Circulating matrix γ-carboxyglutamate protein (MGP) species are refractory to vitamin K treatment in a new case of Keutel syndrome.

Summary.  Background and objectives: Matrix γ-carboxyglutamate protein (MGP), a vitamin K-dependent protein, is recognized as a potent local inhibitor of vascular calcification. Studying patients with Keutel syndrome (KS), a rare autosomal recessive disorder resulting from MGP mutations, provides an opportunity to investigate the functions of MGP. The purpose of this study was (i) to investigate the phenotype and the underlying MGP mutation of a newly identified KS patient, and (ii) to investigate MGP species and the effect of vitamin K supplements in KS patients. Methods: The phenotype of a newly identified KS patient was characterized with specific attention to signs of vascular calcification. Genetic analysis of the MGP gene was performed. Circulating MGP species were quantified and the effect of vitamin K supplements on MGP carboxylation was studied. Finally, we performed immunohistochemical staining of tissues of the first KS patient originally described focusing on MGP species. Results: We describe a novel homozygous MGP mutation (c.61+1G>A) in a newly identified KS patient. No signs of arterial calcification were found, in contrast to findings in MGP knockout mice. This patient is the first in whom circulating MGP species have been characterized, showing a high level of phosphorylated MGP and a low level of carboxylated MGP. Contrary to expectations, vitamin K supplements did not improve the circulating carboxylated MGP levels. Phosphorylated MGP was also found to be present in the first KS patient originally described. Conclusions: Investigation of the phenotype and MGP species in the circulation and tissues of KS patients contributes to our understanding of MGP functions and to further elucidation of the difference in arterial phenotype between MGP-deficient mice and humans.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.