[Assessing the anthropogenic impact on lake Shira from antibiotic resistance of heterotrophic bacteria by neural networks methods].

A general approach to assessing the anthropogenic impact on lake ecosystems is proposed and exemplified for the case of Lake Shira (Republic of Khakasia, Russia). The impact strength is estimated by applying neural network-based methods to samples of data on interdependent marking features of autochthonous and allochthonous bacteria isolated from the lake in 1997-2001. The proposed combination of analysis methods makes it possible to determine the state of an ecosystem from both small- and large-size samples of data having complex interrelations.

[Database on natural and transgenic luminous microorganisms: "BiolumBase"].

The database "BiolumBase" is designed for the selection and systematization of available world information on microorganisms containing bioluminescent systems; it includes two sections: "natural" and "transgenic" luminous microorganisms. At present, logic schemes of divisions, classification of the objects, presentation of characteristics, and the inputs of relative information, as well as the necessary program modules including links to the database, are developed. The database is constructed on the basis of published data and our own experimental results; the subsequent linkage of the database to the Internet is envisaged. Users will be able to obtain not only the catalogues of strains but also information concerning the properties and functions of the known species of luminous bacteria, the structure, regulatory mechanisms, and application of bioluminescent systems and genetically engineered constructions with lux genes, as well as to find references and to search strains by using any set of attributes. The database will provide information that is of interest for the development of microbial ecology and biotechnology, in particular, for the prediction of biological hazard from the application of transgenic strains.