Bioautomation

Volume 17, Issue 4
Editorial
Anniversary
The 60th Birthday of Professor Ilza Pajeva
The 65th Birthday of Associated Professor Ivan Simeonov
In memoriam
Professor Ivan Daskalov (1933-2013)
New Books
Bioinformatics
Generalized Net Model of the Cognitive and Neural Algorithm for Adaptive Resonance Theory 1	207-216
Todor Petkov, Sotir Sotirov [ +/- abstract ][ full text ]
The artificial neural networks are inspired by biological properties of human and animal brains. One of the neural networks type is called ART [4]. The abbreviation of ART stands for Adaptive Resonance Theory that has been invented by Stephen Grossberg in 1976 [5]. ART represents a family of Neural Networks. It is a cognitive and neural theory that describes how the brain autonomously learns to categorize, recognize and predict objects and events in the changing world. In this paper we introduce a GN model that represent ART1 Neural Network learning algorithm [1]. The purpose of this model is to explain when the input vector will be clustered or rejected among all nodes by the network. It can also be used for explanation and optimization of ART1 learning algorithm.
In silico Approach in the Prediction and Analysis of the Three-dimensional Structure of Maleylacetate reductase: A Biodegrading Protein	217-226
Shasank Sekhar Swain [ +/- abstract ][ full text ]
With the advent of biological research in the field of environmental science, several microbes were found to act as the most important biodegradable molecules. Maleylacetate reductase being a member of oxidoreductase is mostly found in Pseudomonas species. This enzyme participates in three metabolic pathways: gamma-hexachlorocyclohexane degradation, benzoate degradation which are higher alkyl compounds involved in chemical pollutions in metropolitan cities. Determining its three-dimensional structure will lead to the structure function analysis and also might be helpful for designing receptors for the degradation of new chemical compounds. For this purpose, the amino acid sequence of the protein had been imported from UniProtKB database and the template was searched in BLASTp2.2.27. This template was then used for the prediction of the three dimensional structure of the protein by using SWISS-MODEL Workspace. The predicted model was then validated using SAVES server which gave an almost result for better prediction.
Identification of Critical Target Protein for Cystic Fibrosis using Systems Biology Network Approach	227-240
Tammanna R. Sahrawat, Sherry Bhalla [ +/- abstract ][ full text ]
The most critical step in drug discovery process is target identification. Loss of function disease such as cystic fibrosis is caused by impairment of one protein and target identification of such diseases becomes a more tedious task. Protein interactions can provide an insight of other proteins which could be targeted to affect the regulation of particular protein involved in disease. The present study was taken to identify the critical protein which could be targeted to affect the regulation of CFTR (Cystic Fibrosis Transmembrane Regulator) which is mutated in Cystic Fibrosis using a systems biology network approach involving STRING and ClusterONE plugin of Cytoscape. Five proteins namely PDZK1, SLC9A3R1, CFTR, CANX and HSPA8 were identified to be the critical proteins which could affect the regulation of CFTR. Calnexin (CANX) and HSPA8/HSC70 were also present in Cluster 2 obtained by ClusterONE which contains proteins responsible for degradation of misfolded protein. Finally HSPA8/HSC70 was selected to be a probable critical target protein as it has been reported that abolishing the interaction of F508del-CFTR with calnexin (CAS treatment) has no major (positive or negative) effect on ERAD of F508del-CFTR. Thus, systems biology approach may hold great promise to identify probable therapeutic targets.
Biomedical systems
On the Study of Indomethacin Release from Pnipam-g-peo Vesicular Nanoparticles	241-248
Victoria Michailova, Rumiana Blagoeva, Assen Nedev [ +/- abstract ][ full text ]
A study of indomethacin release from poly(N-isopropylacrylamide)-graft-poly(ethylene oxide) PNIPAM-g-PEO vesicular nanoparticles was performed. In vitro release studies were conducted in distilled water containing 10% and 20% (v/v) ethanol added with different rate and using the dialysis tube method. The nano-sized polymeric vesicles were characterized with respect to morphology, drug loading content and in vitro drug release kinetics. A mathematical model for indomethacin release, recently proposed by the authors, was validated under the obtained experimental results. It was used for numerical simulation of indomethacin release from nanoparticles of the considered vesicular type in the solution, neglecting the presence of a membrane, within a period of 24 hours.
A Hybrid Gene Selection Method for Multi-category Tumor Classification using Microarray Data	249-258
Xiaobo Li, Huijuan Lu, Mingjun Wang [ +/- abstract ][ full text ]
Microarray technology allows molecular classification of tumors and identification of tumor markers, and it has been used widely in the field of cancer research. Although the problem of binary tumor classification has been addressed extensively, it lacks in-depth research on multi-category tumor classification. In this paper, informative gene selection method, which is a critical step of multi-category tumor classification, was studied. We present a hybrid gene selection strategy aiming to take advantage from the combination of different gene selection algorithms. Top ranked genes of Chi-squared and SVM-RFE algorithms are fused to generate a gene pool, and a genetic algorithm further explores the search space for reduced gene subsets. We tested the proposed model on the multi-category lung cancer microarray gene expression data set. Compared with each individual gene selection algorithm, our hybrid model was able to obtain highest classification performance with much smaller sized subsets of informative genes.