Empirical Bayes accommodation of batch-effects in microarray data using identical replicate reference samples: application to RNA expression profiling of blood from Duchenne muscular dystrophy patients
2008

Batch Effects in Microarray Data Correction for Duchenne Muscular Dystrophy

Sample size: 98 publication Evidence: high

Author Information

Author(s): Walker Wynn L, Liao Isaac H, Gilbert Donald L, Wong Brenda, Pollard Katherine S, McCulloch Charles E, Lit Lisa, Sharp Frank R

Primary Institution: University of California at Davis

Hypothesis

Can an empirical Bayes method effectively correct for batch effects in microarray data to allow comparisons of gene expression between biological groups from independent experiments?

Conclusion

The empirical Bayes method significantly improves the identification of differentially expressed genes in Duchenne Muscular Dystrophy patients compared to controls by correcting for batch effects.

Supporting Evidence

  • The empirical Bayes method identified 629 differentially expressed genes after batch adjustment.
  • Before adjustment, 527 genes were identified as differentially expressed, indicating significant batch effects.
  • Batch adjustment led to a reduction in false positives and improved detection of true differentially expressed genes.

Takeaway

This study shows a new way to fix errors in gene tests that happen when samples are processed at different times, helping doctors better understand diseases like Duchenne Muscular Dystrophy.

Methodology

The study used an empirical Bayes method to adjust for batch effects in microarray data from two experiments involving blood samples from Duchenne Muscular Dystrophy patients and healthy controls.

Potential Biases

Potential bias due to the limited number of reference samples used in each batch.

Limitations

The study primarily focuses on a specific disease and may not generalize to all types of microarray data or other diseases.

Participant Demographics

Participants included teenagers with Duchenne Muscular Dystrophy and healthy controls.

Statistical Information

P-Value

0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2164-9-494

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication