Octavian Arsene

Development of membrane controllers for mobile robots

The main contribution of this paper is the introduction of the new concept of membrane controller based on the structure and functioning of a deterministic numerical P system. The procedure for developing a membrane controller and for using it to control a mobile robot is explained and several test cases are given in which membrane controllers are used to control both simulated and real mobile robots and to generate various desired behaviours (obstacle avoidance, wall following, and follow the leader). The experiments reported in this paper validate the concept and prove that the performance of a membrane controller is comparable to or better than that of other controllers (such as fuzzy logic controllers).

Enzymatic numerical P systems - a new class of membrane computing systems

A P system represents a distributed and parallel computing model in which basic data structures are multi-sets, strings or numerical variables. Numerical P systems have been introduced for possible applications in economics. A simulator for numerical P systems (SNUPS) has been designed, implemented and made available to the scientific community by the authors of this paper. SNUPS allows the development of a wide range of applications, from modeling and simulation of ordinary differential equations, to design and simulation of computational blocks for cognitive architectures and of membrane controllers for autonomous mobile robots. This paper introduces a new class of membrane computing systems, that of enzymatic numerical P systems, in which enzyme-like variables allow the existence of more than one production function in each membrane. The way this new type of deterministic numerical P systems works and the corresponding implementation in SNUPS are detailed, together with an illustrative example.

Medicine expert system dynamic Bayesian Network and ontology based

The paper proposes an application framework to be used for medicine assisted diagnosis based on ontology and Bayesian Network (DBNO). There are two goals: (1) to separate the domain knowledge from the probabilistic information and (2) to create an intuitive user interface. The framework architecture has three layers: knowledge, uncertainty model and user interface. The contributions of the domain experts are decoupled, the ontology builder will create the domain concepts and relationships focusing on the domain knowledge only. The uncertainty model is Bayesian Network and the probabilities of the variables states are stored in a profile repository. The diagnostician will use the user interface feeded with the domain ontology and one uncertainty profile. The application was tested on a sample medicine model for the diagnose of heart disease.

Expert system for medicine diagnosis using software agents

In order to simplify the information exchange within the medical diagnosis process, a collaborative software agents framework is presented.The human body systems (e.g. respiratory, cardiovascular) are embedded into distinct software agents.The automated information exchange between different medicine specialists from different areas of expertise.The framework has three key components: knowledge management, uncertainty reasoning and software agents. In order to simplify the information exchange within the medical diagnosis process, a collaborative software agents framework is presented. The human body systems (e.g. respiratory, cardiovascular) are embedded into distinct software agents. The holistic perspective is given by the all connected agents exchanging information. The purpose of the framework is to allow the automated information exchange between different medicine specialists. The key factor of the exchange is sharing concepts between the areas of expertise. Each human body system expert will act over his concepts (evidences, causes, effects), however the information from other systems will be assimilated as well. The framework has three key components: knowledge management, uncertainty reasoning and software agents. The ontology is chosen to address the management of human body systems knowledge. The Bayesian Network is the graphical model for probabilistic knowledge representation and reasoning about partial beliefs under uncertainty. The software agents, as collaboration framework, are in charge of the belief propagation between system instances.

A software tool for modeling and simulation of numerical P systems

UNLABELLED A P system represents a distributed and parallel bio-inspired computing model in which basic data structures are multi-sets or strings. Numerical P systems have been recently introduced and they use numerical variables and local programs (or evolution rules), usually in a deterministic way. They may find interesting applications in areas such as computational biology, process control or robotics. The first simulator of numerical P systems (SNUPS) has been designed, implemented and made available to the scientific community by the authors of this paper. SNUPS allows a wide range of applications, from modeling and simulation of ordinary differential equations, to the use of membrane systems as computational blocks of cognitive architectures, and as controllers for autonomous mobile robots. This paper describes the functioning of a numerical P system and presents an overview of SNUPS capabilities together with an illustrative example. AVAILABILITY SNUPS is freely available to researchers as a standalone application and may be downloaded from a dedicated website, http://snups.ics.pub.ro/, which includes an user manual and sample membrane structures.

