Dataset Information

Multiple Feature Selection Strategies Identified Novel Cardiac Gene Expression Signature for Heart Failure.

ABSTRACT: Heart failure (HF) is a serious condition in which the support of blood pumped by the heart is insufficient to meet the demands of body at a normal cardiac filling pressure. Approximately 26 million patients worldwide are suffering from heart failure and about 17-45% of patients with heart failure die within 1-year, and the majority die within 5-years admitted to a hospital. The molecular mechanisms underlying the progression of heart failure have been poorly studied. We compared the gene expression profiles between patients with heart failure (n = 177) and without heart failure (n = 136) using multiple feature selection strategies and identified 38 HF signature genes. The support vector machine (SVM) classifier based on these 38 genes evaluated with leave-one-out cross validation (LOOCV) achieved great performance with sensitivity of 0.983 and specificity of 0.963. The network analysis suggested that the hub gene SMOC2 may play important roles in HF. Other genes, such as FCN3, HMGN2, and SERPINA3, also showed great promises. Our results can facilitate the early detection of heart failure and can reveal its molecular mechanisms.

SUBMITTER: Li D

PROVIDER: S-EPMC7693561 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Multiple Feature Selection Strategies Identified Novel Cardiac Gene Expression Signature for Heart Failure.

Li Dan D Lin Hong H Li Luyifei L

Frontiers in physiology 20201111

Heart failure (HF) is a serious condition in which the support of blood pumped by the heart is insufficient to meet the demands of body at a normal cardiac filling pressure. Approximately 26 million patients worldwide are suffering from heart failure and about 17-45% of patients with heart failure die within 1-year, and the majority die within 5-years admitted to a hospital. The molecular mechanisms underlying the progression of heart failure have been poorly studied. We compared the gene expres ...[more]

PMID: 33304275

Similar Datasets

Project description:ObjectivesThe purpose of this study was to identify where ultrasmall superparamagnetic particles of iron oxide (USPIO) locate to in myocardium, develop a methodology that differentiates active macrophage uptake of USPIO from passive tissue distribution; and investigate myocardial inflammation in cardiovascular diseases.BackgroundMyocardial inflammation is hypothesized to be a key pathophysiological mechanism of heart failure (HF), but human evidence is limited, partly because evaluation is challenging. USPIO-magnetic resonance imaging (MRI) potentially allows specific identification of myocardial inflammation but it remains unclear what the USPIO-MRI signal represents.MethodsHistological validation was performed using a murine acute myocardial infarction (MI) model. A multiparametric, multi-time-point MRI methodology was developed, which was applied in patients with acute MI (n = 12), chronic ischemic cardiomyopathy (n = 7), myocarditis (n = 6), dilated cardiomyopathy (n = 5), and chronic sarcoidosis (n = 5).ResultsUSPIO were identified in myocardial macrophages and myocardial interstitium. R1 time-course reflected passive interstitial distribution whereas multi-time-point R2* was also sensitive to active macrophage uptake. R2*/R1 ratio provided a quantitative measurement of myocardial macrophage infiltration. R2* behavior and R2*/R1 ratio were higher in infarcted (p = 0.001) and remote (p = 0.033) myocardium in acute MI and in chronic ischemic cardiomyopathy (infarct: p = 0.008; remote p = 0.010), and were borderline higher in DCM (p = 0.096), in comparison to healthy controls, but were no different in myocarditis or sarcoidosis. An R2*/R1 threshold of 25 had a sensitivity and specificity of 90% and 83%, respectively, for detecting active USPIO uptake.ConclusionsUSPIO are phagocytized by cardiac macrophages but are also passively present in myocardial interstitium. A multiparametric multi-time-point MRI methodology specifically identifies active myocardial macrophage infiltration. Persistent active macrophage infiltration is present in infarcted and remote myocardium in chronic ischemic cardiomyopathy, providing a substrate for HF.

Project description:Background In the last few years, multi-omics data, that is, datasets containing different types of high-dimensional molecular variables for the same samples, have become increasingly available. To date, several comparison studies focused on feature selection methods for omics data, but to our knowledge, none compared these methods for the special case of multi-omics data. Given that these data have specific structures that differentiate them from single-omics data, it is unclear whether different feature selection strategies may be optimal for such data. In this paper, using 15 cancer multi-omics datasets we compared four filter methods, two embedded methods, and two wrapper methods with respect to their performance in the prediction of a binary outcome in several situations that may affect the prediction results. As classifiers, we used support vector machines and random forests. The methods were compared using repeated fivefold cross-validation. The accuracy, the AUC, and the Brier score served as performance metrics. Results The results suggested that, first, the chosen number of selected features affects the predictive performance for many feature selection methods but not all. Second, whether the features were selected by data type or from all data types concurrently did not considerably affect the predictive performance, but for some methods, concurrent selection took more time. Third, regardless of which performance measure was considered, the feature selection methods mRMR, the permutation importance of random forests, and the Lasso tended to outperform the other considered methods. Here, mRMR and the permutation importance of random forests already delivered strong predictive performance when considering only a few selected features. Finally, the wrapper methods were computationally much more expensive than the filter and embedded methods. Conclusions We recommend the permutation importance of random forests and the filter method mRMR for feature selection using multi-omics data, where, however, mRMR is considerably more computationally costly. Supplementary Information The online version contains supplementary material available at 10.1186/s12859-022-04962-x.

Dataset Information

Multiple Feature Selection Strategies Identified Novel Cardiac Gene Expression Signature for Heart Failure.

Publications

Multiple Feature Selection Strategies Identified Novel Cardiac Gene Expression Signature for Heart Failure.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets