BALZARINI MONICA GRACIELA
Congresos y reuniones científicas
Título:
Multivariate techniques for the association of molecular data to other sources of information and its applications from phenomics to conservation biology
Autor/es:
TEICH, I.; BRUNO, C.; BALZARINI, M.
Reunión:
Congreso; 2º Congreso Argentino de Bioinformática y Biología Computacional; 2011
Resumen:
The amount of available information in biological studies has increased dramatically not only at the molecular level but also at other levels of organization. At the molecular level, technologies as DNA sequencing have swamped data bases challenging Moore´s law. At the ecosystem level, satellite imaginery also provides a constant influx of environmental data not easy to deal with (Figure 1). Concomitantly, the need to study organisms as a whole and find associations between data sets of different nature has increased. In phenomics, it is essential to analyze genotype, phenotype, environment and the interactions among them. For non-model species and natural populations, finding associations between genetic markers and environmental characteristics allows the identification of genes and genetic mechanisms underlying adaptation. Statistical models of such complex interactions are difficult for both computational and biological interpretation aspects. On the contrary, algorithmic methods can be used to filter the main signals of genetic data and to study the association between sets of genotypic, phenotypic and environmental data. These methods are more straightforward and provide meaningful insight for later statistical modeling. In this work we present different multivariate techniques that can be used to compare the level of covariation of different data sets between situations or treatments and to find associations. We illustrate the application of Procrustes (PGA) and Coinertia Analysis (PCIA) in association studies between morphological and molecular traits (SSR) in an agronomical important tree species (Prosopis sp.) and between remote sensed environmental data and molecular marker data (AFLP) in a non-model tree species (Polylepis australis).