Enhancing density-based clustering: Parameter reduction and outlier detection

Clustering is a widely used unsupervised data mining technique. It allows to identify structures in collections of objects by grouping them into classes, named clusters, in such a way that similarity of objects within any cluster is maximized and similarity of objects belonging to different clusters is minimized. In density-based clustering, a cluster is defined as a connected dense component and grows in the direction driven by the density. The basic structure of density-based clustering presents some common drawbacks: (i) parameters have to be set; (ii) the behavior of the algorithm is sensitive to the density of the starting object; and (iii) adjacent clusters of different densities could not be properly identified. In this paper, we address all the above problems. Our method, based on the concept of space stratification, efficiently identifies the different densities in the dataset and, accordingly, ranks the objects of the original space. Next, it exploits such a knowledge by projecting the original data into a space with one more dimension. It performs a density based clustering taking into account the reverse-nearest-neighbor of the objects. Our method also reduces the number of input parameters by giving a guideline to set them in a suitable way. Experimental results indicate that our algorithm is able to deal with clusters of different densities and outperforms the most popular algorithms DBSCAN and OPTICS in all the standard benchmark datasets.     

The olfactory organ of the trout Salmo trutta fario: A novel localization for a progestin receptor

A progestin receptor (PR) has been detected in the olfactory organ of the trout Salmo trutta fario. The specificity of this receptor was high for 17α,20β‐dihydroxy‐4‐pregnen‐3‐one (17α,20β‐DP), but it also bound 17α‐hydroxy‐progesterone (17α‐OHP) and 21‐hydroxyprogesterone (21‐OHP), even when present at low concentrations (10‐fold in relative binding affinity assay). Progesterone (P) competed effectively at much higher concentrations (1,000‐fold in relative binding affinity assay). Immunohistochemical studies carried out with three different monoclonal antibodies against human progesterone receptor (hPR), chicken progesterone receptor hinge region (cPR), and chicken progesterone receptor A/B domain (PR22), revealed that immunoreactivity was present in the epithelium of the olfactory organ of females and males of the trout Salmo trutta fario only against hPR. Western blotting showed two hPR immunoreactive bands of about 62 and 66 kDa. Finally, a portion of the cDNA of about 300 nucleotides extending over the DNA binding domain and the ligand binding domain was cloned and sequenced, revealing a high degree of sequence homology of the PR in Salmo trutta fario with the PR in other teleosts. Microsc. Res. Tech., 2010. © 2009 Wiley‐Liss, Inc.     

Effects of milk high pressure homogenization on biogenic amine accumulation during ripening of ovine and bovine Italian cheeses

The aim of this work was to evaluate the biogenic amine (BA) content during the ripening of both bovine and ovine cheeses obtained using milk subjected to a homogenization treatment at 100 MPa before cheese-making. The data obtained were compared with those from cheeses produced by the same milks without any treatment or thermized. The results showed that both microbial ecology and BA concentrations of cheeses during ripening were significantly influenced by the type of milk used for cheese-making and by the treatment applied to the raw materials. In particular, the microbial counts found in Caciotta indicated that the high pressure homogenization (HPH) of milk significantly reduced the presence of the yeasts, Micrococcaceae and lactobacilli at the end of ripening. On the other hand, the HPH treatment of milk favoured the proliferation of yeasts in ovine cheese. Moreover, the ovine cheeses were characterized by a remarkably higher accumulation of BA than bovine cheeses. However, the HPH treatment of milk was able to drastically reduce the biogenic amine concentrations in both cheese typologies at the end of ripening.     

Penalised regression splines: theory and application to medical research

Generalised additive models (GAMs) allow for flexible functional dependence of a response variable on covariates. The aim of this article is to provide an accessible overview of GAMs based on the penalised likelihood approach with regression splines. In contrast to the classical backfitting, the penalised likelihood framework taken here provides researchers with an efficient computational method for automatic multiple smoothing parameter selection, which can determine the functional form of any relationship from the data. We illustrate through an example how the use of this methodology can help to gain insights into medical research.     

Role of surface-inoculated Debaryomyces hansenii and Yarrowia lipolytica strains in dried fermented sausage manufacture. Part 2: Evaluation of their effects on sensory quality and biogenic amine content

The aim was to study the effects of Debaryomyces hansenii and Yarrowia lipolytica strains, used with lactic acid starter cultures (Lactobacillus plantarum), in the manufacture of dried fermented sausages in order to understand their effects on volatile profile, biogenic amine content and sensory properties. The experimental data showed that every yeast strain produced a specific profile of volatile metabolic products. The yeasts also gave sausages with distinctive sensory properties. The degree of mincing also influenced these properties, but none of these factors had significant influence upon the accumulation of biogenic amines.     

