Showing papers on "Selection (genetic algorithm) published in 1974"

A Model for Certain Types of Selection Experiments

Abstract: A model is suggested for certain types of experiments that involve individuals being selected one-byone from a population consisting of two or more different classes of individuals. Selection with and without replacement is considered and equations are provided for the estimation of the selectivity of the selection process. The equations are illustrated on data from experiments involving selective predation and food competition.

Prediction and evaluation of response to selection with overlapping generations

Abstract: In a population in which generations overlap the improvement in performance in successive years resulting from a single year of selection is not constant, for the genes from a group of selected individuals may take many years to pass through the population. A formal method is developed for predicting responses and discounted returns from improvement in populations with overlapping generations including, if necessary, generations of multiplication of breeding stock.The method is based on a matrix which specifies the passage of genes between the different age groups and sexes. Simple matrix operations can be used to compute the proportion of genes in animals of both sexes and each age in the population at any time which derive from a group of selected animals at an earlier time. The response produced by these selected animals equals the product of their genetic selection differential and the proportion of genes deriving from them. Comparisons are made between responses predicted using this theory and the classical theory of uniform rates of response, and a method is given for computing the time lag of genes passing through the population.The results are extended to enable computation of discounted returns from improvement.

Cannabis: An Example of Taxonomic Neglect

Abstract: Native apparently somewhere in central Asia, cannabis is at present one of the most widely disseminated cultivated plants. Cannabis as we know it has developed together with man as a multi-purpose economic plant; and, as a result of selection for desirable characteristics, it has become one of the most variable of cultivated plants. Due to the extraordinary plasticity and variability of cultivated cannabis, there can be no progress in unravelling the taxonomic complexities in the genus until the biology of the wild populations is investigated.

Interdemic selection and the evolution of altruism: a computer simulation study.

Abstract: In the classical interpretation, natural selection is an egoistic process. Individual organisms compete for representation of their genes in future generations. The genomes of those individuals better able to survive and reproduce occur in higher frequencies. This process of natural selection working at the level of individual organisms is capable of explaining the vast majority of adaptations of species to their environment. There are, however, a significant minority of characters which are difficult to explain by this mechanism. The evolution of such diverse phenomena as social behavior in hymenopterous and isopterous insects, warning calls in birds and even sexual reproduction in eukaryotes, cannot readily be explained by natural selection operating at the level of individuals. In general, characters which benefit the group as a whole, but are selectively neutral or deleterious to the bearers, fail to conform to the requirements for individual natural selection. Either for want of a better term, or because of its anthropocentric appeal, this class of characters have become known as altruistic. Two primary mechanisms have been used to explain the evolution of altruistic characters: kin selection and interdemic selection. Both of these processes require some form of non-random association among the individual members of the population. For this reason they have been appropriately combined under the general heading of group selection (see Wilson, 1973). Kin selection groups are composed of individuals of common ancestry. Interdemic selection groups are spatially or temporaly bounded subsets of the population forming primarily endogamous breeding units or

Selection in dual purpose cattle populations: defining the breeding objective

Abstract: The selection objective in choosing dual-purpose bulls for use in artificial insemination should be the economic benefit conferred on the population with each insemination. The value of a unit of genetic superiority in the bull for a beef or a dairy trait, as realized through one insemination, depends on two factors—the economic value per unit of superiority and the number of times that superiority is expressed. The economic values are easily established. A standard unit of expression for a trait is defined to be one progeny expression in the year in which the insemination is carried out. A method is developed for summarizing the total genetic consequences of an insemination in these standard units. It takes account of the timing of the insemination, the number of years over which the evaluation is carried out, a discount factor, the dilution of the bull's genotype in his descendants, the female replacement rate, the calf survival rate, and the probability that a surviving calf becomes a dairy cow. The method is independent of the full testing procedures, the intensity of selection, and the level of bull usage. It applies, equally, to selection of bulls for general use and for use in planned matings to produce young sires for testing.

Adaptive coloration in Peromyscus polionotus experimental selection by owls

Abstract: Predation experiments were conducted in field enclosures using two naturally occurring phenotypes of the old-field mouse ( Peromyscus polionotus ) as prey, with barn owls and screech owls as predators. Two mice, one light brown and one dark brown, were released into enclosures each containing one owl. The phenotype of the first mouse captured was recorded. Effectiveness of selection against the conspicuous prey by owls under the experimental conditions was estimated using the selection index from Dice (1947). Owls captured more of the conspicuous than the matching phenotype on both dark and light soils. Changes in intensity of illumination altered the intensity of selection and this effect differed on light and dark soils.

Density-dependent selection and character displacement

Abstract: A model is proposed to study the effect of natural selection on a metrical character subject both to selection for an optimal value and to density-dependent selection due to competition. The model incorporates the idea that competition is more severe between phenotypically similar individuals, perhaps through resource specialization. This may explain the maintenance of genetic variability in the population despite the tendency of optimizing selection to eliminate it. This suggestion is examined mathematically, and I show that the population will tend to evolve by shifting its variance until it reaches a critical value at which the direct effect of selection on the phenotypic variance is minimized. Numerical results show that genetic variability can be maintained in the population under this model, but that quite a high degree of differential competition (or resource specialization) is required. The model is then extended to two noninterbreeding populations as a model of character displacement through comp...

Habitat selection in titmice

Abstract: HABITAT selection has important ecological and evolutionary consequences for birds as for other animals1–4, yet although there is abundant evidence from natural observation that it occurs, little is known of the cues by which habitats are recognised, nor the role of experience in determining such preferences5–10.

Selection of the best method in stereology.

Abstract: SUMMARY A brief historical survey of the origin of stereological methods is followed by a synoptic characterization of their nature. They fall into two main classes depending on the structural models at their origin. Those methods which aim at a size analysis of the particle population provide more information, but are more subject to limitations, than those which estimate bulk parameters of the structure as a whole. The choice between the different methods must be guided by a clear definition of the most informative parameters and of the geometric properties of the structure.

Changing stepsize in the integration of differential equations using modified divided differences.

Abstract: Multistep methods for solving differential equations based on numerical integration formulas or numerical differentiation formulas (for stiff equations) require special provision for changing the stepsize. New algorithms are given which make the use of modified divided differences an attractive way to carry out the change in stepsize for such methods. Error estimation and some of the important factors in stepsize selection and the selection of integration order are also considered.

A note on increasing the limit of selection through selection within families

Abstract: It is shown that, in a finite population, by ensuring an equal contribution of offspring from each family we achieve a higher selection limit than by using mass selection, given that the correlation between selection criterion and additive genetic value is sufficiently high. The difference between the selection limits of the two schemes increases with more intense selection. The theoretical results were verified by Monte Carlo Simulation, and the influences of several factors were investigated. It is shown that such a breeding scheme might be useful in dairy cattle.