Accounting for multiple comparisons in a genome-wide association study (GWAS)
2010

Improving Statistical Significance in Genome-Wide Association Studies

Sample size: 1514 publication 10 minutes Evidence: moderate

Author Information

Author(s): Randall C. Johnson, George W. Nelson, Jennifer L. Troyer, James A. Lautenberger, Bailey D. Kessing, Cheryl A. Winkler, Stephen J. O'Brien

Primary Institution: Basic Research Program, SAIC-Frederick, Inc.

Hypothesis

How can we improve the statistical significance thresholds in genome-wide association studies (GWAS) considering multiple comparisons?

Conclusion

Using methods like SLIDE and simpleℳ provides more accurate significance thresholds in GWAS compared to traditional Bonferroni adjustments.

Supporting Evidence

  • The study found that traditional Bonferroni corrections were overly conservative.
  • SLIDE and simpleℳ methods showed improved power in detecting associations.
  • The results suggest that using LD blocks for corrections can lead to anti-conservative adjustments.

Takeaway

This study looks at how to better understand the results of genetic studies by using smarter math to avoid mistakes when testing many genes at once.

Methodology

The study used data from 1514 European Americans genotyped for 700,078 SNPs and compared various statistical correction methods.

Potential Biases

Potential bias due to the over-reliance on specific statistical methods that may not apply universally.

Limitations

The study's findings may not be generalizable to populations outside of European Americans.

Participant Demographics

1514 European Americans, including 1,255 HIV-infected and 259 HIV-negative individuals.

Statistical Information

P-Value

p < 5 × 10-8

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2164-11-724

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication