Search

Biogeography and the occurrence of small mammals are usually hard to investigate due to the small size and secretive habits of these mammals. Available data are particularly insufficient on minor islands and at national borders, where research efforts are usually scarce. Here we briefly updated the knowledge on murid rodents on two remote Italian small islands (Lampedusa and Pantelleria) at the southernmost Italian borders. During summer 2019, house mice and rats were sampled in Lampedusa and Pantelleria and molecular markers were sequenced for species identification. The new sequences of Mus domesticus were also compared with samples from previous works collected in Lampedusa, Pantelleria, and several localities in the Mediterranean basin. Moreover, our analyses provided the first genetic evidence of the occurrence of Rattus norvegicus in Lampedusa. To conclude, R. rattus was confirmed to be present in Pantelleria. The newly collected haplotype of M. domesticus from Pantelleria is similar to those currently known for Sicily, whereas the new haplotype from Lampedusa partially diverges from the ones previously described, and clusters with haplotypes from North Africa and the Middle East.

The site of Coste San Giacomo, in the Anagni Basin, is renowned for its mammal fauna of Early Pleistocene (Gelasian) age, including 32 taxa of small and large mammals. Adding to these, a small amount of remains of amphibians and reptiles was also recovered during recent excavations focused on collecting small vertebrates as well. The analysis of these remains resulted in the identification of a total of at least six different taxa: three anurans (Bufonidae indet., Pelophylax sp., and Rana sp.) and three reptiles (Lacertidae indet., Pseudopus sp., and Natrix sp.). This assemblage is comprised of taxa that are common in the Italian Quaternary, but also presents the first occurrence of the large anguid Pseudopus from the Apennine Peninsula in this period. On the whole, the herpetofauna indicates the presence of humid conditions in CSG at the moment of deposition of the bone bed that yielded the vertebrate remains. [ABSTRACT FROM AUTHOR]

New evidence arising from the study of the microtheriological materials from the Late Neolithic site of Tosina of Monzambano (Province of Mantova, northern Italy) shows that Mus musculus ssp. was present in Western Europe in the Late Neolithic -- at least 1500 years earlier than previously clearly demonstrated. This discovery provides evidence for a first migratory wave, following human populations moving from east to west, of this highly synanthropic animal species. The data are consistent with the evidence of Mus musculus musculus -- hypothesised to belong to a second wave of house mice reaching Europe -- from the Romanian Copper Age site of Buc§ani La Pod tell. [ABSTRACT FROM AUTHOR]

In this study a single specimen of a species of the genus Sicista, collected from the Caverna Generosa, is described: a right mandible with first molar and m/2-m/3 alveoli. The specimen shows a morphological and morphometrical similarity to Sicista subtilis and it may represent the first record in Italy and the third in the whole Europe of this species. The discovery defines advancement in the distribution of Sicista in northern Italy and it certainly demonstrates that an important faunal migration took place during the Last Glacial Maximum in the western part of the Po valley. [ABSTRACT FROM AUTHOR]

We have developed a machine learning toolbox, called SHOGUN, which is designed for unified large-scale learning for a broad range of feature types and learning settings. It offers a considerable number of machine learning models such as support vector machines, hidden-Markov models, multiple kernel learning, linear discriminant analysis, and more. Most of the specific algorithms are able to deal with several different data classes. We have used this toolbox in several applications from computational biology, some of them coming with no less than 50 million training examples and others with 7 billion test examples. With more than a thousand installations worldwide, SHOGUN is already widely adopted in the machine learning community and beyond. SHOGUN is implemented in C++ and interfaces to MATLAB,™ R, Octave, Python, and has a stand-alone command line interface. The source code is freely available under the GNU General Public License, Version 3 at http://www.shogun-toolbox.org. [ABSTRACT FROM AUTHOR]