Enhancing the efficiency of a polymerase chain reaction using gold nanoparticles

Random Forest is a prediction technique based on growing trees on bootstrap samples of data, in conjunction with a random selection of explanatory variables to define the best split at each node. In the case of a quantitative outcome, the tree predictor takes on a numerical value. We applied Random Forest to the first replicate of the Genetic Analysis Workshop 13 simulated data set, with the sibling pairs as our units of analysis and identity by descent (IBD) at selected loci as our explanatory variables. With the knowledge of the true model, we performed two sets of analyses on three phenotypes: HDL, triglycerides, and glucose. The goal was to approach the mapping of complex traits from a multivariate perspective. The first set of analyses mimics a candidate gene approach with a high proportion of true genes among the predictors while the second set represents a genome scan analysis using microsatellite markers. Random Forest was able to identify a few of the major genes influencing the phenotypes, such as baseline HDL and triglycerides, but failed to identify the major genes regulating baseline glucose levels. Background Trees are nonparametric prediction models. In the context of genetic linkage analysis, Zhang et al. [1] used single trees to identify markers where mean identity-by-descent (IBD) sharing in sib pairs is predictive of the affection status of the pair. Breiman [2] and others have reported that significant improvements in prediction accuracy are achieved by using a collection of trees, called a random forest. We applied Random Forest to the first replicate of the Genetic Analysis Workshop 13 simulated data set, with the sibling pairs as our units of analysis. We used the IBD scores (mean IBD or probability of sharing two alleles IBD) at various loci on the genome to predict the absolute difference between phenotypic values in each pair of siblings. In the candidate gene approach, IBD was estimated at the location of the candidates while in the genome scan approach, IBD was harvested at each microsatellite marker. Methods We used MEGA2 [3] to create all nuclear families from the data set and then computed multipoint IBD probabilities on the whole genome for each family with GENEHUNTER [4]. We created two variables out of the IBD sharing probabilities: mean IBD and probability of sharing two alleles (Z2). We performed separate analyses using the mean IBD only and Z2 only, and a joint analysis using both mean IBD and Z2 as explanatory variables. from Genetic Analysis Workshop 13: Analysis of Longitudinal Family Data for Complex Diseases and Related Risk Factors New Orleans Marriott Hotel, New Orleans, LA, USA, November 11–14, 2002 Published: 31 December 2003 BMC Genetics 2003, 4(Suppl 1):S64 Genetic Analysis Workshop 13: Analysis of Longitudinal Family Data for Complex Diseases and Related Risk Factors Laura Almasy, Christopher I Amos, Joan E Bailey-Wilson, Rita M Cantor, Cashell E Jaquish, Maria Martinez, Rosalind J Neuman, Jane M Olson, Lyle J Palmer, Stephen S Rich, M Anne Spence, Jean W MacCluer This article is available from: http://www.biomedcentral.com/1471-2156/4/s1/S64