Techniques de recherche des polymorphismes génétiques

Sequencing the human genome has allowed the discovery of millions of DNA sequence variants. Sequence variations in human DNA are mainly present asSingle Nucleotide Polymorphisms (SNPs); this common form of variation is found about once every 1,000 bases in the human genome and 1.8 million SNPs have now been identified and located. The accessibility of databases of SNPs opens the possibility of studying the influence of these polymorphisms on disease risks as well as on drug responses. Numerous approaches have been set up for the identification of SNPs. In this review we describe the main techniques used for the identification of these polymorphisms. They rely on two major consequences of sequence variations: the apparition or the disappearance of restriction enzyme sites or the alteration of DNA strand hybridization due to the presence of a mismatch. Southern blotting and restriction endonucleases have allowed the development of the technique ofrestriction fragment length polymorphisms (RFLPs), now performed on PCR products. Several other approaches such as denaturing high-performance liquid chromatography or real-time PCR can detect allele differences upon re-hybridization and heteroduplex formation. However, DNA sequencing remains the obligate step for the positive identification of known or unknown SNPs. At last, the development of high-throughput methods allows a large increase in the rate of discovery of SNPs likely.     

Une méthode simple de prévision des tendances appliquée aux séries temporelles de qualité des eaux courantes

Missing values, non-regular sampling, censored data and departure from normality are common problems in the analysis of ecological or environmental time series. For over 20 years, in the scope of the National Basin Network in France, numerous water quality time series have been collected. A method is proposed for the regularization of the series, the detection of trend changes and short-term forecasting, on the basis of the last detected trend, using a non-parametric regression. The method is applied to phosphates and nitrogen concentrations measured on the Sambre river (northern France).