Dataset Information

A Deep Neural Network Model using Random Forest to Extract Feature Representation for Gene Expression Data Classification.

ABSTRACT: In predictive model development, gene expression data is associated with the unique challenge that the number of samples (n) is much smaller than the amount of features (p). This "n???p" property has prevented classification of gene expression data from deep learning techniques, which have been proved powerful under "n?>?p" scenarios in other application fields, such as image classification. Further, the sparsity of effective features with unknown correlation structures in gene expression profiles brings more challenges for classification tasks. To tackle these problems, we propose a newly developed classifier named Forest Deep Neural Network (fDNN), to integrate the deep neural network architecture with a supervised forest feature detector. Using this built-in feature detector, the method is able to learn sparse feature representations and feed the representations into a neural network to mitigate the overfitting problem. Simulation experiments and real data analyses using two RNA-seq expression datasets are conducted to evaluate fDNN's capability. The method is demonstrated a useful addition to current predictive models with better classification performance and more meaningful selected features compared to ordinary random forests and deep neural networks.

SUBMITTER: Kong Y

PROVIDER: S-EPMC6220289 | biostudies-literature | 2018 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Deep Neural Network Model using Random Forest to Extract Feature Representation for Gene Expression Data Classification.

Kong Yunchuan Y Yu Tianwei T

Scientific reports 20181107 1

In predictive model development, gene expression data is associated with the unique challenge that the number of samples (n) is much smaller than the amount of features (p). This "n ≪ p" property has prevented classification of gene expression data from deep learning techniques, which have been proved powerful under "n > p" scenarios in other application fields, such as image classification. Further, the sparsity of effective features with unknown correlation structures in gene expression profil ...[more]

PMID: 30405137

Similar Datasets

Project description:Development of machine learning solutions for prediction of functional and clinical significance of cancer driver genes and mutations are paramount in modern biomedical research and have gained a significant momentum in a recent decade. In this work, we integrate different machine learning approaches, including tree based methods, random forest and gradient boosted tree (GBT) classifiers along with deep convolutional neural networks (CNN) for prediction of cancer driver mutations in the genomic datasets. The feasibility of CNN in using raw nucleotide sequences for classification of cancer driver mutations was initially explored by employing label encoding, one hot encoding, and embedding to preprocess the DNA information. These classifiers were benchmarked against their tree-based alternatives in order to evaluate the performance on a relative scale. We then integrated DNA-based scores generated by CNN with various categories of conservational, evolutionary and functional features into a generalized random forest classifier. The results of this study have demonstrated that CNN can learn high level features from genomic information that are complementary to the ensemble-based predictors often employed for classification of cancer mutations. By combining deep learning-generated score with only two main ensemble-based functional features, we can achieve a superior performance of various machine learning classifiers. Our findings have also suggested that synergy of nucleotide-based deep learning scores and integrated metrics derived from protein sequence conservation scores can allow for robust classification of cancer driver mutations with a limited number of highly informative features. Machine learning predictions are leveraged in molecular simulations, protein stability, and network-based analysis of cancer mutations in the protein kinase genes to obtain insights about molecular signatures of driver mutations and enhance the interpretability of cancer-specific classification models.

Project description:Feedforward deep neural networks (DNNs), artificial neural networks with multiple hidden layers, have recently demonstrated a record-breaking performance in multiple areas of applications in computer vision and speech processing. Following the success, DNNs have been applied to neuroimaging modalities including functional/structural magnetic resonance imaging (MRI) and positron-emission tomography data. However, no study has explicitly applied DNNs to 3D whole-brain fMRI volumes and thereby extracted hidden volumetric representations of fMRI that are discriminative for a task performed as the fMRI volume was acquired. Our study applied fully connected feedforward DNN to fMRI volumes collected in four sensorimotor tasks (i.e., left-hand clenching, right-hand clenching, auditory attention, and visual stimulus) undertaken by 12 healthy participants. Using a leave-one-subject-out cross-validation scheme, a restricted Boltzmann machine-based deep belief network was pretrained and used to initialize weights of the DNN. The pretrained DNN was fine-tuned while systematically controlling weight-sparsity levels across hidden layers. Optimal weight-sparsity levels were determined from a minimum validation error rate of fMRI volume classification. Minimum error rates (mean±standard deviation; %) of 6.9 (±3.8) were obtained from the three-layer DNN with the sparsest condition of weights across the three hidden layers. These error rates were even lower than the error rates from the single-layer network (9.4±4.6) and the two-layer network (7.4±4.1). The estimated DNN weights showed spatial patterns that are remarkably task-specific, particularly in the higher layers. The output values of the third hidden layer represented distinct patterns/codes of the 3D whole-brain fMRI volume and encoded the information of the tasks as evaluated from representational similarity analysis. Our reported findings show the ability of the DNN to classify a single fMRI volume based on the extraction of hidden representations of fMRI volumes associated with tasks across multiple hidden layers. Our study may be beneficial to the automatic classification/diagnosis of neuropsychiatric and neurological diseases and prediction of disease severity and recovery in (pre-) clinical settings using fMRI volumes without requiring an estimation of activation patterns or ad hoc statistical evaluation.

Project description:Background:Objective assessments of Parkinson's disease (PD) patients' motor state using motion capture techniques are still rarely used in clinical practice, even though they may improve clinical management. One major obstacle relates to the large dimensionality of motor abnormalities in PD. We aimed to extract global motor performance measures covering different everyday motor tasks, as a function of a clinical intervention, i.e., deep brain stimulation (DBS) of the subthalamic nucleus. Methods:We followed a data-driven, machine-learning approach and propose performance measures that employ Random Forests with probability distributions. We applied this method to 14 PD patients with DBS switched-off or -on, and 26 healthy control subjects performing the Timed Up and Go Test (TUG), the Functional Reach Test (FRT), a hand coordination task, walking 10-m straight, and a 90° curve. Results:For each motor task, a Random Forest identified a specific set of metrics that optimally separated PD off DBS from healthy subjects. We noted the highest accuracy (94.6%) for standing up. This corresponded to a sensitivity of 91.5% to detect a PD patient off DBS, and a specificity of 97.2% representing the rate of correctly identified healthy subjects. We then calculated performance measures based on these sets of metrics and applied those results to characterize symptom severity in different motor tasks. Task-specific symptom severity measures correlated significantly with each other and with the Unified Parkinson's Disease Rating Scale (UPDRS, part III, correlation of r2?=?0.79). Agreement rates between different measures ranged from 79.8 to 89.3%. Conclusion:The close correlation of PD patients' various motor abnormalities quantified by different, task-specific severity measures suggests that these abnormalities are only facets of the underlying one-dimensional severity of motor deficits. The identification and characterization of this underlying motor deficit may help to optimize therapeutic interventions, e.g., to "automatically" adapt DBS settings in PD patients.

Dataset Information

A Deep Neural Network Model using Random Forest to Extract Feature Representation for Gene Expression Data Classification.

Publications

A Deep Neural Network Model using Random Forest to Extract Feature Representation for Gene Expression Data Classification.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets