Genome Wide Association Study to predict severe asthma exacerbations in children using random forests classifiers
2011

Predicting Severe Asthma Exacerbations in Children Using Genetic Data

Sample size: 417 publication Evidence: moderate

Author Information

Author(s): Xu Mousheng, Tantisira Kelan G, Wu Ann, Litonjua Augusto A, Chu Jen-hwa, Himes Blanca E, Damask Amy, Weiss Scott T

Primary Institution: Channing Laboratory, Brigham and Women's Hospital, Harvard Medical School

Hypothesis

Using random forests classifiers to select SNPs would result in an improved predictive model of asthma exacerbations.

Conclusion

A random forests algorithm can effectively extract and use information from a small number of samples to predict severe asthma exacerbations.

Supporting Evidence

  • The study identified 127 cases of severe asthma exacerbations among 417 children.
  • Using 160 SNPs improved the predictive model's accuracy to an AUC of 0.66.
  • Clinical traits alone yielded a lower AUC of 0.54, indicating the importance of genetic factors.

Takeaway

Scientists used a computer program to look at many tiny differences in genes to help predict when kids with asthma might have a bad attack.

Methodology

The study used random forests classifiers to analyze SNPs and clinical traits to predict severe asthma exacerbations in children.

Limitations

The study had a limited sample size and potential for false positives with more SNPs used.

Participant Demographics

Participants were Caucasian children with mild to moderate asthma.

Statistical Information

P-Value

0.000266

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2350-12-90

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication