Comparison of approaches for machine‐learning optimization of neural networks for detecting gene‐gene interactions in genetic epidemiology

Motsinger‐Reif, Alison A.; Dudek, Scott M.; Hahn, Lance W.; Ritchie, Marylyn D.

doi:10.1002/gepi.20307

Cited by 103 publications

(88 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, even if a linear model holds at least locally, a standard fixed effects ANOVA of a highly dimensional, multifactorial system is not feasible, because one ' runs out ' of degrees of freedom (df) (Gianola et al, 2006). Nevertheless, methodologies for dealing with complex epistatic systems are becoming available and these include, for example, machine learning, regularized neural networks (Lee, 2004), neural networks optimized with grammatical evolution computations (Motsinger-Reif et al, 2008) and non-parametric regression (e.g. Gianola et al, 2006 ;Gianola & van Kaam, 2008).…”

Section: Confronting Complexitymentioning

confidence: 98%

Inferring genetic values for quantitative traits non-parametrically

Gianola

Campos

2008

Genet. Res.

View full text Add to dashboard Cite

Inferences about genetic values and prediction of phenotypes for a quantitative trait in the presence of complex forms of gene action, issues of importance in animal and plant breeding, and in evolutionary quantitative genetics, are discussed. Current methods for dealing with epistatic variability via variance component models are reviewed. Problems posed by cryptic, non-linear, forms of epistasis are identified and discussed. Alternative statistical procedures are suggested. Non-parametric definitions of additive effects (breeding values), with and without employing molecular information, are proposed, and it is shown how these can be inferred using reproducing kernel Hilbert spaces regression. Two stylized examples are presented to demonstrate the methods numerically. The first example falls in the domain of the infinitesimal model of quantitative genetics, with additive and dominance effects inferred both parametrically and non-parametrically. The second example tackles a non-linear genetic system with two loci, and the predictive ability of several models is evaluated.

show abstract

Section: Confronting Complexitymentioning

confidence: 98%

Inferring genetic values for quantitative traits non-parametrically

Gianola

Campos

2008

Genet. Res.

View full text Add to dashboard Cite

show abstract

“…Potential avenues for addressing this problem include nonparametric regression (e.g., Gianola et al 2006a, b;Gianola and van Kaam 2008), methods based on regularized neural networks (Lee 2004) or neural networks optimized with grammatical evolution computations (Motsinger-Reif et al 2008). For example, Gianola et al (2006a, b) and Gianola and van Kaam (2008) suggested using reproducing kernel Hilbert spaces (RKHS) regression, where dense molecular information enters into a positive-definite kernel (incidence) matrix whose dimension is equal to the number of individuals with genotype information.…”

Section: Other Forms Of Dealing With Complexitymentioning

confidence: 99%

A non-parametric mixture model for genome-enabled prediction of genetic value for a quantitative trait

Gianola

Manfredi³

et al. 2010

Genetica

View full text Add to dashboard Cite

A Bayesian nonparametric form of regression based on Dirichlet process priors is adapted to the analysis of quantitative traits possibly affected by cryptic forms of gene action, and to the context of SNP-assisted genomic selection, where the main objective is to predict a genomic signal on phenotype. The procedure clusters unknown genotypes into groups with distinct genetic values, but in a setting in which the number of clusters is unknown a priori, so that standard methods for finite mixture analysis do not work. The central assumption is that genetic effects follow an unknown distribution with some "baseline" family, which is a normal process in the cases considered here. A Bayesian analysis based on the Gibbs sampler produces estimates of the number of clusters, posterior means of genetic effects, a measure of credibility in the baseline distribution, as well as estimates of parameters of the latter. The procedure is illustrated with a simulation representing two populations. In the first one, there are 3 unknown QTL, with additive, dominance and epistatic effects; in the second, there are 10 QTL with additive, dominance and additive × additive epistatic effects. In the two populations, baseline parameters are inferred correctly. The Dirichlet process model infers the number of unique genetic values correctly in the first population, but it produces an understatement in the second one; here, the true number of clusters is over 900, and the model gives a posterior mean estimate of about 140, probably because more replication of genotypes is needed for correct inference. The impact on inferences of the prior distribution of a key parameter (M), and of the extent of replication, was examined via an analysis of mean body weight in 192 paternal half-sib families of broiler chickens, where each sire was genotyped for nearly 7,000 SNPs. In this small sample, it was found that inference about the number of clusters was affected by the prior distribution of M. For a set of combinations of parameters of a given prior distribution, the effects of the prior dissipated when the number of replicate samples per genotype was increased. Thus, the Dirichlet process model seems to be useful for gauging the number of QTLs affecting the trait: if the number of clusters inferred is small, probably just a few QTLs code for the trait. If the number of clusters inferred is large, this may imply that standard parametric models based on the baseline distribution may suffice. However, priors may be influential, especially if sample size is not large and if only a few genotypic configurations have replicate phenotypes in the sample.

show abstract

“…The steps of GENN have been previously described in detail [1]. For the purposes of the current study, an option was added to the configuration file to specify the fitness function used: classification error (CE) or balanced error (BE).…”

Section: Grammatical Evolution Neural Networkmentioning

confidence: 99%

“…Grammatical Evolution Neural Networks (GENN) uses grammatical evolution to evolve neural networks to detect gene-gene interactions in studies of complex human diseases [1].…”

Section: Introductionmentioning

confidence: 99%

A balanced accuracy fitness function leads to robust analysis using grammatical evolution neural networks in the case of class imbalance

Hardison

Fanelli

Dudek

et al. 2008

Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

Grammatical Evolution Neural Networks (GENN) is a computational method designed to detect gene-gene interactions in genetic epidemiology, but has so far only been evaluated in situations with balanced numbers of cases and controls. Real data, however, rarely has such perfectly balanced classes. In the current study, we test the power of GENN to detect interactions in data with a range of class imbalance using two fitness functions (classification error and balanced error), as well as data re-sampling. We show that when using classification error, class imbalance greatly decreases the power of GENN. Re-sampling methods demonstrated improved power, but using balanced accuracy resulted in the highest power. Based on the results of this study, balanced error has replaced classification error in the GENN algorithm.

show abstract

Comparison of approaches for machine‐learning optimization of neural networks for detecting gene‐gene interactions in genetic epidemiology

Cited by 103 publications

References 38 publications

Inferring genetic values for quantitative traits non-parametrically

Inferring genetic values for quantitative traits non-parametrically

A non-parametric mixture model for genome-enabled prediction of genetic value for a quantitative trait

A balanced accuracy fitness function leads to robust analysis using grammatical evolution neural networks in the case of class imbalance

Contact Info

Product

Resources

About