Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Dealing with gene expression missing data.

L P Brás¹, J C Menezes

¹Centre for Chemical and Biological Engineering, Department of Chemical and Biological Engineering, IST, Technical University of Lisbon, Portugal. bsel@ist.utl.pt

Systems Biology

|September 21, 2006

Summary

This study compares microarray data imputation methods, finding that Local Least Squares (LLS) imputation excels for time-series data, while Partial Least Squares (PLS) methods are better for non-time-series data. Dataset properties significantly impact performance.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Semi-quantitative prediction of a multiple API solid dosage form with a combination of vibrational spectroscopy methods.

Journal of pharmaceutical and biomedical analysis·2016

Same author

Advanced qualification of pharmaceutical excipient suppliers by multiple analytics and multivariate analysis combined.

International journal of pharmaceutics·2015

Same author

A PAT-based qualification of pharmaceutical excipients produced by batch or continuous processing.

Journal of pharmaceutical and biomedical analysis·2015

Same author

A novel in-line NIR spectroscopy application for the monitoring of tablet film coating in an industrial scale process.

Talanta·2012

Same author

A multivariate approach for the statistical evaluation of near-infrared chemical images using Symmetry Parameter Image Analysis (SPIA).

European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V·2011

Same author

A new PAT/QbD approach for the determination of blend homogeneity: combination of on-line NIRS analysis with PC Scores Distance Analysis (PC-SDA).

European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V·2010

Area of Science:

Bioinformatics
Computational Biology
Genomics

Background:

Missing data is a common challenge in microarray analysis.
Accurate imputation is crucial for reliable downstream analysis and interpretation.

Purpose of the Study:

To compare the performance of various missing value imputation methods for microarray data.
To elucidate the influence of experimental factors on imputation accuracy.

Main Methods:

Weighted K-nearest neighbours imputation (KNNimpute)
Regression-based methods: Local Least Squares (LLSimpute) and Partial Least Squares (PLSimpute)
Bayesian Principal Component Analysis (BPCA)
Proposed iterative and modified methods (ILLSimpute, IPLSimpute, MPLSimpute, HPLSimpute, APLSimpute)

Related Experiment Videos

Main Results:

Dataset properties (experiment type, missing rate/pattern) influence method performance.
LLSimpute and iterative LLS (ILLSimpute) are optimal for time-series (TS) and mixed (MIX) experiments due to stronger data similarity.
PLS-based methods (MPLSimpute, IPLSimpute, APLSimpute) are preferable for non-time-series (NTS) data.

Conclusions:

No single imputation method is universally superior; method selection depends on data characteristics.
Understanding data structure is key to choosing the most effective imputation strategy for microarray analysis.