CMA – a comprehensive Bioconductor package for supervised classification with high dimensional data
2008

CMA: A Bioconductor Package for Classifying High-Dimensional Data

Sample size: 65 publication 10 minutes Evidence: moderate

Author Information

Author(s): Martin Slawski, Alfred-Ludwig Boulesteix, Daumer M

Primary Institution: Sylvia Lawry Centre for Multiple Sclerosis Research, Munich, Germany

Hypothesis

Can a new Bioconductor package improve classification accuracy for high-dimensional data?

Conclusion

CMA is a user-friendly package that simplifies classifier construction and evaluation for high-dimensional data.

Supporting Evidence

  • CMA provides an overview of the unbiased accuracy of top-performing classifiers.
  • The package includes over twenty different classifiers and various evaluation schemes.
  • CMA automates hyperparameter tuning to prevent bias in model selection.
  • Users can visualize results easily with built-in functions.

Takeaway

CMA helps scientists easily classify data with many variables, like genes, without needing to be experts in statistics.

Methodology

The CMA package automates variable selection, parameter tuning, classifier construction, and evaluation using various methods.

Potential Biases

Inexperienced users may introduce errors by not understanding the tuning parameters or data formats.

Limitations

The package may not perform well with very small sample sizes or when users do not follow good practices in model selection.

Participant Demographics

The study involved 65 samples from four tumor classes.

Digital Object Identifier (DOI)

10.1186/1471-2105-9-439

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication