ProJect: a powerful mixed-model missing value imputation method.

Kong, W; Wong, BJH; Hui, HWH; Lim, KP; Wang, Y; Wong, L; Goh, WWB

ProJect: a powerful mixed-model missing value imputation method.

Kong, W Wong, BJH Hui, HWH Lim, KP Wang, Y Wong, L Goh, WWB

Permalink

Publisher:: Oxford University Press (OUP)
Publication Type:: Journal Article
Citation:: Brief Bioinform, 2023, 24, (4), pp. bbad233
Issue Date:: 2023-07-20

Closed Access

	Filename	Description	Size
	ProJect a powerful mixed-model missing value imputation method.pdf		2.19 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Kong, W
dc.contributor.author	Wong, BJH
dc.contributor.author	Hui, HWH
dc.contributor.author	Lim, KP
dc.contributor.author	Wang, Y
dc.contributor.author	Wong, L
dc.contributor.author	Goh, WWB
dc.date.accessioned	2024-01-16T04:25:43Z
dc.date.available	2023-06-05
dc.date.available	2024-01-16T04:25:43Z
dc.date.issued	2023-07-20
dc.identifier.citation	Brief Bioinform, 2023, 24, (4), pp. bbad233
dc.identifier.issn	1467-5463
dc.identifier.issn	1477-4054
dc.identifier.uri	http://hdl.handle.net/10453/174576
dc.description.abstract	Missing values (MVs) can adversely impact data analysis and machine-learning model development. We propose a novel mixed-model method for missing value imputation (MVI). This method, ProJect (short for Protein inJection), is a powerful and meaningful improvement over existing MVI methods such as Bayesian principal component analysis (PCA), probabilistic PCA, local least squares and quantile regression imputation of left-censored data. We rigorously tested ProJect on various high-throughput data types, including genomics and mass spectrometry (MS)-based proteomics. Specifically, we utilized renal cancer (RC) data acquired using DIA-SWATH, ovarian cancer (OC) data acquired using DIA-MS, bladder (BladderBatch) and glioblastoma (GBM) microarray gene expression dataset. Our results demonstrate that ProJect consistently performs better than other referenced MVI methods. It achieves the lowest normalized root mean square error (on average, scoring 45.92% less error in RC_C, 27.37% in RC_full, 29.22% in OC, 23.65% in BladderBatch and 20.20% in GBM relative to the closest competing method) and the Procrustes sum of squared error (Procrustes SS) (exhibits 79.71% less error in RC_C, 38.36% in RC full, 18.13% in OC, 74.74% in BladderBatch and 30.79% in GBM compared to the next best method). ProJect also leads with the highest correlation coefficient among all types of MV combinations (0.64% higher in RC_C, 0.24% in RC full, 0.55% in OC, 0.39% in BladderBatch and 0.27% in GBM versus the second-best performing method). ProJect's key strength is its ability to handle different types of MVs commonly found in real-world data. Unlike most MVI methods that are designed to handle only one type of MV, ProJect employs a decision-making algorithm that first determines if an MV is missing at random or missing not at random. It then employs targeted imputation strategies for each MV type, resulting in more accurate and reliable imputation outcomes. An R implementation of ProJect is available at https://github.com/miaomiao6606/ProJect.
dc.format	Print
dc.language	eng
dc.publisher	Oxford University Press (OUP)
dc.relation.ispartof	Brief Bioinform
dc.relation.isbasedon	10.1093/bib/bbad233
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0601 Biochemistry and Cell Biology, 0802 Computation Theory and Mathematics, 0899 Other Information and Computing Sciences
dc.subject.classification	Bioinformatics
dc.subject.classification	3101 Biochemistry and cell biology
dc.subject.classification	3102 Bioinformatics and computational biology
dc.subject.classification	3105 Genetics
dc.subject.mesh	Bayes Theorem
dc.subject.mesh	Oligonucleotide Array Sequence Analysis
dc.subject.mesh	Algorithms
dc.subject.mesh	Genomics
dc.subject.mesh	Mass Spectrometry
dc.subject.mesh	Oligonucleotide Array Sequence Analysis
dc.subject.mesh	Bayes Theorem
dc.subject.mesh	Genomics
dc.subject.mesh	Algorithms
dc.subject.mesh	Mass Spectrometry
dc.subject.mesh	Bayes Theorem
dc.subject.mesh	Oligonucleotide Array Sequence Analysis
dc.subject.mesh	Algorithms
dc.subject.mesh	Genomics
dc.subject.mesh	Mass Spectrometry
dc.title	ProJect: a powerful mixed-model missing value imputation method.
dc.type	Journal Article
utslib.citation.volume	24
utslib.location.activity	England
utslib.for	0601 Biochemistry and Cell Biology
utslib.for	0802 Computation Theory and Mathematics
utslib.for	0899 Other Information and Computing Sciences
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
utslib.copyright.status	closed_access	*
dc.date.updated	2024-01-16T04:25:42Z
pubs.issue	4
pubs.publication-status	Published
pubs.volume	24
utslib.citation.issue	4

Abstract:

Missing values (MVs) can adversely impact data analysis and machine-learning model development. We propose a novel mixed-model method for missing value imputation (MVI). This method, ProJect (short for Protein inJection), is a powerful and meaningful improvement over existing MVI methods such as Bayesian principal component analysis (PCA), probabilistic PCA, local least squares and quantile regression imputation of left-censored data. We rigorously tested ProJect on various high-throughput data types, including genomics and mass spectrometry (MS)-based proteomics. Specifically, we utilized renal cancer (RC) data acquired using DIA-SWATH, ovarian cancer (OC) data acquired using DIA-MS, bladder (BladderBatch) and glioblastoma (GBM) microarray gene expression dataset. Our results demonstrate that ProJect consistently performs better than other referenced MVI methods. It achieves the lowest normalized root mean square error (on average, scoring 45.92% less error in RC_C, 27.37% in RC_full, 29.22% in OC, 23.65% in BladderBatch and 20.20% in GBM relative to the closest competing method) and the Procrustes sum of squared error (Procrustes SS) (exhibits 79.71% less error in RC_C, 38.36% in RC full, 18.13% in OC, 74.74% in BladderBatch and 30.79% in GBM compared to the next best method). ProJect also leads with the highest correlation coefficient among all types of MV combinations (0.64% higher in RC_C, 0.24% in RC full, 0.55% in OC, 0.39% in BladderBatch and 0.27% in GBM versus the second-best performing method). ProJect's key strength is its ability to handle different types of MVs commonly found in real-world data. Unlike most MVI methods that are designed to handle only one type of MV, ProJect employs a decision-making algorithm that first determines if an MV is missing at random or missing not at random. It then employs targeted imputation strategies for each MV type, resulting in more accurate and reliable imputation outcomes. An R implementation of ProJect is available at https://github.com/miaomiao6606/ProJect.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/174576