GUEST: an R package for handling estimation of graphical structure and multiclassification for error-prone gene expression data
2024

GUEST: An R Package for Gene Expression Data Analysis

Sample size: 83 publication Evidence: high

Author Information

Author(s): Chen Li-Pang, Tsao Hui-Shan

Primary Institution: National Chengchi University

Hypothesis

Can we develop a reliable R package to estimate graphical structures and classify error-prone gene expression data?

Conclusion

The GUEST R package effectively addresses measurement errors in high-dimensional gene expression data and improves classification accuracy.

Supporting Evidence

  • The GUEST package outperformed existing methods in classification accuracy.
  • Measurement error correction was shown to significantly impact results.
  • The package is designed to handle ultrahigh-dimensional data effectively.

Takeaway

This study created a tool called GUEST that helps scientists understand gene data better, even when the data is messy.

Methodology

The GUEST package uses boosting algorithms to estimate network structures and improve classification accuracy in gene expression data.

Potential Biases

Potential biases may arise from incorrect assumptions about measurement error.

Limitations

The package may not be valid for all types of data and relies on user-specified parameters for measurement error.

Participant Demographics

The study analyzed gene expression data from 83 subjects with four types of childhood tumors.

Digital Object Identifier (DOI)

10.1093/bioinformatics/btae731

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication