Dataset Information

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

ABSTRACT: Time series gene expression data is widely used to study different dynamic biological processes. Although gene expression datasets share many of the characteristics of time series data from other domains, most of the analyses in this field do not fully leverage the time-ordered nature of the data and focus on clustering the genes based on their expression values. Other domains, such as financial stock and weather prediction, utilize time series data for forecasting purposes. Moreover, many studies have been conducted to classify generic time series data based on trend, seasonality, and other patterns. Therefore, an assessment of these approaches on gene expression data would be of great interest to evaluate their adequacy in this domain. Here, we perform a comprehensive evaluation of different traditional unsupervised and supervised machine learning approaches as well as deep learning based techniques for time series gene expression classification and forecasting on five real datasets. In addition, we propose deep learning based methods for both classification and forecasting, and compare their performances with the state-of-the-art methods. We find that deep learning based methods generally outperform traditional approaches for time series classification. Experiments also suggest that supervised classification on gene expression is more effective than clustering when labels are available. In time series gene expression forecasting, we observe that an autoregressive statistical approach has the best performance for short term forecasting, whereas deep learning based methods are better suited for long term forecasting.

SUBMITTER: Tripto NI

PROVIDER: S-EPMC7647064 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Tripto Nafis Irtiza NI Kabir Mohimenul M Bayzid Md Shamsuzzoha MS Rahman Atif A

PloS one 20201106 11

Time series gene expression data is widely used to study different dynamic biological processes. Although gene expression datasets share many of the characteristics of time series data from other domains, most of the analyses in this field do not fully leverage the time-ordered nature of the data and focus on clustering the genes based on their expression values. Other domains, such as financial stock and weather prediction, utilize time series data for forecasting purposes. Moreover, many studi ...[more]

PMID: 33156855

Similar Datasets

Project description:ImportanceForecasting the volume of hospital discharges has important implications for resource allocation and represents an opportunity to improve patient safety at periods of elevated risk.ObjectiveTo determine the performance of a new time-series machine learning method for forecasting hospital discharge volume compared with simpler methods.DesignA retrospective cohort study of daily hospital discharge volumes at 2 large, New England academic medical centers between January 1, 2005, and December 31, 2014 (hospital 1), or January 1, 2005, and December 31, 2010 (hospital 2), comparing time-series forecasting methods for prediction was performed. Data analysis was conducted from February 28, 2017, to August 30, 2018. Group-level data for all discharges from inpatient units were included. In addition to conventional methods, a technique originally developed for allocating data center resources, and comparison strategies for incorporating prior data and frequency of model updates, was conducted to identify the model application that optimized forecast accuracy.Main outcomes and measuresModel calibration as measured by R2 and, secondarily, number of days with errors greater than 1 SD of daily volume.ResultsDuring the forecasted year, hospital 1 had 54?411 discharges (daily mean, 149) and hospital 2 had 47?456 discharges (daily mean, 130). The machine learning method was well calibrated at both sites (R2, 0.843 and 0.726, respectively) and made errors greater than 1 SD of daily volume on only 13 and 22 days, respectively, of the forecast year at the 2 sites. Last-value-carried-forward models performed somewhat less well (calibration R2, 0.781 and 0.596, respectively) with 13 and 46 errors of 1 SD or greater, respectively. More frequent retraining and training sets of longer than 1 year had minimal effects on the machine learning method's performance.Conclusions and relevanceVolume of hospital discharges can perhaps be reliably forecasted using simple carry-forward models as well as methods drawn from machine learning. The benefit of the latter does not appear to be dependent on extensive training data and may enable forecasts up to 1 year in advance with superior absolute accuracy to carry-forward models.

Project description:BackgroundComparing metabolic profiles under different biological perturbations has become a powerful approach to investigating the functioning of cells. The profiles can be taken as single snapshots of a system, but more information is gained if they are measured longitudinally over time. The results are short time series consisting of relatively sparse data that cannot be analyzed effectively with standard time series techniques, such as autocorrelation and frequency domain methods. In this work, we study longitudinal time series profiles of glucose consumption in the yeast Saccharomyces cerevisiae under different temperatures and preconditioning regimens, which we obtained with methods of in vivo nuclear magnetic resonance (NMR) spectroscopy. For the statistical analysis we first fit several nonlinear mixed effect regression models to the longitudinal profiles and then used an ANOVA likelihood ratio method in order to test for significant differences between the profiles.ResultsThe proposed methods are capable of distinguishing metabolic time trends resulting from different treatments and associate significance levels to these differences. Among several nonlinear mixed-effects regression models tested, a three-parameter logistic function represents the data with highest accuracy. ANOVA and likelihood ratio tests suggest that there are significant differences between the glucose consumption rate profiles for cells that had been--or had not been--preconditioned by heat during growth. Furthermore, pair-wise t-tests reveal significant differences in the longitudinal profiles for glucose consumption rates between optimal conditions and heat stress, optimal and recovery conditions, and heat stress and recovery conditions (p-values <0.0001).ConclusionWe have developed a nonlinear mixed effects model that is appropriate for the analysis of sparse metabolic and physiological time profiles. The model permits sound statistical inference procedures, based on ANOVA likelihood ratio tests, for testing the significance of differences between short time course data under different biological perturbations.

Project description:In this paper, we introduce a novel model selection approach to time series forecasting. For linear stationary processes, such as AR processes, the direction of time is independent of the model parameters. By combining theoretical principles of time-reversibility in time series with conventional modeling approaches such as information criteria, we construct a criterion that employs the backwards prediction (backcast) as a proxy for the forecast. Hereby, we aim to adopt a theoretically grounded, data-driven approach to model selection. The novel criterion is named the backwards validated information criterion (BVIC). The BVIC identifies suitable models by trading off a measure of goodness-of-fit and a models ability to predict backwards. We test the performance of the BVIC by conducting experiments on synthetic and real data. In each experiment, the BVIC is examined in contrast to conventionally employed criteria. Our experimental results suggest that the BVIC has comparable performance as conventional information criteria. Specifically, in most of the experiments performed, we did not find statistically significant differences between the forecast error of the BVIC under certain parameterizations and that of the different information criteria. Nonetheless, it is worth emphasizing that the BVIC guarantees are established by design where the model order penalization term depends on strong mathematical properties of time-reversible time series forecasting properties and a finite data assessment. In particular, the penalization term is replaced by a weighted trade-off between functional dimensions pertaining to forecasting.That said, we observed that the BVIC recovered more accurately the real order of the underlying process than the other criteria, which rely on a static penalization of the model order. Lastly, leveraging the latter property we perform the assessment of the order model (or, memory) of time series pertaining to epileptic seizures recorded using electrocorticographic data. Our results provide converging evidence that the order of the model increases during the epileptic events.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets