Consensus Machine Learning Research Articles

Genetic Analysis Uncovers Potential Mechanisms Linking Juvenile ldiopathic Arthritisto Breast Cancer: A BioinformaticPilot Study

BackgroundIn recent years, concerns have emerged regarding the potential link between Juvenile idiopathic arthritis (JIA) and an elevated risk of developing breast cancer. However, the potential relationship between JIA and breast cancer is currently unclear. The objective of this study is to investigate the mechanism of JIA on cancer risk. MethodsUse the Bulk-seq data related to JIA, selected from the GEO database, to explore potential candidate genes using methods such as WGCNA and consensus machine learning labeling. Verify using breast cancer Bulk-seq data from TCGA and scRNA-seq analyses. ResultsA total of 2050 genes potentially related to JIA were identified by WGCNA, and after merged with differentially expressed genes, 43 potential candidate genes were finding. Subsequently, consensus machine learning label analysis was conducted on the aforementioned genes, and a total of 6 genes closely related to JIA were identified. Subsequently, in breast cancer, we found that PRRG4, NCR3 and CREB5 also had significant differences in TCGA. And it is closely related to prognosis. ScRNA-seq analysis showed that the expression of PRRG4 was different in T cells in JIA, and PRRG4 was mainly expressed in T cells in breast cancer. ConclusionsThe findings of this study support a mechanism between JIA and an increased risk of breast cancer.

Advancing lung adenocarcinoma prognosis and immunotherapy prediction with a multi-omics consensus machine learning approach.

Lung adenocarcinoma (LUAD) is a tumour characterized by high tumour heterogeneity. Although there are numerous prognostic and immunotherapeutic options available for LUAD, there is a dearth of precise, individualized treatment plans. We integrated mRNA, lncRNA, microRNA, methylation and mutation data from the TCGA database for LUAD. Utilizing ten clustering algorithms, we identified stable multi-omics consensus clusters (MOCs). These data were then amalgamated with ten machine learning approaches to develop a robust model capable of reliably identifying patient prognosis and predicting immunotherapy outcomes. Through ten clustering algorithms, two prognostically relevant MOCs were identified, with MOC2 showing more favourable outcomes. We subsequently constructed a MOCs-associated machine learning model (MOCM) based on eight MOCs-specific hub genes. Patients characterized by a lower MOCM score exhibited better overall survival and responses to immunotherapy. These findings were consistent across multiple datasets, and compared to many previously published LUAD biomarkers, our MOCM score demonstrated superior predictive performance. Notably, the low MOCM group was more inclined towards 'hot' tumours, characterized by higher levels of immune cell infiltration. Intriguingly, a significant positive correlation between GJB3 and the MOCM score (R = 0.77, p < 0.01) was discovered. Further experiments confirmed that GJB3 significantly enhances LUAD proliferation, invasion and migration, indicating its potential as a key target for LUAD treatment. Our developed MOCM score accurately predicts the prognosis of LUAD patients and identifies potential beneficiaries of immunotherapy, offering broad clinical applicability.

Unraveling the immunogenetic landscape of autism spectrum disorder: a comprehensive bioinformatics approach.

Autism spectrum disorder (ASD) is a disease characterized by social disorder. Recently, the population affected by ASD has gradually increased around the world. There are great difficulties in diagnosis and treatment at present. The ASD datasets were obtained from the Gene Expression Omnibus database and the immune-relevant genes were downloaded from a previously published compilation. Subsequently, we used WGCNA to screen the modules related to the ASD and immune. We also choose the best combination and screen out the core genes from Consensus Machine Learning Driven Signatures (CMLS). Subsequently, we evaluated the genetic correlation between immune cells and ASD used GNOVA. And pleiotropic regions identified by PLACO and CPASSOC between ASD and immune cells. FUMA was used to identify pleiotropic regions, and expression trait loci (EQTL) analysis was used to determine their expression in different tissues and cells. Finally, we use qPCR to detect the gene expression level of the core gene. We found a close relationship between neutrophils and ASD, and subsequently, CMLS identified a total of 47 potential candidate genes. Secondly, GNOVA showed a significant genetic correlation between neutrophils and ASD, and PLACO and CPASSOC identified a total of 14 pleiotropic regions. We annotated the 14 regions mentioned above and identified a total of 6 potential candidate genes. Through EQTL, we found that the CFLAR gene has a specific expression pattern in neutrophils, suggesting that it may serve as a potential biomarker for ASD and is closely related to its pathogenesis. In conclusion, our study yields unprecedented insights into the molecular and genetic heterogeneity of ASD through a comprehensive bioinformatics analysis. These valuable findings hold significant implications for tailoring personalized ASD therapies.

Predicting Elimination of Small-Molecule Drug Half-Life in Pharmacokinetics Using Ensemble and Consensus Machine Learning Methods.

Half-life is a significant pharmacokinetic parameter included in the excretion phase of absorption, distribution, metabolism, and excretion. It is one of the key factors for the successful marketing of drug candidates. Therefore, predicting half-life is of great significance in drug design. In this study, we employed eXtreme Gradient Boosting (XGboost), randomForest (RF), gradient boosting machine (GBM), and supporting vector machine (SVM) to build quantitative structure-activity relationship (QSAR) models on 3512 compounds and evaluated model performance by using root-mean-square error (RMSE), R2, and mean absolute error (MAE) metrics and interpreted features by SHapley Additive exPlanation (SHAP). Furthermore, we developed consensus models through integrating four individual models and validated their performance using a Y-randomization test and applicability domain analysis. Finally, matched molecular pair analysis was used to extract the transformation rules. Our results revealed that XGboost outperformed other individual models (RMSE = 0.176, R2 = 0.845, MAE = 0.141). The consensus model integrating all four models continued to enhance prediction performance (RMSE = 0.172, R2 = 0.856, MAE = 0.138). We evaluated the reliability, robustness, and generalization ability via Y-randomization test and applicability domain analysis. Meanwhile, we utilized SHAP to interpret features and employed matched molecular pair analysis to extract chemical transformation rules that provide suggestions for optimizing drug structure. In conclusion, we believe that the consensus model developed in this study serve as a reliable tool to evaluate half-life in drug discovery, and the chemical transformation rules concluded in this study could provide valuable suggestions in drug discovery.

In Silico Prediction of Oral Acute Rodent Toxicity Using Consensus Machine Learning.

Acute oral toxicity (AOT) is required for the classification and labeling of chemicals according to the global harmonized system (GHS). Acute oral toxicity studies are optimized to minimize the use of animals. However, with the advent of the three Rs principles and machine learning in toxicology, alternative in silico methods became a reasonable alternative approach for addressing the AOT of new chemical matter. Here, we describe the compilation of AOT data from a commercial database and the development of a consensus classification model after evaluating different combinations of molecular representations and machine learning algorithms. The model shows significantly better performance compared to publicly available AOT models. Its performance was evaluated on an external validation data set, which was compiled from the literature, and an applicability domain was deduced.

Combining bulk and single-cell RNA-sequencing data to develop an NK cell-related prognostic signature for hepatocellular carcinoma based on an integrated machine learning framework

BackgroundThe application of molecular targeting therapy and immunotherapy has notably prolonged the survival of patients with hepatocellular carcinoma (HCC). However, multidrug resistance and high molecular heterogeneity of HCC still prevent the further improvement of clinical benefits. Dysfunction of tumor-infiltrating natural killer (NK) cells was strongly related to HCC progression and survival benefits of HCC patients. Hence, an NK cell-related prognostic signature was built up to predict HCC patients’ prognosis and immunotherapeutic response.MethodsNK cell markers were selected from scRNA-Seq data obtained from GSE162616 data set. A consensus machine learning framework including a total of 77 algorithms was developed to establish the gene signature in TCGA–LIHC data set, GSE14520 data set, GSE76427 data set and ICGC–LIRI–JP data set. Moreover, the predictive efficacy on ICI response was externally validated by GSE91061 data set and PRJEB23709 data set.ResultsWith the highest C-index among 77 algorithms, a 11-gene signature was established by the combination of LASSO and CoxBoost algorithm, which classified patients into high- and low-risk group. The prognostic signature displayed a good predictive performance for overall survival rate, moderate to high predictive accuracy and was an independent risk factor for HCC patients’ prognosis in TCGA, GEO and ICGC cohorts. Compared with high-risk group, low-risk patients showed higher IPS–PD1 blocker, IPS–CTLA4 blocker, common immune checkpoints expression but lower TIDE score, which indicated low-risk patients might be prone to benefiting from ICI treatment. Moreover, a real-world cohort, PRJEB23709, also revealed better immunotherapeutic response in low-risk group.ConclusionsOverall, the present study developed a gene signature based on NK cell-related genes, which offered a novel platform for prognosis and immunotherapeutic response evaluation of HCC patients.

The Development of a Consensus Machine Learning Model for Hurricane Rapid Intensification Forecasts with Hurricane Weather Research and Forecasting (HWRF) Data

Abstract This study focused on developing a consensus machine learning (CML) model for tropical cyclone (TC) intensity-change forecasting, especially for rapid intensification (RI). This CML model was built upon selected classical machine learning models with the input data extracted from a high-resolution hurricane model, the Hurricane Weather Research and Forecasting (HWRF) system. The input data contained 21 or 34 RI-related predictors extracted from the 2018 version of HWRF (H218). This study found that TC inner-core predictors can be critical for improving RI predictions, especially the inner-core relative humidity. Moreover, this study emphasized the importance of performing resampling on an imbalanced input dataset. Edited nearest-neighbor and synthetic minority oversampling techniques improved the probability of detection (POD) by ∼10% for the RI class. This study also showed that the CML model has satisfactory performance on RI predictions compared to the operational models. CML reached 56% POD and 46% false alarm ratio (FAR), while the operational models had only 10%–30% POD but 50%–60% FAR. The CML performance on the non-RI classes was comparable to the operational models. The results indicated that, with proper and sufficient training data and RI-related predictors, CML has the potential to provide reliable probabilistic RI forecasts during a hurricane season.

Development of a Robust Consensus Modeling Approach for Identifying Cellular and Media Metabolites Predictive of Mesenchymal Stromal Cell Potency.

Mesenchymal stromal cells (MSCs) have shown promise in regenerative medicine applications due in part to their ability to modulate immune cells. However, MSCs demonstrate significant functional heterogeneity in terms of their immunomodulatory function because of differences in MSC donor/tissue source, as well as non-standardized manufacturing approaches. As MSC metabolism plays a critical role in their ability to expand to therapeutic numbers ex vivo, we comprehensively profiled intracellular and extracellular metabolites throughout the expansion process to identify predictors of immunomodulatory function (T-cell modulation and indoleamine-2,3-dehydrogenase (IDO) activity). Here, we profiled media metabolites in a non-destructive manner through daily sampling and nuclear magnetic resonance (NMR), as well as MSC intracellular metabolites at the end of expansion using mass spectrometry (MS). Using a robust consensus machine learning approach, we were able to identify panels of metabolites predictive of MSC immunomodulatory function for 10 independent MSC lines. This approach consisted of identifying metabolites in 2 or more machine learning models and then building consensus models based on these consensus metabolite panels. Consensus intracellular metabolites with high predictive value included multiple lipid classes (such as phosphatidylcholines, phosphatidylethanolamines, and sphingomyelins) while consensus media metabolites included proline, phenylalanine, and pyruvate. Pathway enrichment identified metabolic pathways significantly associated with MSC function such as sphingolipid signaling and metabolism, arginine and proline metabolism, and autophagy. Overall, this work establishes a generalizable framework for identifying consensus predictive metabolites that predict MSC function, as well as guiding future MSC manufacturing efforts through identification of high-potency MSC lines and metabolic engineering.

In silico drug repurposing by combining machine learning classification model and molecular dynamics to identify a potential OGT inhibitor

O-linked N-acetylglucosamine (O-GlcNAc) is a unique intracellular post-translational glycosylation at the hydroxyl group of serine or threonine residues in nuclear, cytoplasmic and mitochondrial proteins. The enzyme O-GlcNAc transferase (OGT) is responsible for adding GlcNAc, and anomalies in this process can lead to the development of diseases associated with metabolic imbalance, such as diabetes and cancer. Repurposing approved drugs can be an attractive tool to discover new targets reducing time and costs in the drug design. This work focuses on drug repurposing to OGT targets by virtual screening of FDA-approved drugs through consensus machine learning (ML) models from an imbalanced dataset. We developed a classification model using docking scores and ligand descriptors. The SMOTE approach to resampling the dataset showed excellent statistical values in five of the seven ML algorithms to create models from the training set, with sensitivity, specificity and accuracy over 90% and Matthew’s correlation coefficient greater than 0.8. The pose analysis obtained by molecular docking showed only H-bond interaction with the OGT C-Cat domain. The molecular dynamics simulation showed the lack of H-bond interactions with the C- and N-catalytic domains allowed the drug to exit the binding site. Our results showed that the non-steroidal anti-inflammatory celecoxib could be a potentially OGT inhibitor.

A Novel Weighted Consensus Machine Learning Model for COVID-19 Infection Classification Using CT Scan Images.

As COVID-19 has spread rapidly, detection of the COVID-19 infection from radiology and radiography images is probably one of the quickest ways to diagnose the patients. Many researchers found the necessity to utilize chest X-ray and chest computed tomography imaging to diagnose COVID-19 infection. In this paper, our objective is to minimize the false negatives and false positives in the detection process. Reduction in the number of false negatives minimizes community spread of the COVID-19 pandemic. Reducing false positives help people avoid mental trauma and wasteful expenses. This paper proposes a novel weighted consensus model to minimize the number of false negatives and false positives without compromising accuracy. In the proposed novel weighted consensus model, the accuracy of individual classification models is normalized. While predicting, different models predict different classes, and the sum of the normalized accuracy for a particular class is then considered based on a predefined threshold value. We used traditional Machine Learning classification algorithms like Linear Regression, Support Vector Machine, k-Nearest Neighbours, Decision Tree, and Random Forest for the weighted consensus experimental evaluation. We predicted the classes, which provided better insights into the condition. The proposed model can perform as well as the existing state-of-the-art technique in terms of accuracy (99.64%) and reduce false negatives and false positives.

A diagnostic miRNA signature for pulmonary arterial hypertension using a consensus machine learning approach.

BackgroundPulmonary arterial hypertension (PAH) is a rare but life shortening disease, the diagnosis of which is often delayed, and requires an invasive right heart catheterisation. Identifying diagnostic biomarkers may improve screening to identify patients at risk of PAH earlier and provide new insights into disease pathogenesis. MicroRNAs are small, non-coding molecules of RNA, previously shown to be dysregulated in PAH, and contribute to the disease process in animal models.MethodsPlasma from 64 treatment naïve patients with PAH and 43 disease and healthy controls were profiled for microRNA expression by Agilent Microarray. Following quality control and normalisation, the cohort was split into training and validation sets. Four separate machine learning feature selection methods were applied to the training set, along with a univariate analysis.Findings20 microRNAs were identified as putative biomarkers by consensus feature selection from all four methods. Two microRNAs (miR-636 and miR-187-5p) were selected by all methods and used to predict PAH diagnosis with high accuracy. Integrating microRNA expression profiles with their associated target mRNA revealed 61 differentially expressed genes verified in two independent, publicly available PAH lung tissue data sets. Two of seven potentially novel gene targets were validated as differentially expressed in vitro in human pulmonary artery smooth muscle cells.InterpretationThis consensus of multiple machine learning approaches identified two miRNAs that were able to distinguish PAH from both disease and healthy controls. These circulating miRNA, and their target genes may provide insight into PAH pathogenesis and reveal novel regulators of disease and putative drug targets.

Pushing the limits of solubility prediction via quality-oriented data selection.

SummaryAccurate prediction of the solubility of chemical substances in solvents remains a challenge. The sparsity of high-quality solubility data is recognized as the biggest hurdle in the development of robust data-driven methods for practical use. Nonetheless, the effects of the quality and quantity of data on aqueous solubility predictions have not yet been scrutinized. In this study, the roles of the size and the quality of data sets on the performances of the solubility prediction models are unraveled, and the concepts of actual and observed performances are introduced. In an effort to curtail the gap between actual and observed performances, a quality-oriented data selection method, which evaluates the quality of data and extracts the most accurate part of it through statistical validation, is designed. Applying this method on the largest publicly available solubility database and using a consensus machine learning approach, a top-performing solubility prediction model is achieved.

A Distance-Based Framework for the Characterization of Metabolic Heterogeneity in Large Sets of Genome-Scale Metabolic Models

SummaryGene expression and protein abundance data of cells or tissues belonging to healthy and diseased individuals can be integrated and mapped onto genome-scale metabolic networks to produce patient-derived models. As the number of available and newly developed genome-scale metabolic models increases, new methods are needed to objectively analyze large sets of models and to identify the determinants of metabolic heterogeneity. We developed a distance-based workflow that combines consensus machine learning and metabolic modeling techniques and used it to apply pattern recognition algorithms to collections of genome-scale metabolic models, both microbial and human. Model composition, network topology and flux distribution provide complementary aspects of metabolic heterogeneity in patient-specific genome-scale models of skeletal muscle. Using consensus clustering analysis we identified the metabolic processes involved in the individual responses to endurance training in older adults.

Lipophilicity prediction of peptides and peptide derivatives by consensus machine learning.

Lipophilicity prediction is routinely applied to small molecules and presents a working alternative to experimental log P or log D determination. For compounds outside the domain of classical medicinal chemistry these predictions lack accuracy, advocating the development of bespoke in silico approaches. Peptides and their derivatives and mimetics fill the structural gap between small synthetic drugs and genetically engineered macromolecules. Here, we present a data-driven machine learning method for peptide log D 7.4 prediction. A model for estimating the lipophilicity of short linear peptides consisting of natural amino acids was developed. In a prospective test, we obtained accurate predictions for a set of newly synthesized linear tri- to hexapeptides. Further model development focused on more complex peptide mimetics from the AstraZeneca compound collection. The results obtained demonstrate the applicability of the new prediction model to peptides and peptide derivatives in a log D 7.4 range of approximately -3 to 5, with superior accuracy to established lipophilicity models for small molecules.

Can human experts predict solubility better than computers?

In this study, we design and carry out a survey, asking human experts to predict the aqueous solubility of druglike organic compounds. We investigate whether these experts, drawn largely from the pharmaceutical industry and academia, can match or exceed the predictive power of algorithms. Alongside this, we implement 10 typical machine learning algorithms on the same dataset. The best algorithm, a variety of neural network known as a multi-layer perceptron, gave an RMSE of 0.985 log S units and an R2 of 0.706. We would not have predicted the relative success of this particular algorithm in advance. We found that the best individual human predictor generated an almost identical prediction quality with an RMSE of 0.942 log S units and an R2 of 0.723. The collection of algorithms contained a higher proportion of reasonably good predictors, nine out of ten compared with around half of the humans. We found that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median generated excellent predictivity. While our consensus human predictor achieved very slightly better headline figures on various statistical measures, the difference between it and the consensus machine learning predictor was both small and statistically insignificant. We conclude that human experts can predict the aqueous solubility of druglike molecules essentially equally well as machine learning algorithms. We find that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median is a powerful way of benefitting from the wisdom of crowds.

CAGI4 Crohn's exome challenge: Marker SNP versus exome variant models for assigning risk of Crohn disease.

Understanding the basis of complex trait disease is a fundamental problem in human genetics. The CAGI Crohn's Exome challenges are providing insight into the adequacy of current disease models by requiring participants to identify which of a set of individuals has been diagnosed with the disease, given exome data. For the CAGI4 round, we developed a method that used the genotypes from exome sequencing data only to impute the status of genome wide association studies marker SNPs. We then used the imputed genotypes as input to several machine learning methods that had been trained to predict disease status from marker SNP information. We achieved the best performance using Naïve Bayes and with a consensus machine learning method, obtaining an area under the curve of 0.72, larger than other methods used in CAGI4. We also developed a model that incorporated the contribution from rare missense variants in the exome data, but this performed less well. Future progress is expected to come from the use of whole genome data rather than exomes.

Shared Consensus Machine Learning Models for Predicting Blood Stage Malaria Inhibition.

The development of new antimalarial therapies is essential, and lowering the barrier of entry for the screening and discovery of new lead compound classes can spur drug development at organizations that may not have large compound screening libraries or resources to conduct high-throughput screens. Machine learning models have been long established to be more robust and have a larger domain of applicability with larger training sets. Screens over multiple data sets to find compounds with potential malaria blood stage inhibitory activity have been used to generate multiple Bayesian models. Here we describe a method by which Bayesian quantitative structure-activity relationship models, which contain information on thousands to millions of proprietary compounds, can be shared between collaborators at both for-profit and not-for-profit institutions. This model-sharing paradigm allows for the development of consensus models that have increased predictive power over any single model and yet does not reveal the identity of any compounds in the training sets.

In Silico Mining for Antimalarial Structure-Activity Knowledge and Discovery of Novel Antimalarial Curcuminoids.

Malaria is a parasitic tropical disease that kills around 600,000 patients every year. The emergence of resistant Plasmodium falciparum parasites to artemisinin-based combination therapies (ACTs) represents a significant public health threat, indicating the urgent need for new effective compounds to reverse ACT resistance and cure the disease. For this, extensive curation and homogenization of experimental anti-Plasmodium screening data from both in-house and ChEMBL sources were conducted. As a result, a coherent strategy was established that allowed compiling coherent training sets that associate compound structures to the respective antimalarial activity measurements. Seventeen of these training sets led to the successful generation of classification models discriminating whether a compound has a significant probability to be active under the specific conditions of the antimalarial test associated with each set. These models were used in consensus prediction of the most likely active from a series of curcuminoids available in-house. Positive predictions together with a few predicted as inactive were then submitted to experimental in vitro antimalarial testing. A large majority from predicted compounds showed antimalarial activity, but not those predicted as inactive, thus experimentally validating the in silico screening approach. The herein proposed consensus machine learning approach showed its potential to reduce the cost and duration of antimalarial drug discovery.

Integrative analysis of large-scale biological data sets

AbstractWe present two novel web-applications for microarray and gene/protein set analysis, ArrayMining.net and TopoGSA. These bioinformatics tools use integrative analysis methods, including ensemble and consensus machine learning techniques, as well as modular combinations of different analysis types, to extract new biological insights from experimental transcriptomics and proteomics data. They enable researchers to combine related algorithms and datasets to increase the robustness and accuracy of statistical analyses and exploit synergies of different computational methods, ranging from statistical learning to optimization and topological network analysis.

Integrative analysis of large-scale biological data sets

We present two novel web-applications for microarray and gene/protein set analysis, ArrayMining.net and TopoGSA. These bioinformatics tools use integrative analysis methods, including ensemble and consensus machine learning techniques, as well as modular combinations of different analysis types, to extract new biological insights from experimental transcriptomics and proteomics data. They enable researchers to combine related algorithms and datasets to increase the robustness and accuracy of statistical analyses and exploit synergies of different computational methods, ranging from statistical learning to optimization and topological network analysis.