Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs137 activity given the measurements taken in the region of Briansk following the Chernobyl accident.

Broad-scale adaptive genetic variation in alpine plants is driven by temperature and precipitation.

Identifying adaptive genetic variation is a challenging task, in particular in non-model species for which genomic information is still limited or absent. Here, we studied distribution patterns of amplified fragment length polymorphisms (AFLPs) in response to environmental variation, in 13 alpine plant species consistently sampled across the entire European Alps. Multiple linear regressions were performed between AFLP allele frequencies per site as dependent variables and two categories of independent variables, namely Moran's eigenvector map MEM variables (to account for spatial and unaccounted environmental variation, and historical demographic processes) and environmental variables. These associations allowed the identification of 153 loci of ecological relevance. Univariate regressions between allele frequency and each environmental factor further showed that loci of ecological relevance were mainly correlated with MEM variables. We found that precipitation and temperature were the best environmental predictors, whereas topographic factors were rarely involved in environmental associations. Climatic factors, subject to rapid variation as a result of the current global warming, are known to strongly influence the fate of alpine plants. Our study shows, for the first time for a large number of species, that the same environmental variables are drivers of plant adaptation at the scale of a whole biome, here the European Alps.

History of biological warfare and bioterrorism.

Bioterrorism literally means using microorganisms or infected samples to cause terror and panic in populations. Bioterrorism had already started 14 centuries before Christ, when the Hittites sent infected rams to their enemies. However, apart from some rare well-documented events, it is often very difficult for historians and microbiologists to differentiate natural epidemics from alleged biological attacks, because: (i) little information is available for times before the advent of modern microbiology; (ii) truth may be manipulated for political reasons, especially for a hot topic such as a biological attack; and (iii) the passage of time may also have distorted the reality of the past. Nevertheless, we have tried to provide to clinical microbiologists an overview of some likely biological warfare that occurred before the 18th century and that included the intentional spread of epidemic diseases such as tularaemia, plague, malaria, smallpox, yellow fever, and leprosy. We also summarize the main events that occurred during the modern microbiology era, from World War I to the recent 'anthrax letters' that followed the World Trade Center attack of September 2001. Again, the political polemic surrounding the use of infectious agents as a weapon may distort the truth. This is nicely exemplified by the Sverdlovsk accident, which was initially attributed by the authorities to a natural foodborne outbreak, and was officially recognized as having a military cause only 13 years later.