Shrinkage regression-based methods for microarray missing value imputation

Wang, Hsiuying; Chiu, Chia Chun; Wu, Yi Ching; Wu, Wei

doi:10.1186/1752-0509-7-s6-s11

Cited by 8 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With regard to invalid measurements, the calculated rate of 3.79% invalid measurements on the microarray was already satisfactory. Microarray data have been described to contain usually more than 5% missing values [51] or to vary between 0.8% and 10% [52]. The percentage of invalid measurements would even be lower if antigen concentrations that were found to be redundant could be excluded from the microarray layout.…”

Section: Discussionmentioning

confidence: 99%

Development of a miniaturized protein microarray as a new serological IgG screening test for zoonotic agents and production diseases in pigs

Loreck¹,

Mitrenga²,

Meemken³

et al. 2019

PLoS ONE

View full text Add to dashboard Cite

In order to monitor the occurrence of zoonotic agents in pig herds as well as to improve herd health management, the development of new cost-effective diagnostic methods for pigs is necessary. In this study, a protein microarray-based assay for the simultaneous detection of immunoglobulin G (IgG) antibodies against different zoonotic agents and pathogens causing production diseases in pigs was developed. Therefore, antigens of ten different important swine pathogens ( Toxoplasma gondii , Yersinia enterocolitica , Salmonella spp., Trichinella spp., Mycobacterium avium , Hepatitis E virus, Mycoplasma hyopneumoniae , Actinobacillus pleuropneumoniae , the porcine reproductive and respiratory syndrome virus, Influenza A virus) were spotted and covalently immobilized as ‘antigen-spots’ on microarray chips in order to test pig serum for the occurrence of antibodies. Pig serum was sampled at three German abattoirs and ELISA tests for the different pathogens were conducted with the purpose of creating a panel of reference samples for microarray analysis. To evaluate the accuracy of the antigens on the microarray, receiver operating characteristic (ROC) curve analysis using the ELISA test results as reference was performed for the different antigens. High area under curve values were achieved for the antigens of two zoonotic agents: Toxoplasma gondii (0.91), Yersinia enterocolitica (0.97) and for three production diseases: Actinobacillus pleuropneumoniae (0.77), Mycoplasma hyopneumoniae (0.94) and the porcine reproductive and respiratory syndrome virus (0.87). With the help of the newly developed microarray assay, collecting data on the occurrence of antibodies against zoonotic agents and production diseases in pig herds could be minimized to one measurement, resulting in an efficient screening test.

show abstract

Section: Discussionmentioning

confidence: 99%

Development of a miniaturized protein microarray as a new serological IgG screening test for zoonotic agents and production diseases in pigs

Loreck¹,

Mitrenga²,

Meemken³

et al. 2019

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Zhang et al [31] proposed the sequentially local least squares-based imputation method (SLLSimpute). Wang et al [32] proposed the shrinkage regression-based method (ShrinkageLLS) that first selected similar genes by Pearson correlation coefficients and then adjusted the regression coefficients with a shrinkage estimation operator. Compared with KNNimpute and its variations, regression-based methods generally obtain better results, especially when a larger number of neighbor genes are used [33].…”

Section: Related Workmentioning

confidence: 99%

Regularized Sparse Modelling for Microarray Missing Value Estimation

Wang

Yang

2021

IEEE Access

View full text Add to dashboard Cite

The existence of missing values in microarray data inevitably hinders downstream biological analyses that expect complete data as input, therefore how to effectively explore the underlying structure of data to accurately estimate missing entries remains crucial and meaningful. In this study, we formalize the problem under a regularized sparse framework and accordingly propose local learning-based imputation models to capture the relationships that are hidden in gene expression profiles towards better imputation. Specifically, in view of the simultaneous variable selection and grouping effect of the elastic net penalty, we present an elastic net regularized local least squares-based imputation method to estimate the missing entries of a target gene with its neighbors. Besides, we investigate different similarity filtering metrics to select neighbor genes and develop another four imputation methods under the framework. Furthermore, the proposed methods process the target genes in ascending order of their associated missing rates. Finally, extensive comparative experiments against other eight commonly-used methods are conducted on multiple microarray datasets having varying missing rates. Results indicate the power of sparse regularization techniques and the superiority of elastic net over its competitors in terms of statistical analysis metrics.

show abstract

“…The percentage of invalid measurements in the laboratories correspond to the percentages of missing values for microarray data that have been described to be usually higher than 5% [35] or to vary between 0.8 and 10% [36]. The higher percentage of invalid measurements for meat juice might be attributed to protein and fat residues in the sample material.…”

Section: Number Of Positive and Negative Samples In Roc Analysismentioning

confidence: 99%

Use of meat juice and blood serum with a miniaturised protein microarray assay to develop a multi-parameter IgG screening test with high sample throughput potential for slaughtering pigs

et al. 2020

View full text Add to dashboard Cite

Background: Serological screening of pig herds at the abattoir is considered a potential tool to improve meat inspection procedures and herd health management. Therefore, we previously reported the feasibility of a miniaturised protein microarray as a new serological IgG screening test for zoonotic agents and production diseases in pigs. The present study investigates whether the protein microarray-based assay is applicable for high sample throughput using either blood serum or meat juice.Material and methods: Microarrays with 12 different antigens were produced by Abbott (formerly Alere Technologies GmbH) Jena, Germany in a previously offered 'ArrayTube' platform and in an 'ArrayStrip' platform for large-scale use. A test protocol for the use of meat juice on both microarray platforms was developed. Agreement between serum and meat juice was analysed with 88 paired samples from three German abattoirs. Serum was diluted 1:50 and meat juice 1:2. ELISA results for all tested antigens from a preceding study were used as reference test to perform Receiver Operating Characteristic analysis for both test specimens on both microarray platforms.Results: High area under curve values (AUC > 0.7) were calculated for the analysis of T. gondii (0.87), Y. enterocolitica (0.97), Mycoplasma hyopneumoniae (0.84) and Actinobacillus pleuropneumoniae (0.71) with serum as the test specimen and for T. gondii (0.99), Y. enterocolitica (0.94), PRRSV (0.88), A. pleuropneumoniae (0.78) and Salmonella spp. (0.72) with meat juice as the test specimen on the ArrayStrip platform. Cohens kappa values of 0.92 for T. gondii and 0.82 for Y. enterocolitica were obtained for the comparison between serum and meat juice. When applying the new method in two further laboratories, kappa values between 0.63 and 0.94 were achieved between the laboratories for these two pathogens.(Continued on next page) Conclusion: Further development of a miniaturised pig-specific IgG protein microarray assay showed that meat juice can be used on microarray platforms. Two out of twelve tested antigens (T. gondii, Y. enterocolitica) showed high test accuracy on the ArrayTube and the ArrayStrip platform with both sample materials.

show abstract

Shrinkage regression-based methods for microarray missing value imputation

Cited by 8 publications

References 29 publications

Development of a miniaturized protein microarray as a new serological IgG screening test for zoonotic agents and production diseases in pigs

Development of a miniaturized protein microarray as a new serological IgG screening test for zoonotic agents and production diseases in pigs

Regularized Sparse Modelling for Microarray Missing Value Estimation

Use of meat juice and blood serum with a miniaturised protein microarray assay to develop a multi-parameter IgG screening test with high sample throughput potential for slaughtering pigs

Contact Info

Product

Resources

About