Search

There are many instances in genetics in which we wish to determine whether two candidate populations are distinguishable on the basis of their genetic structure. Examples include populations which are geographically separated, case-control studies and quality control (when participants in a study have been genotyped at different laboratories). This latter application is of particular importance in the era of large scale genome wide association studies, when collections of individuals genotyped at different locations are being merged to provide increased power. The traditional method for detecting structure within a population is some form of exploratory technique such as principal components analysis. Such methods, which do not utilise our prior knowledge of the membership of the candidate populations. are termed unsupervised. Supervised methods, on the other hand are able to utilise this prior knowledge when it is available.In this paper we demonstrate that in such cases modern supervised approaches are a more appropriate tool for detecting genetic differences between populations. We apply two such methods, (neural networks and support vector machines) to the classification of three populations (two from Scotland and one from Bulgaria). The sensitivity exhibited by both these methods is considerably higher than that attained by principal components analysis and in fact comfortably exceeds a recently conjectured theoretical limit on the sensitivity of unsupervised methods. In particular, our methods can distinguish between the two Scottish populations, where principal components analysis cannot. We suggest, on the basis of our results that a supervised learning approach should be the method of choice when classifying individuals into pre-defined populations, particularly in quality control for large scale genome wide association studies.

Introduction: The aim of this study was to describe the reproductive intentions of newlywed Bulgarian families, build predictive models of the desired number of children by artificial neural networks (ANN) and compare these models' overall prediction accuracy with conventional statistical analyses., Material and Methods: Data from a large cross-sectional study on family planning knowledge and habits of newlywed couples from the region of Plovdiv (Bulgaria) in the year 2000 were used. A subset of 17 variables was extracted from the database, in order to construct and train a back-propagation multilayer ANN. Desired number of children, separately for male and female spouses, was used as an output variable. The following statistical analyses were used: descriptive statistics, percentage distribution, nonparametric tests and logit regression analysis., Results and Discussion: Factors that determine the opinion of male and female spouses on desired number of children have different relative importance in the phase space. The reproductive planning of the newly-married family can be better modeled and predicted in women than in men. The reproductive intentions of the modern newlywed Bulgarian family are decided on the basis of mutual agreement and understanding. There is no difference between desired number of children by both partners. The age at marriage of spouses is increasing with a tendency to reach the Western European patterns. Artificial neural networks are found to be a very useful tool in cross-sectional population studies for factor classification and outcome modeling, especially in cases where data are not entirely suitable for logistic regression analyses.