Katherine Bowers

Publications (2)5.63 Total impact

• Article: Glutathione pathway gene variation and risk of autism spectrum disorders.

  Katherine Bowers, Qing Li, Joseph Bressler, Dimitrios Avramopoulos, Craig Newschaffer, M Daniele Fallin
  [show abstract] [hide abstract]
  ABSTRACT: Despite evidence that autism is highly heritable with estimates of 15 or more genes involved, few studies have directly examined associations of multiple gene interactions. Since inability to effectively combat oxidative stress has been suggested as a mechanism of autism, we examined genetic variation 42 genes (308 single-nucleotide polymorphisms (SNPs)) related to glutathione, the most important antioxidant in the brain, for both marginal association and multi-gene interaction among 318 case-parent trios from The Autism Genetic Resource Exchange. Models of multi-SNP interactions were estimated using the trio Logic Regression method. A three-SNP joint effect was observed for genotype combinations of SNPs in glutaredoxin, glutaredoxin 3 (GLRX3), and cystathione gamma lyase (CTH); OR = 3.78, 95% CI: 2.36, 6.04. Marginal associations were observed for four genes including two involved in the three-way interaction: CTH, alcohol dehydrogenase 5, gamma-glutamylcysteine synthetase, catalytic subunit and GLRX3. These results suggest that variation in genes involved in counterbalancing oxidative stress may contribute to autism, though replication is necessary.
  Journal of Neurodevelopmental Disorders 06/2011; 3(2):132-43. · 3.06 Impact Factor
• Article: Importance measures for epistatic interactions in case-parent trios.

  Holger Schwender, Katherine Bowers, M Daniele Fallin, Ingo Ruczinski
  [show abstract] [hide abstract]
  ABSTRACT: Ensemble methods (such as Bagging and Random Forests) take advantage of unstable base learners (such as decision trees) to improve predictions, and offer measures of variable importance useful for variable selection. LogicFS has been proposed as such an ensemble learner for case-control studies when interactions of single nucleotide polymorphisms (SNPs) are of particular interest. LogicFS uses bootstrap samples of the data and employs the Boolean trees derived in logic regression as base learners to create ensembles of models that allow for the quantification of the contributions of epistatic interactions to the disease risk. In this article, we propose an extension of logicFS suitable for case-parent trio data, and derive an additional importance measure that is much less influenced by linkage disequilibrium between SNPs than the measure originally used in logicFS. We illustrate the performance of the novel procedure in simulation studies and in a case study of 461 case-parent trios with autistic children.
  Annals of Human Genetics 01/2011; 75(1):122-32. · 2.57 Impact Factor