Welcome to the webpage of the Research Group on Computational Intelligence and Bioinformatics. !!!
CIB is a team of people interested in Computational Intelligence as a general set of tools, and Bioinformatics as its main field of application. Although, we have been working together for several years in different research projects, we have just recently constitute this research group.
Our main research interests are:
- Automatic design of neural network architectures. We have worked in different evolutionary computation based methods for the automatic design of neural networks. We have used cooperative coevolution, genetic algorithms and artificial immune systems. We are also working in constructive algorithms based on evolutionary computation methods.
- Crossover operator in real-coded genetic algorithms. We are specially interested in this operator, as it is the most innovative within the field of evolutionary computation. We have study the effect of this operator on real-coded genetic algorithms for the optimization of functions and proposed a new approach for mating individuals based on confidence intervals.
- Classification. Different aspects of the classification problem have been the subject of our research. We work in ensembles of classifiers, from both classical and evolutionary points of view, output coding, and classification by means of evolutionary response surfaces. This one of most active research lines of the group.
- Multi-objective optimisation. We have used evolutionary multiobjective methods as a tool form several years. Now, we are developing an Artificial Immune System based on Immune Network Theory for multiobjective optimisation.
And, of course, we are working in Bioinformatics. In this area we are focused in two main subjects:
- Gene structure prediction. Gene recognition is our main objective within the field of Bioinformatics. We are working on introducing evolutionary computation techniques to solve this problem. Recently, this research line has been funded with a project in the Excellence Research Projects' Program of the local government of Andalusia.
- Translation initiation site (TIS) prediction. As this can be considered as a first step in gene recognition we are working on applying different classification algorithms to this task.
- DNA separation of different organisms. In several tasks, especially in the study of plant/pathogen interactions, there is a need to separate the DNA of two species. In this problem we are developing models based on Hidden Markov Models and one-class methods.
We are open to any collaboration with other groups and researchers. If you are interested in our work do not hesitate to contact us in:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
.
As a final word we must say that we are highly committed to Open Source. All our research is done under GNU/Linux machines and using only Open Source tools. Of course, all the code we develop is licensed under GPL. This software will be soon available on this web. In the meantime you can request it to us by e-mail and we will sent it to you.
Greetings and enjoy our webpage.
