Putative null distributions corresponding to tests of differential expression in the Golden Spike dataset are intensity dependent
2007

Re-analysis of the Golden Spike Dataset

Sample size: 152 publication 10 minutes Evidence: moderate

Author Information

Author(s): Gaile Daniel P, Miecznikowski Jeffrey C

Primary Institution: University at Buffalo

Hypothesis

Are the null p-values for tests of differential expression in the Golden Spike dataset uniformly distributed?

Conclusion

The null p-values in the Golden Spike dataset are non-uniform and should not be used as a reference for evaluating false discovery rate controlling methodologies.

Supporting Evidence

  • The original analysis raised questions about the performance of statistical methods for controlling false discovery rates.
  • Previous studies indicated that the null p-values were not uniformly distributed.
  • The study demonstrated that the distributions of p-values vary with signal intensity.

Takeaway

The study looked at a dataset used to test gene expression and found that the results were not reliable because the data was not processed correctly.

Methodology

The study involved re-analyzing the Golden Spike dataset using various statistical methods to assess the distribution of null p-values.

Potential Biases

Potential biases due to the experimental design and the choice of normalization methods.

Limitations

The analysis is limited to the Golden Spike dataset and may not generalize to other datasets.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2164-8-105

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication