Identification of human circadian genes based on time course gene expression profiles by using a deep learning method

Abstract

Circadian genes express periodically in an approximate 24-h period and the identification and study of these genes can provide deep understanding of the circadian control which plays significant roles in human health. Although many circadian gene identification algorithms have been developed, large numbers of false positives and low coverage are still major problems in this field. In this study we constructed a novel computational framework for circadian gene identification using deep neural networks (DNN) - a deep learning algorithm which can represent the raw form of data patterns without imposing assumptions on the expression distribution. Firstly, we transformed time-course gene expression data into categorical-state data to denote the changing trend of gene expression. Two distinct expression patterns emerged after clustering of the state data for circadian genes from our manually created learning dataset. DNN was then applied to discriminate the aperiodic genes and the two subtypes of periodic genes. In order to assess the performance of DNN, four commonly used machine learning methods including k-nearest neighbors, logistic regression, naïve Bayes, and support vector machines were used for comparison. The results show that the DNN model achieves the best balanced precision and recall. Next, we conducted large scale circadian gene detection using the trained DNN model for the remaining transcription profiles. Comparing with JTK_CYCLE and a study performed by Möller-Levet et al. (doi: https://doi.org/10.1073/pnas.1217154110), we identified 1132 novel periodic genes. Through the functional analysis of these novel circadian genes, we found that the GTPase superfamily exhibits distinct circadian expression patterns and may provide a molecular switch of circadian control of the functioning of the immune system in human blood. Our study provides novel insights into both the circadian gene identification field and the study of complex circadian-driven biological control. This article is part of a Special Issue entitled: Accelerating Precision Medicine through Genetic and Genomic Big Data Analysis edited by Yudong Cai & Tao Huang.

An Optimal Time for Treatment—Predicting Circadian Time by Machine Learning and Mathematical Modelling

“…It considers signal to noise (S/N) levels and uses harmonic regression to model the changes over time. In recent years, deep neural networks (DNNs) for the detection of circadian rhythmicity have been used in several studies [ 148 , 155 ]. BIO_CYCLE, which is also based on DNNs, works firstly by training an algorithm with periodic data.…”

Evaluation of classification and forecasting methods on time series gene expression data

“…In this study, we are performing classification on labelled datasets where each gene belongs to a specific class and consider each expression to consist of the same number of time points. The methods for classification evaluated in this study are described below where the CNN and LSTM architectures are proposed in this paper, and SVM, One-Class SVM, DNN [30] and DeepTrust [31] are from the literature.…”

“…Deep Neural Network (DNN). DNN [30] first transforms the raw expression data to categorical state to discover the underlying distinct expression patterns of the generated circadian genes for validating the manually labelled dataset. However, labels of each gene expression are already available for all of our datasets.…”

“…Therefore, we directly implemented the DNN based on the time series expression data. We employed the same architecture for DNN as [30] with ten hidden layers of 100 nodes. For activation function, the rectified linear unit (ReLU) was used, and the learning rate was set at 0.0001 using an Adam gradient-based optimizer [44].…”

“…Orsenigo and Vercellis suggested a temporal extension of L1-norm Support Vector Machines (SVM) that uses dynamic time warping distance for measuring time series similarity [29]. Cui et al presented a classification of human circadian genes based on time-course gene expression profiles using deep neural networks (DNN) [30]. They transformed time series into categorical-state data to denote the changing trend of gene expression and then applied DNN to discriminate between aperiodic and two subclasses of periodic genes.…”