6839, p=0.006), but not in the remaining three cohorts. Pre-deployment risk scores from all models (eMRS, beta = 1.92; MoRS, beta = 1.99 and MoRSAE, beta = 1.77) displayed a significant (p < 0.001) predictive power for post-deployment PTSD. CONCLUSION: The inclusion of exposure variables adds to the predictive power of MRS. Classification-based MRS may be useful in predicting risk of future PTSD in populations with anticipated trauma exposure. As more data become available, including additional molecular, environmental, and psychosocial factors in these scores may enhance their accuracy in predicting PTSD and, relatedly, improve their performance in independent cohorts.

Humans↗

PubMed2003

A new perspective on V3 phenotype prediction.

The particular coreceptor used by a strain of HIV-1 to enter a host cell is highly indicative of its pathology. HIV-1 coreceptor usage is primarily determined by the amino add sequences of the V3 loop region of the viral envelope glycoprotein. The canonical approach to sequence-based prediction of coreceptor usage was derived via statistical analysis of a less reliable and significantly smaller data set than is presently available. We aimed to produce a superior phenotypic classifier by applying modern machine learning (ML) techniques to the current database of V3 loop sequences with known phenotype. The trained classifiers along with the sequence data are available for public use at the supplementary website: http://genomiac2.ucsd.edu:8080/wetcat/v3.html and http://www.cs.waikato.ac.nz/ml/weka[corrected].

Amino Acid Sequence↗

PubMedDate not supplied

Granular support vector machines with association rules mining for protein homology prediction.

OBJECTIVE: Protein homology prediction between protein sequences is one of critical problems in computational biology. Such a complex classification problem is common in medical or biological information processing applications. How to build a model with superior generalization capability from training samples is an essential issue for mining knowledge to accurately predict/classify unseen new samples and to effectively support human experts to make correct decisions. METHODOLOGY: A new learning model called granular support vector machines (GSVM) is proposed based on our previous work. GSVM systematically and formally combines the principles from statistical learning theory and granular computing theory and thus provides an interesting new mechanism to address complex classification problems. It works by building a sequence of information granules and then building support vector machines (SVM) in some of these information granules on demand. A good granulation method to find suitable granules is crucial for modeling a GSVM with good performance. In this paper, we also propose an association rules-based granulation method. For the granules induced by association rules with high enough confidence and significant support, we leave them as they are because of their high "purity" and significant effect on simplifying the classification task. For every other granule, a SVM is modeled to discriminate the corresponding data. In this way, a complex classification problem is divided into multiple smaller problems so that the learning task is simplified. RESULTS AND CONCLUSIONS: The proposed algorithm, here named GSVM-AR, is compared with SVM by KDDCUP04 protein homology prediction data. The experimental results show that finding the splitting hyperplane is not a trivial task (we should be careful to select the association rules to avoid overfitting) and GSVM-AR does show significant improvement compared to building one single SVM in the whole feature space. Another advantage is that the utility of GSVM-AR is very good because it is easy to be implemented. More importantly and more interestingly, GSVM provides a new mechanism to address complex classification problems.

Algorithms↗

PubMed2003

A data analysis competition to evaluate machine learning algorithms for use in brain-computer interfaces.

We present three datasets that were used to conduct an open competition for evaluating the performance of various machine-learning algorithms used in brain-computer interfaces. The datasets were collected for tasks that included: 1) detecting explicit left/right (L/R) button press; 2) predicting imagined L/R button press; and 3) vertical cursor control. A total of ten entries were submitted to the competition, with winning results reported for two of the three datasets.

Algorithms↗

PubMed2025

CAKL: Commutative algebra k-mer learning of genomics.

Despite the availability of various sequence analysis models, comparative genomic analysis remains a challenge in genomics, genetics, and phylogenetics. Commutative algebra, a fundamental tool in algebraic geometry and number theory, has rarely been used in data and biological sciences. In this study, we introduce commutative algebra k-mer learning (CAKL) as the first-ever nonlinear algebraic framework for analyzing genomic sequences. CAKL bridges between commutative algebra, algebraic topology, combinatorics, and machine learning to establish a new mathematical paradigm for comparative genomic analysis. We evaluate its effectiveness on three tasks-genetic variant identification, phylogenetic tree analysis, and viral genome classification-typically requiring alignment-based, alignment-free, and machine-learning approaches, respectively. Across eleven datasets, CAKL outperforms five state-of-the-art sequence analysis methods, particularly in viral classification, and maintains stable predictive accuracy as dataset size increases, underscoring its scalability and robustness. This work ushers in a new era in commutative algebraic data analysis and learning.

Journal Article↗

PubMed2026

Multi-omics dynamic profiling reveals predictive biomarkers for first-line immunochemotherapy in extensive-stage small-cell lung cancer.

BACKGROUND: Extensive-stage small-cell lung cancer (ES-SCLC) is associated with a poor prognosis. Although first-line immunochemotherapy improves clinical outcomes, robust prognostic biomarkers for this treatment modality remain unavailable. The aim of this study was to identify non-invasive, easily accessible, and dynamically monitored biomarkers of ES-SCLC by machine learning integrating serum metabolomics, lipidomics, and proteomics at multiple time points. METHODS: A total of 816 serum samples were collected from ES-SCLC patients receiving first-line immunotherapy combined with chemotherapy or first-line chemotherapy for metabolomics, lipidomics, and proteomics analysis. The immunochemotherapy cohort was randomly divided into training and validation subsets at a 6:4 ratio. Biomarkers were identified using machine learning algorithms, and their prognostic significance was evaluated through receiver operating characteristic (ROC) analysis, Kaplan–Meier survival analysis, and multivariate Cox regression. Potential metabolic pathways and mechanisms were further explored via integrated multi-omic analysis. RESULTS: The immunochemotherapy exhibited a prolonged median progression-free survival (PFS) and higher objective response rate (ORR) compared to the chemotherapy group. A total of 5 serum metabolites (uric acid, L-aspartate-semialdehyde, dimethisterone, xanthine, L-cysteine), 6 lipids (Cer d18:1/26:0, Cer d18:2/25:0, SM d18:1/20:1, SM d17:1/25:1, DG O-18:1_16:0, PS 18:0_24:0), and 3 proteins (ACIN1, ACSL4, PHGDH) were identified and constructed into independent prognostic models. Among patients receiving immunochemotherapy, those categorized as low-risk based on the model demonstrated significantly longer PFS compared with those in the high-risk group. These prognostic signatures also retained predictive value in patients who underwent second-line treatment with anlotinib plus immunochemotherapy. Integrated analysis revealed that glycine, serine, and threonine metabolism was the commonly enriched pathway across all three omics layers. Notably, PHGDH (protein), L-aspartate-semialdehyde and L-cysteine (metabolites), and PS (18:0_24:0) (lipid), key elements in this pathway, were all incorporated in the predictive model. In addition, models of the composition of these substances after one cycle of treatment can still predict the prognosis of patients. CONCLUSION: In this study, we constructed and validated a set of non-invasive, dynamically monitorable prognostic models (containing 5 metabolites, 6 lipids, and 3 proteins) using machine learning by integrating multiple time point data from the serum metabolome, lipid panel, and proteome to accurately distinguish the prognostic risk of patients with ES-SCLC receiving immunochemotherapy. PFS was significantly prolonged in patients in the low-risk group, and this model remains predictive in the subsequent second-line treatment with anlotinib in combination with immunochemotherapy. Glycine-serine-threonine metabolic pathway may be the key mechanism, of which PHGDH, L-aspartate semialdehyde, L-cysteine and PS (18:0_24:0) are the core predictors. This study provides the first multi-omics dynamic prognostic tool for ES-SCLC immunochemotherapy and reveals potential therapeutic targets.

Humans↗