Modeling of biological nanostructured surfaces

The paper presents a methodology using atom or amino acid hydrophobicities to describe the surface properties of proteins in order to predict their interactions with other proteins and with artificial nanostructured surfaces. A standardized pattern is built around each surface atom of the protein for a radius depending on the molecule type and size. The atom neighborhood is characterized in terms of the hydrophobicity surface density. A clustering algorithm is used to classify the resulting patterns and to identify the possible interactions. The methodology has been implemented in a software package based on Java technology deployed in a Linux environment.

Functional nanoscale imaging of protein surfaces

The paper presents an image-oriented modality to functionally describe artificially and biologically nanostructured surfaces, which can be used for the characterization of the atom neighborhoods on the surface of proteins. The property which is mainly analyzed in this paper is the hydrophobicity distribution on protein surface, but the distributions of charges and mutual electrical potentials can also be considered. The actual discrete hydrophobicity distribution attached to the atoms that form a surface atoms vicinity is replaced by an approximately equivalent hydrophobicity density distribution, computed in a standardized octagonal pattern around each atom. These representation of hydrophobicities is used to compute the resemblance of surface atom neighborhoods belonging to a protein, defined as the sum of the products of hydrophobicity densities of the corresponding patches (the patterns central circles or angular sectors having the same position). The similitude and the interaction of a pair of atom neighborhoods are defined as their resemblance for parallel, respectively, anti-parallel orientations of the normals on the molecular surfaces in the points where the central atoms are located. The purpose of this work is to create a database of selected protein surfaces that will be used for nanotechnology research and applications purposes.

Multi-threading protein surface functional description

The paper presents an image-oriented description of artificial and biological nanostructured surfaces, with applicability to the functional characterization of atom neighborhoods at the surface of proteins. The property which is considered is the hydrophobicity around each surface atom. The actual hydrophobicity distribution on the atoms that form an atoms vicinity is replaced by an equivalent hydrophobicity density distribution, computed in a standardized hexagonal or octagonal pattern around the atom. The software implementation is a desktop multi-threading application, able to process a large number of atom properties, such as type, 3D coordinates, charge and hydrophobicity. The atoms at the surface of a molecule are divided among the execution threads and a feature vector is created for each of them. The purpose of this work is to create a database of molecular surfaces that will be used in several nanotechnology research fields.

Protein Surface Functional Imaging

The paper presents an image-oriented functional description of protein surfaces in terms of amphiphilicity (hydrophobicity / hydrophilicity) distribution. The actual discrete surface atom amphiphilicity distribution is replaced by an approximately equivalent amphiphilicity density distribution, computed in a standardized octagonal pattern around each atom. This representation is used to compute the resemblance of the neighborhoods of a pair of surface atoms – defined as the sum of the products of amphiphilicity densities of the corresponding patches (the patterns central circles and the angular sectors in the same positions) in the two neighborhoods. The similitude and the interaction of a pair of atom neighborhoods are defined as their resemblance for parallel, respectively, anti-parallel orientations of the unit vectors perpendicular on the molecular surfaces in the points where the central atoms are located. These parameters, as well as the vector description of the neighborhoods, are used for the functional classification of surface atoms and for the study of protein interactions.

Protein surface atom neighbourhoods classification

The paper presents a classification of the protein surface atom neighbourhoods from the hydrophobicity perspective. Hydrophobicity is the property which is considered around each surface atom. The actual hydrophobicity distribution on the atoms that form an atoms vicinity is replaced by an equivalent hydrophobicity density distribution, computed in a standardized octagonal pattern around the atom. All atoms hydrophobicity densities are clustered using K-means algorithm. A three layers neural network is trained for classification of the atoms vicinities having as many nodes in the output layers as clusters are.