Class prediction and feature selection are two learning tasks that are strictly paired in the search of molecular profiles from microarray data. Researchers have become aware how easy it is to incur a selection bias effect and complex validation setups are required to avoid overly optimistic estimates of the predictive accuracy of the models and incorrect gene selections. This paper describes a semi-supervised pattern discovery approach that uses the by-products of complete validation studies on experimental setups for gene profiling. In particular, we introduce the study of the patterns of single sample responses (sample-tracking profiles) to the gene selection process induced by typical supervised learning tasks in microarray studies. We originate sample-tracking profiles as the aggregated off-training evaluation of SVM models of increasing gene panel sizes. Genes are ranked by E-RFE, an entropy-based variant of the recursive feature elimination for support vector machines (RFE-SVM). A Dynamic Time Warping (DTW) algorithm is then applied to define a metric between sample-tracking profiles. An unsupervised clustering based on the DTW metric allows automating the discovery of outliers and of subtypes of different molecular profiles. Applications are described on synthetic data and in two gene expression studies.

This paper reports on the processing of time series of MODIS NDVI/EVI and LST satellite data in a Geographical Information System (GIS). The required data preparations for the integration of MODIS data in GIS is described with focus on the reprojection from MODIS/Sinusoidal projection to national coordinate systems. To remove low quality pixels, the MODIS quality maps are utilised. We explain subsequent filtering of Land Surface Temperature maps with an outlier detector to eliminate originally undetected cloud pixels. Further analysis of time series is briefly discussed as well as related landscape epidemiological applications in the field of tick-borne diseases.

In this paper we present a suite of new image processing tools for the GRASS Geographic Information System. New modules are suggested to support improved and semi-automated geocoding of vertical imagery. The ortho-rectification procedures have been extended to rectify oblique imagery from digital hand-held cameras for rendering purposes. Multi- and hyperspectral image analysis has been implemented to derive landuse/landcover maps at subpixel resolution. Image fusion with the Brovey transform is shown. We finally show high performance SMAP image classification on an openMosix cluster.

Cos'è il software Open Source? Senza perdersi in eccessivi tecnicismi, si può dire che è quello che tutti possono usare, senza limitazioni. Il termine equivale all'incirca a 'software liberò, per cui ci si riferisce di solito a questi programmi con l'acronimo FOSS (Free and Open Source Software). Anche se di solito il FOSS è distribuito gratuitamente, non c'è corrispondenza fra FOSS e p rogrammi gratuiti; esistono infatti molti programmi gratuiti che impongono forti limitazioni al loro uso.

Remote sensing data can easily provide images of snow covered areas and, therefore, in the near future it will be possible to follow the time evolution of snow melting spatial patterns with increasing spatial and temporal resolution. While during the accumulation phase of the processes air temperature patterns dominate the snow distribution, during the melting period the patterns are dominated by the complex interplay of topography, available radiation forcings and atmosphere turbulent transfer processes. In this contribution the snow covered area of an alpine basin in Trentino (Italy), extended approximately 200 square kilometers, is studied during the melting time using the distributed hydrological model GEOtop and data from some satellite platforms currently available. GEOtop describes the soil-snow-atmosphere energy and mass exchanges taking into account the topography effects (slope and aspect), the solar radiation dependence on the weather conditions, and the snow physics, and has already been tested with point data. Snow cover extent can be provided by remote-sensed data at different resolution (Quick Bird 1 meter, SPOT 10 meters and MODIS 500 meters), and by data from in situ measurements. The model reproduces quite well the physical features of snow melting and shows a fair agreement with the data, although it does not describe the snow redistribution due to wind drift, and it seems to be an effective tool to upscale and downscale remote sensing data at different resolution.

In this paper we described the efforts towards Internationalization (i18n) of GRASS GIS and discuss present status. Development of GRASS-i18n has been carried out in the following four broadways. a) Tcltkgrass-i18n and Nviz-i18n Graphical User Interface (GUI) b) i18n of GRASS command help system c) i18n of GRASS text drawing command such as d.text and d.label d) i18n of GRASS PostScript print command (ps.map) As a result of the above developments, it is now easily possible to localize GRASS GUI to other languages by creating appropriate message files in other languages. Further, display and printing for other languages are supported using Unicode (UTF-8) encoding scheme. The i18n of GRASS and localization to Japanese and Vietnamese were completed for version 5.0.3. The i18n functionality and localization to other are being incorporated into the GRASS version 6.0.0 release.