Identification of fish species by 5S rRNA gene amplification

The 5S ribosomal RNA is a very suitable target for easy, rapid and inexpensive fish species identification due to its structure, consisting of a conserved region followed by a species-specific noncoding region called ‘nontranscribed spacer’. We have exploited this species-specificity in length and sequence to discriminate among fish species which can be subjected to substitution in the fish markets. After sequencing and alignment of the corresponding portions of the 5S rRNAs of different fish species, we have designed the primer pairs necessary for PCR amplification on the DNA traits which most diverged and a primer pair on conserved regions. Our results have shown the feasibility, simplicity and reliability of the proposed approach for the detection of mislabelling or fraudulent substitution of fish species.     

Microbiological quality of filled pasta in relation to the nature of heat treatment

BACKGROUND: The role of nitric oxide in the ischemia/reperfusion injury of the pancreas is still unclear. In other organs, protective as well as aggravating effects have been described. We have, therefore, investigated the effect of the nitric oxide donor sodium nitroprusside on pancreatic ischemia/reperfusion injury. METHODS: In Landrace pigs, after transsection of the pancreas, complete vascular isolation of the pancreatic tail was performed. The tail was subjected to 3 hr of warm ischemia and thereafter reperfusion (6 hr). The animals were divided into a control group (n=7) and a treatment group (n=7) that received 15 mg of sodium nitroprusside after reperfusion intra-arterially into the splenic artery. RESULTS: The morphological tissue damage and lipase activity in the venous effluent of the pancreas were significantly lower in the treatment group. Partial oxygen tension in the tissue after reperfusion was markedly reduced in the control group, indicating an impairment of microcirculation. In the treatment group, however, partial oxygen tension in the tissue was significantly higher (43 vs. 20 mmHg; P<0.014). Furthermore, total blood flow through the pancreatic tail in the treatment group was found to be significantly higher in the late reperfusion period (14 vs. 9.5 ml/min at 5 hr after reperfusion; P<0.05). CONCLUSION: There is a marked impairment of pancreatic microcirculation after reperfusion. Sodium nitroprusside counteracts this impairment and has a protective effect on ischemia/reperfusion injury of the pancreas.     

Scalability analysis of a two level domain decomposition approach in space and time solving data assimilation models

We are concerned with the mapping on high performance hybrid architectures of a parallel software implementing a two level overlapping domain decomposition, that is, along space and time directions, of the four dimensional variational data assimilation model. The reference architecture belongs to the SCoPE (Sistema Cooperativo Per Elaborazioni scientifiche multidisciplinari) data center, located at University of Naples Federico II. We consider the initial boundary problem of the shallow water equation and analyse both strong and weak scaling. Keeping the efficiency always greater than $$ and about $$ in most cases, we experimentally find that the isoefficiency function grows a little more than linearly with respect to the number of processes. Results, obtained by using the parallel computing toolbox of MATLABR2013a, are in agreement with the algorithm's performance prevision based on the scale up factor, confirming the appropriate mapping of the algorithm on the hybrid architecture.     

Characterization of metal-supported Al2O3 thin films by scanning electrochemical microscopy

In this paper, physiochemical properties of amorphous alumina thin films, grown by the metal organic chemical vapour deposition process on the surface of platinum (Pt/Al₂O₃) and stainless steel (SS/Al₂O₃), were investigated in aqueous media. The study was performed by the use of scanning electrochemical microscopy (SECM), which allowed obtaining information on uniformity, topography and chemical stability/reactivity of the alumina coatings with high spatial resolution. In particular, the effects due to local acid, base and fluoride ions attack on alumina layers of thickness of about 250 nm (in the Pt/Al₂O₃ sample) and 1000 nm (in the SS/Al₂O₃ sample) were investigated. In the acid and base attack, high concentrations of H₂SO₄ and KOH were electrogenerated locally by the use of a 25 μm diameter platinum microelectrode. The latter was also used as SECM tip to monitor the chemical effect on the alumina layers. It was found that, regardless of the thickness of the film, alumina provided good resistance against local attack of concentrated H₂SO₄; instead, the film dissolved when subjected to KOH attack. The dissolution rate depended on several experimental parameters, such as SECM-tip to substrate distance, electrolysis time and alumina film thickness. The alumina layer proved also relatively poor resistance to etching in 0.1 M NaF solutions.