Robust Method for Filling in Missing Values in Microarray Data
Author Information
Author(s): Yoon Dankyu, Lee Eun-Kyung, Park Taesung
Primary Institution: Seoul National University
Hypothesis
Can a robust least squares estimation with principal components (RLSP) method improve the accuracy of missing value imputation in microarray data?
Conclusion
The RLSP method is more robust and accurate than commonly used methods for imputing missing values in microarray data.
Supporting Evidence
- The RLSP method outperformed the kNNimpute and LLSimpute methods in terms of accuracy.
- RLSP showed competitive results with Bayesian principal component analysis (BPCA).
- Using principal components improved the robustness of the imputation method.
Takeaway
This study created a new way to guess missing information in gene data, which works better than older methods.
Methodology
The study developed a robust least squares estimation method using principal components and quantile regression to impute missing values.
Potential Biases
The method assumes that missing data is missing completely at random, which may not hold true in all cases.
Limitations
The performance of the RLSP method may depend on the choice of the number of principal components.
Participant Demographics
The study used gene expression data from various datasets, including yeast.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website