Gene-gene interaction filtering with ensemble of filters
2011

Improving Gene-Gene Interaction Filtering with Ensemble Methods

Sample size: 400 publication Evidence: high

Author Information

Author(s): Yang Pengyi, Ho Joshua WK, Yang Yee Hwa, Zhou Bing B

Primary Institution: University of Sydney

Hypothesis

Aggregating results from multiple runs of SNP filtering algorithms may improve filtering performance.

Conclusion

The ensemble approach significantly stabilizes SNP ranking and improves the retention of functional SNP pairs compared to traditional methods.

Supporting Evidence

  • The ensemble of TuRF achieved the highest success rate in retaining gene-gene interactions.
  • The proposed ensemble approach can produce much more stable SNP ranking.
  • TuRF-E performed the best in retaining interaction SNP pairs across various datasets.

Takeaway

This study shows that using a group of results from different runs of a filtering algorithm can help find important gene interactions better than using just one result.

Methodology

The study used simulated and real datasets to compare the performance of ensemble filtering methods against traditional SNP filtering algorithms.

Limitations

The study may not account for all possible biases in SNP filtering methods.

Participant Demographics

The study involved simulated datasets with varying sample sizes and heritability, as well as a real dataset of AMD samples.

Digital Object Identifier (DOI)

10.1186/1471-2105-12-S1-S10

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication