Resumen La mejora genética vegetal tiene una doble índole científica: Ciencia empírica de la naturaleza y ciencia aplicada de diseño. En este contexto el presente trabajo pretende indagar en la predicción científica como rasgo esencial de esta disciplina en cuanto ciencia aplicada de diseño con especial referencia la mejora de frutales del género Prunus (también denominados frutales de hueso). En mejora genética vegetal podemos hablar de tres niveles de conocimiento asociados a la biología molecular (nivel micro), la constitución genética de un individuo (nivel meso) y al fenotipo o aspecto global de la nueva variedad (nivel macro), que afectan tanto a los tipos de predicción como a las metodologías a aplicar. La predicción constituye el objetivo principal de la mejora genética como ciencia de diseño. Es clave tener un conocimiento del futuro posible para poder hacer un nuevo diseño que estará completo al cabo de algunos años, más de 12 años en el caso de los frutales de hueso. Asimismo, es necesario tener en cuenta en el desarrollo de esta predicción científica aplicada a la mejora de Prunus diversas variables internas (la naturaleza genética del material vegetal de partida, las metodologías disponibles, etc.) y externas (la aceptación social, los factores medioambientales, los estreses bióticos y abióticos, etc.) a la hora de llevar a cabo esta predicción. El grado de conocimiento de estas variables determinará la calidad de la predicción para el diseño de nuevas variedades de Prunus.

Abstract Plant genetic improvement has a double scientific nature: it is both an empirical science of nature and an applied science of design. In this context, this paper provides an investigation of scientific prediction as an essential feature of plant genetic improvement as an applied science of design with special emphasis in the improvement of stone fruits (Prunus genus). In considering plant genetic improvement, we can speak of three levels of knowledge. These levels are associated with molecular biology (micro level); the genetic constitution of an individual (meso level); and the phenotype and overall appearance of the new release (macro level). These levels of reality affect both the type of prediction and the methodology applied. The prediction is the main objective of the plant genetic improvement as science of design. It is very important to have a knowledge of possible future to make a new design that will be complete within a few years, more than 12 years in the case of stone fruit that is the subject of this work. Finally, in developing a scientific prediction applied to Prunus breeding it is necessary to consider different internal (the genetic nature of the starting plant material, the available methodologies, etc.) and external (social acceptation, environmental factors, biotic and abiotic stresses, etc.) variables. The degree of knowledge of these variables determines the quality of the prediction in the design of the new Prunus varieties.

DNA-based molecular markers have been extensively utilized for a variety of studies in both plant and animal systems. One of the major uses of these markers is the construction of genome-wide molecular maps and the genetic analysis of simple and complex traits. However, these studies are generally based on linkage analysis in mapping populations, thus placing serious limitations in using molecular markers for genetic analysis in a variety of plant populations. Therefore, alternative approach has been suggested, linkage disequilibrium-based association analysis which detects and locates quantitative trait loci (QTL) by the strength of the correlation between a trait and a marker. Although association analysis has already been used for studies on genetics of complex traits in humans, its use in plants has newly started. In the present review, we describe what is known about variation in linkage disequilibrium (LD) and summarize published results on association studies in crop plant species. We give a list of different factors affecting LD, and discuss the current issues of LD research in plants. Later, we also describe the various uses of LD in crop plants research and summarize the present status of LD researches in different plant genomes. Finally, future key issues about the application of these studies on the localization of genes in these crop plants have been also discussed.