Catálogo de publicaciones - libros

Functionally related genes involved in the same molecular-genetic, biochemical, or physiological process are often regulated coordinately Such regulation is provided by precisely organized binding of a multiplicity of spe cial proteins- transcription factors to their target sites (cis-elements) in regulatory regions of genes. Cis-element combinations provide a structural basis for the generation of unique patterns of gene expression. Here we present new method based on genetic algorithm for prediction of class-specific composite modules in promoters of functionally related or coexpressed genes and it’s testing on simulated data.

A simulation program is a useful tool to predict the hybridization error propagated through a microfluidic system. This paper shows the hybridization event between solution strands and capture probes. The program provides a Graphical User Interface that allows the user to insert parameters as well as to observe results showing critical values and additional visualization of the simulation process. The program and its interface have been developed in Java, making it platform independent and web-based.

A novel adaptive ant colony clustering algorithm based on digraph (ACD) is presented. Inspired by the swarm intelligence shown through the social insects’ self-organizing behavior, in ACD we assign acceptance weights on the directed edges of a pheromone digraph. The weights of the digraph is adaptively updated by the pheromone left by ants in the seeking process. Finally, strong connected components are extracted as clusters under a certain threshold. ACD has been implemented and tested on several clustering benchmarks and real datasets to compare the performance with the classical K-means clustering algorithm and LF algorithm which is also based on ACO. Experimental results show that our algorithm is easier to implement, more efficient and performs faster and has better clustering quality than other methods.

Ant Colony optimization takes inspiration from the behavior of real ant colony to solve optimization problems. This paper presents a parallel model for ant colony to solve the optimal packing problem. The problem is represented by a directed graph so that the objective of the original problem becomes to find the shortest closed circuit on the graph under the problem-specific constraints. A number of artificial ants are distributed on the graph and communicate with one another through the pheromone trails which are a form of the long-term memory guiding the future exploration of the graph. The algorithm supports the parallel computation and facilitates quick convergence to the optimal solution. The performance of the proposed method as compared to those of the genetic-based approaches is very promising.

An adaptive parallel ant colony optimization is presented by improving the critical factor influencing the performance of the parallel algorithm. We propose two different strategies for information exchange between processors: selection based on sorting and on difference, which make each processor choose another processor to communicate and update the pheromone adaptively. In order to increase the ability of search and avoid early convergence, we also propose a method of adjusting the time interval of information exchange adaptively according to the diversity of the solutions. These techniques are applied to the traveling salesman problem on the massive parallel processors (MPP) Dawn 2000. Experimental results show that our algorithm has high convergence speed, high speedup and efficiency.

This paper presents a new method for hierarchical image segmentation. The hierarchical structure is represented by a binary tree with the main image as its root. At the lower levels, each node stands as one image segment, which is described by a weighted graph and may be divided into two new segments at the next level through a specific cut. Graph bi-sectioning is done by the self organizing property of ant systems. Ants are free to wander over one image segment graph to find the best cut on it. When an ant finds a suitable cut, it returns to its colony and leaves a proper value of pheromone over its trail to attract other ants to that cut. By using the Chemical Computing approach in this paper, it is assumed the mobile hormones (pheromone) are secreted which can diffuse around initial positions and attract more ants to the found cut. The advantages of this assumption are reducing the noise effects and improving the convergence speed of ants to find a new selected image segment, which can be seen in the practical results.

Many algorithms that solve optimization problems are being developed and used. However, large and complex optimization problems still exist, and it is often difficult to obtain the desired results with one of these algorithms alone. This paper applies tabu search and ant colony optimization method to the container load sequencing problem. We also propose a hybrid algorithm, which can combine the merits of these two algorithms by running them alternately. Experiments have shown that the proposed hybrid algorithm is superior to both tabu search and ant colony optimization individually.

By applying a candidate set strategy based on minimum 1-tree and a self-adaptive hybrid mutation operator to the ant colony system, a novel ant colony system for TSP (MMACS) is proposed. Under the condition that all the edges in the global optimal tour are nearly all contained in the candidate sets, the candidate set strategy based on minimum 1-tree can limit the selection scope of ants at each step to six cities and thus substantially reduce the size of search space. Meanwhile, the self-adaptive hybrid mutation operator that consists of inversion mutation, insertion mutation and swap mutation can effectively prevent MMACS from being trapped in local optimal areas. The simulation of TSP shows that MMACS can avoid the premature convergence phenomenon effectively while greatly increasing the convergence speed. Although MMACS takes TSP as an example for explaining its mechanism, its ideas can be used for other related algorithms.