Search

The motivating application for this research is the problem of recognizing a planar object consisting of points from a noisy observation of that object. Given is a planar Gaussian mixture model T (x) representing an object along with a noise model for the observation process (the template). Also given are points representing the observation of the object (the query). We propose a method to determine if these points were drawn from a Gaussian mixture Q(x) with the same shape as the template. The method consists in comparing samples from the distribution of distances of T (x) and Q(x), respectively. The distribution of distances is a faithful representation of the shape of generic Gaussian mixtures. Since it is invariant under rotations and translations of the Gaussian mixture, it provides a workaround to the problem of aligning objects before recognizing their shape without sacrificing accuracy. Experiments using synthetic data show a robust performance against type I errors, and few type II errors when the given template Gaussian mixtures are well distinguished.

Rapports

in Fiches Jean-Luc (dir.) : Ambrussum, fouilles programmées 2005-2007, compte-rendu de la campagne 2005; Rapports

In North America, the pea aphid Acyrthosiphon pisum encompasses ecologically and genetically distinct host races that offer an ideal biological system for studies on sympatric speciation. In addition to its obligate symbiont Buchnera, pea aphids harbour several facultative and phylogenetically distant symbionts. We explored the relationships between host races of A. pisum and their symbiotic microbiota to gain insights into the historical process of ecological specialization and symbiotic acquisition in this aphid. We used allozyme and microsatellite markers to analyse the extent of genetic differentiation between populations of A. pisum on pea, alfalfa and clover in France. In parallel, we examined: (i) the distribution of four facultative symbionts; and (ii) the genetic variation in the Buchnera genome across host-associated populations of A. pisum. Our study clearly demonstrates that populations of A. pisum on pea, clover and alfalfa in France are genetically divergent, which indicates that they constitute distinct host races. We also found a very strong association between host races of A. pisum and their symbiotic microbiota. We stress the need for phylogeographic studies to shed light on the process of host-race formation and acquisition of facultative symbionts in A. pisum. We also question the effects of these symbionts on aphid host fitness, including their role in adaptation to a host plant.

Discrete variation in wing morphology is a very common phenomenon in insects and has been used extensively in the past 50 years as a model to study the ecology and evolution of dispersal. Wing morph determination can be purely genetic, purely environmental, or some combination of the two. The precise genetic determinants of genetically based wing morph variation are unknown. Here we explore the genetic basis of wing polymorphism in the pea aphid, which can produce either winged or wingless males. We confirm that three types of pea aphid clones coexist in natural populations, those producing winged males only, those producing wingless males only, and those producing a mixture of both. A Mendelian genetic analysis reveals that male wing polymorphism in pea aphids is determined by a single locus, two alleles system. Using microsatellite loci of known location, we show that this locus is on the X chromosome. The existence of a simple genetic determinism for wing polymorphism in a system in which genetic investigation is possible may help investigations on the physiological and molecular mechanisms of genetically-based wing morph variation. This locus could also be used in the search for genes involved in the wing polyphenism described in parthenogenetic females and to investigate the interplay between polymorphisms and polyphenisms.