Statistical validation of megavariate effects in ASCA
2007

Validating ASCA for Metabolomics Data

Sample size: 3 publication 10 minutes Evidence: high

Author Information

Author(s): Daniel J Vis, Westerhuis Johan A, Smilde Age K, van der Greef Jan

Primary Institution: University of Amsterdam

Hypothesis

Can a permutation approach effectively validate megavariate effects in ASCA?

Conclusion

The proposed procedure provides approximate p-values for testing effects in metabolomics data sets, enabling model validation.

Supporting Evidence

  • The permutation approach successfully validated significant effects in real-life metabolomics data.
  • Histological screening of treated rats' liver confirmed the findings of the study.
  • The method allows for model validation without the common assumptions of normality or equal variance.

Takeaway

This study shows a way to check if the results from metabolomics experiments are real or just random chance by using a special method called permutation.

Methodology

A permutation approach was used to validate megavariate effects observed with ASCA by permuting class labels and calculating a distribution of no-effect.

Potential Biases

Potential biases may arise from small sample sizes and the assumptions made in the statistical tests.

Limitations

The study assumes normality and equal variances, which may not hold true in biological variations.

Participant Demographics

Rats were used in the study, specifically those dosed with bromobenzene.

Statistical Information

P-Value

0.00012

Statistical Significance

p ≤ 0.0001

Digital Object Identifier (DOI)

10.1186/1471-2105-8-322

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication