An assessment of false discovery rates and statistical significance in label-free quantitative proteomics with combined filters
2009

Assessing Statistical Significance in Proteomics Without Replicates

publication 10 minutes Evidence: moderate

Author Information

Author(s): Li Qingbo, Roxas Bryan AP

Primary Institution: University of Illinois at Chicago

Hypothesis

Can differentially expressed proteins be detected with statistical significance from unlabeled protein samples without replicates?

Conclusion

Statistical significance can be inferred without protein sample replicates in label-free quantitative proteomics.

Supporting Evidence

  • A combination of fold-change and statistical tests was not sufficient to control false discovery rates below 5%.
  • The introduction of minimum number of permuted significant pairings improved specificity.
  • The t-test was found to be more effective than the Wilcoxon ranksum test for this analysis.
  • ROC analysis confirmed the effectiveness of the proposed method for identifying differentially expressed proteins.

Takeaway

This study shows that you can find important proteins in samples even if you don't have multiple copies of each sample to compare.

Methodology

The study used a combination of fold-change, statistical tests, and a minimum number of permuted significant pairings to assess protein significance.

Potential Biases

Potential biases may arise from the reliance on statistical tests without sufficient sample replicates.

Limitations

The method may not be as effective when sample sizes are very small or when there are significant biological variations.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-10-43

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication