Nieves Pedreira

Diversity and multimodal search with a hybrid two-population GA: an application to ANN development

Being based on the theory of evolution and natural selection, the Genetic Algorithms (GA) represent a technique that has been proved as good enough for the resolution of those problems that require a search through a complex space of possible solutions. The maintenance of a population of possible solutions that are in constant evolution may lead to its diversity being lost, consequently it would be more difficult, not only the achievement of a final solution but also the supply of more than one solution The method that is described here tries to overcome those difficulties by means of a modification in traditional GAs. Such modification involves the inclusion of an additional population that might avoid the mentioned loss of diversity of classical GAs. This new population would also provide the piece of exhaustive search that allows to provide more than one solution.

A computational morphogenesis approach to simple structure development

This paper presents a new model for computational embryology that mimics the behaviour of biological cells, whose characteristics can be applied to the solution of computational problems. The presented tests apply the model to simple structure generation and provide promising results with regard to its behaviour and applicability to more complex problems.

LECTINPred: web Server that Uses Complex Networks of Protein Structure for Prediction of Lectins with Potential Use as Cancer Biomarkers or in Parasite Vaccine Design.

Lectins (Ls) play an important role in many diseases such as different types of cancer, parasitic infections and other diseases. Interestingly, the Protein Data Bank (PDB) contains +3000 protein 3D structures with unknown function. Thus, we can in principle, discover new Ls mining non‐annotated structures from PDB or other sources. However, there are no general models to predict new biologically relevant Ls based on 3D chemical structures. We used the MARCH‐INSIDE software to calculate the Markov‐Shannon 3D electrostatic entropy parameters for the complex networks of protein structure of 2200 different protein 3D structures, including 1200 Ls. We have performed a Linear Discriminant Analysis (LDA) using these parameters as inputs in order to seek a new Quantitative Structure‐Activity Relationship (QSAR) model, which is able to discriminate 3D structure of Ls from other proteins. We implemented this predictor in the web server named LECTINPred, freely available at http://bio‐aims.udc.es/LECTINPred.php. This web server showed the following goodness‐of‐fit statistics: Sensitivity=96.7 % (for Ls), Specificity=87.6 % (non‐active proteins), and Accuracy=92.5 % (for all proteins), considering altogether both the training and external prediction series. In mode 2, users can carry out an automatic retrieval of protein structures from PDB. We illustrated the use of this server, in operation mode 1, performing a data mining of PDB. We predicted Ls scores for +2000 proteins with unknown function and selected the top‐scored ones as possible lectins. In operation mode 2, LECTINPred can also upload 3D structural models generated with structure‐prediction tools like LOMETS or PHYRE2. The new Ls are expected to be of relevance as cancer biomarkers or useful in parasite vaccine design.

Interpretation of geometric shapes: an eye movement study

This paper describes a study that seeks to explore the correlation between eye movements and the interpretation of geometric shapes. This study is intended to inform the development of an eye tracking interface for computational tools to support and enhance the natural interaction required in creative design. A common criticism of computational design tools is that they do not enable manipulation of designed shapes according to all perceived features. Instead the manipulations afforded are limited by formal structures of shapes. This research examines the potential for eye movement data to be used to recognise and make available for manipulation the perceived features in shapes. The objective of this study was to analyse eye movement data with the intention of recognising moments in which an interpretation of shape is made. Results suggest that fixation duration and saccade amplitude prove to be consistent indicators of shape interpretation.

Hybrid Two-Population Genetic Algorithm

Genetic Algorithms are non-deterministic, stochastic-search adaptive methods which use the theories of natural evolution and selection in order to solve a problem within a complex range of possible solutions. The aim is to control the distribution of the search space by incorporating an exhaustive method in order to maintain a constant evolution of the population. The main goal is that of redesigning the algorithm in order to add to the classic genetic algorithm method those characteristics which favour exhaustive search methods. The method explained guarantees the achievement of reasonably satisfactory solutions in short time-spans and in a deterministic way, which entails that successive repetitions of the algorithm will achieve the same solutions in almost constant time-spans. We are, therefore, dealing with an evolutionary technique which makes the most of the characteristics of genetic algorithms and exhaustive methods.

Knowledge management for chronic patient control and monitoring

Knowledge Management (KM) can be seen as the process of capturing, developing, sharing, and effectively using organizational knowledge. In this context, the work presented here proposes a KM System to be used in the scope of chronic patient control and monitoring for distributed research projects. It was designed in order to enable communication between patient and doctors, as well as to be usedbythe researchers involved in the project for its management. The proposed model integrates all the information concerning every patient and project management tasks in the Institutional Memory of a KMSystem and uses an ontology to maintain the information and its categorization independently. Furthermore, taking the philosophy of intelligent agents, the system will interact with the user to show him the information according to his preferences and access rights. Finally, three different scenarios of application are described.

Knowledge Management and Interactive Learning

This work presents a proposal for an e-learning model that facilitate the learning process. Using a Knowledge Management System as support, learning is planned through action, which results in the execution of tasks based on computer games strategies. Both contents of the course and tasks are included into the Institutional Memory of System. The design of this Institutional Memory shows as a basis for the system. Intelligent agents will adapt the tasks to the level and preferences of the student.

A model of virtual 'learning to learn'

Even though information and communications technology arises as a powerful tool for the improvement of education, at present, e-learning does not improve learning methods but merely repeats the problems of traditional education. This work proposes a new educational model that tries to tackle these problems. Through the use of a global ontology as basis of a knowledge management system, this system allows us to establish the largest number of relationships within the available information, and its classification. From this knowledge support, learning is planned through action, which results in the execution of tasks that are based on computer games strategies. The system interacts with the student to motivate him, showing him the information adapted to his preferences.