Efficient feature selection filters for high-dimensional data

Feature selection is a central problem in machine learning and pattern recognition. On large datasets (in terms of dimension and/or number of instances), using search-based or wrapper techniques can be cornputationally prohibitive. Moreover, many filter methods based on relevance/redundancy assessment also take a prohibitively long time on high-dimensional. datasets. In this paper, we propose efficient unsupervised and supervised feature selection/ranking filters for high-dimensional datasets. These methods use low-complexity relevance and redundancy criteria, applicable to supervised, semi-supervised, and unsupervised learning, being able to act as pre-processors for computationally intensive methods to focus their attention on smaller subsets of promising features. The experimental results, with up to 10(5) features, show the time efficiency of our methods, with lower generalization error than state-of-the-art techniques, while being dramatically simpler and faster.

Protein-Polysaccharides of trametes versocolor: production and biological activities

Extracellular-(E-PPS) and intracellular-protein-polysaccharides (I-PPS) complexes were produced by Trametes versicolor in submerged cultures with different carbon sources. The highest extracellular-(EPS) and intracellular-polysaccharide (IPS) concentration in the complexes was obtained with tomato pomace culture. DPPH radical scavenging for E-PPS and I-PPS produced by liter of culture was equivallent to 2.115 +/- A 0.227 and 1.374 +/- A 0.364 g of ascorbic acid, respectively. These complexes showed a protector effect in the oxidation of erythrocyte membranes and had ability to inhibit the hemolysis and methemoglobin synthesis in stressed erythrocytes. These results suggest that extracellular- and intracellular- polysaccharides produced are important bioactive compounds with medicinal potential.

Production of recombinant protein hLIF in static and dynamic systems of animal cell culture

Thesis for the Degree of Master of Science in Biotechnology Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

Integrating protein structural information

Dissertação apresentada para obtenção de Grau de Doutor em Bioquímica,Bioquímica Estrutural, pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

Protein folding, metal ions and conformational states: the case of a di-cluster ferredoxin

Dissertation presented to obtain the PhD degree in Biochemistry at the Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa

Phenylketonuria: Protein content and amino acids profile of dishes for phenylketonuric patients. The relevance of phenylalanine

Phenylketonuria is an inborn error of metabolism, involving, in most cases, a deficient activity of phenylalanine hydroxylase. Neonatal diagnosis and a prompt special diet (low phenylalanine and natural-protein restricted diets) are essential to the treatment. The lack of data concerning phenylalanine contents of processed foodstuffs is an additional limitation for an already very restrictive diet. Our goals were to quantify protein (Kjeldahl method) and amino acid (18) content (HPLC/fluorescence) in 16 dishes specifically conceived for phenylketonuric patients, and compare the most relevant results with those of several international food composition databases. As might be expected, all the meals contained low protein levels (0.67–3.15 g/100 g) with the highest ones occurring in boiled rice and potatoes. These foods also contained the highest amounts of phenylalanine (158.51 and 62.65 mg/100 g, respectively). In contrast to the other amino acids, it was possible to predict phenylalanine content based on protein alone. Slight deviations were observed when comparing results with the different food composition databases.

Nutritional composition of low protein and phenylalanine-restricted dishes prepared for phenylketonuric patients

Nutritional management is essential for Phenylketonuria (PKU) treatment, consisting in a semi-synthetic and low phenylalanine (Phe) diet, which includes strictly controlled amounts of low protein natural foods (essentially fruits and vegetables) supplemented with Phe-free protein substitutes and dietetic low-protein products. PKU diet has to be carefully planned, providing the best ingredient combinations, so that patients can achieve good metabolic control and an adequate nutritional status. Hereupon, it is mandatory to know the detailed composition of natural and/or cooked foodstuffs prepared specifically for these patients. We intended to evaluate sixteen dishes specifically prepared for PKU patients, regarding the nutritional composition, Phe and tyrosine (Tyr) contents, fatty acids profile, and vitamins E and B12 amounts. The nutritional composition of the cooked samples was 15.5–92.0 g/100 g, for moisture; 0.7–3.2 g/100 g, for protein; 0.1–25.0 g/100 g, for total fat; and 5.0–62.0 g/100 g, for total carbohydrates. Fatty acids profile and vitamin E amount reflected the type of fat used. All samples were poor in vitamin B12 (0.3–0.8 μg/100 g). Boiled rice presented the highest Phe content: 50.3 mg/g of protein. These data allow a more accurate calculation of the diet portions to be ingested by the patients according to their individual tolerance.

Non-enzymatic assay for glucose by using immobilized whole-cells of E. coli containing glucose binding protein fused to fluorescent proteins

Glucose monitoring in vivo is a crucial issue for gaining new understanding of diabetes. Glucose binding protein (GBP) fused to two fluorescent indicator proteins (FLIP) was used in the present study such as FLIP-glu- 3.2 mM. Recombinant Escherichia coli whole-cells containing genetically encoded nanosensors as well as cell-free extracts were immobilized either on inner epidermis of onion bulb scale or on 96-well microtiter plates in the presence of glutaraldehyde. Glucose monitoring was carried out by Förster Resonance Energy Transfer (FRET) analysis due the cyano and yellow fluorescent proteins (ECFP and EYFP) immobilized in both these supports. The recovery of these immobilized FLIP nanosensors compared with the free whole-cells and cell-free extract was in the range of 50–90%. Moreover, the data revealed that these FLIP nanosensors can be immobilized in such solid supports with retention of their biological activity. Glucose assay was devised by FRET analysis by using these nanosensors in real samples which detected glucose in the linear range of 0–24 mM with a limit of detection of 0.11 mM glucose. On the other hand, storage and operational stability studies revealed that they are very stable and can be re-used several times (i.e. at least 20 times) without any significant loss of FRET signal. To author's knowledge, this is the first report on the use of such immobilization supports for whole-cells and cell-free extract containing FLIP nanosensor for glucose assay. On the other hand, this is a novel and cheap high throughput method for glucose assay.

Robot visual localization through local feature fusion: an evaluation of multiple classifiers combination approaches

In the last decade, local image features have been widely used in robot visual localization. In order to assess image similarity, a strategy exploiting these features compares raw descriptors extracted from the current image with those in the models of places. This paper addresses the ensuing step in this process, where a combining function must be used to aggregate results and assign each place a score. Casting the problem in the multiple classifier systems framework, in this paper we compare several candidate combiners with respect to their performance in the visual localization task. For this evaluation, we selected the most popular methods in the class of non-trained combiners, namely the sum rule and product rule. A deeper insight into the potential of these combiners is provided through a discriminativity analysis involving the algebraic rules and two extensions of these methods: the threshold, as well as the weighted modifications. In addition, a voting method, previously used in robot visual localization, is assessed. Furthermore, we address the process of constructing a model of the environment by describing how the model granularity impacts upon performance. All combiners are tested on a visual localization task, carried out on a public dataset. It is experimentally demonstrated that the sum rule extensions globally achieve the best performance, confirming the general agreement on the robustness of this rule in other classification problems. The voting method, whilst competitive with the product rule in its standard form, is shown to be outperformed by its modified versions.

Feature Discretization with Relevance and Mutual Information Criteria

Feature discretization (FD) techniques often yield adequate and compact representations of the data, suitable for machine learning and pattern recognition problems. These representations usually decrease the training time, yielding higher classification accuracy while allowing for humans to better understand and visualize the data, as compared to the use of the original features. This paper proposes two new FD techniques. The first one is based on the well-known Linde-Buzo-Gray quantization algorithm, coupled with a relevance criterion, being able perform unsupervised, supervised, or semi-supervised discretization. The second technique works in supervised mode, being based on the maximization of the mutual information between each discrete feature and the class label. Our experimental results on standard benchmark datasets show that these techniques scale up to high-dimensional data, attaining in many cases better accuracy than existing unsupervised and supervised FD approaches, while using fewer discretization intervals.

Exploiting the bin-class histograms for feature selection on discrete data

In machine learning and pattern recognition tasks, the use of feature discretization techniques may have several advantages. The discretized features may hold enough information for the learning task at hand, while ignoring minor fluctuations that are irrelevant or harmful for that task. The discretized features have more compact representations that may yield both better accuracy and lower training time, as compared to the use of the original features. However, in many cases, mainly with medium and high-dimensional data, the large number of features usually implies that there is some redundancy among them. Thus, we may further apply feature selection (FS) techniques on the discrete data, keeping the most relevant features, while discarding the irrelevant and redundant ones. In this paper, we propose relevance and redundancy criteria for supervised feature selection techniques on discrete data. These criteria are applied to the bin-class histograms of the discrete features. The experimental results, on public benchmark data, show that the proposed criteria can achieve better accuracy than widely used relevance and redundancy criteria, such as mutual information and the Fisher ratio.

Schistosoma mansoni circulating polysaccharide and protein antigens recognized by sheep antisera in patients with different clinical forms of schistosomiasis before and after treatment

Two sheep antisera, one of which raised against polysaccharide (Po) and other against protein (Pt) components of Schistosoma mansoni adult worms, were assessed by ELISA for their ability to detect circulating parasite antigens in patients with different clinical forms of chronic schistosomiasis mansoni. The former antiserum detected parasite antigens in liver granulomata and the latter in renal glomeruli from schistosomiasis patients and mice experimentally infected with S. mansoni. In general, the levels and/or positivity rate of circulating antigens and specific IgG antibodies were significantly higher in patients with hepatointestinal (HI) and hepatosplenic (HS) forms than in mild intestinal (I) forms. An association between Po antigens and clinical features of the disease was observed, as the level of these antigens was low (137 ng/ml) as well as the positivity rate (7.9%) in patients with I forms; values that were intermediate (593 ng/ml and 33.3%) in those with HI forms, and high (1.563 ng/ml and 50.0%) in more severe HS forms. The Pt antigens were detected in the studied clinical forms not differing statistically but, the positivity rate was significantly higher in HS forms comparatively to I forms. The antisera studied revealed distinct circulating antigen profiles, and the prognostic value of Po and Pt antigens was suggested.

The web feature service: transaction standard: an example of prototype implementation

Trabalho de Projecto apresentado como requisito parcial para obtenção do grau de Mestre em Ciência e Sistemas de Informação Geográfica

Defective protein folding and function in metabolic disorders: studies on the mitochondrial flavoenzyme ETF

Dissertation presented to obtain the PhD degree in Biochemistry at the Instituto de Tecnologia Química e Biológica, Universidade Nova de Lisboa