A hybrid imputation approach for microarray missing value estimation

Abstract

BackgroundMissing data is an inevitable phenomenon in gene expression microarray experiments due to instrument failure or human error. It has a negative impact on performance of downstream analysis. Technically, most existing approaches suffer from this prevalent problem. Imputation is one of the frequently used methods for processing missing data. Actually many developments have been achieved in the research on estimating missing values. The challenging task is how to improve imputation accuracy for data with a large missing rate.MethodsIn this paper, induced by the thought of collaborative training, we propose a novel hybrid imputation method, called Recursive Mutual Imputation (RMI). Specifically, RMI exploits global correlation information and local structure in the data, captured by two popular methods, Bayesian Principal Component Analysis (BPCA) and Local Least Squares (LLS), respectively. Mutual strategy is implemented by sharing the estimated data sequences at each recursive process. Meanwhile, we consider the imputation sequence based on the number of missing entries in the target gene. Furthermore, a weight based integrated method is utilized in the final assembling step.ResultsWe evaluate RMI with three state-of-art algorithms (BPCA, LLS, Iterated Local Least Squares imputation (ItrLLS)) on four publicly available microarray datasets. Experimental results clearly demonstrate that RMI significantly outperforms comparative methods in terms of Normalized Root Mean Square Error (NRMSE), especially for datasets with large missing rates and less complete genes.ConclusionsIt is noted that our proposed hybrid imputation approach incorporates both global and local information of microarray genes, which achieves lower NRMSE values against to any single approach only. Besides, this study highlights the need for considering the imputing sequence of missing entries for imputation methods.

Efficient technique of microarray missing data imputation using clustering and weighted nearest neighbour

“…In a hybrid approach Recursive Mutation Imputation, global approach(BPCA), and local approach(LLS) are combined for making imputation 17 . Missing values in the target gene decide the order of the imputation process.…”

Regularized Sparse Modelling for Microarray Missing Value Estimation

“…For example, Jornsten et al [19] proposed to combine global and local learning-based methods (LinCmb) to estimate the missing values. He et al [23] combined the predictions from local least squares-based imputation method and Bayesian principal component analysis imputation method under the ensemble framework and inferred the missing values from the weighted outputs of its components. Meng et al [24] proposed to combine Bayesian principal component analysis and bicluster analysis, where the latter filtered the selection of neighbors and the former was applied on the biclusters to explore the local data structure.…”

R Package imputeTestbench to Compare Imputation Methods for Univariate Time Series

“…Table 1 reproduces the results, where the rows show root-mean squared error (RMSE) between observed and predicted data for each imputation method after removing and predicting 10% and 80% of the complete dataset. Additional studies have used a similar workflow to compare the performance of imputation methods (Jörnsten et al, 2007;Li et al, 2015;Nguyen et al, 2013;Ran et al, 2015;Tak et al, 2016).…”