Abstract. Machine Learning methods have been widely used in bioinformatics, mainly for data classification and pattern recognition. The detection of genes in DNA sequences is still an open problem. Identifying the promoter region laying prior the gene itself is an important aid to detect a gene. This paper aims at applying several Machine Learning methods to the construction of classifiers for detection of promoters in the DNA of Escherichia coli. A thorough comparison of methods was done. In general, probabilistic and neural network-based methods were those that performed better regarding accuracy rate.

The amount of data produced by the several genomic sequencing projects has increased dramatically in recent years. One of the main goals of bioinformatics is to analyze biological data aiming at identifying genes. The splice junction recognition problem is an important part of the gene detection problem. This work evaluates the performance of two classification models, derived from the Weight Matrix Model, when applied to the splice junction recognition problem. Two splice junction data sets were used in this work and some measures of predictive accuracy were reported. Based on the experiments, classification thresholds were established, which can be useful for further implementation of an automatic gene detection system.

No abstract

In this paper we study how the connectivity affects the performance of insular Parallel Genetic Algorithms (PGAs). Seven topologies PGAs were proposed, with growing number of connections. We used three instances of the well-known traveling salesman problem as benchmark. Each island of the PGA had different parameters and we established a fixed migration policy for all islands. Experiments were done and average results were reported. The effect of coevolution in PGAs was evidenced. The convergence time increased with the number of connections of the topology. The quality of solutions also increased in the same way. Although topologies with large connectivity increases the overall processing time, they take benefits to the quality of solutions found.

Background: This paper presents an electronic stethoscope model for cardiac auscultation, that adds innovative functionality to the conventional stethoscope, nominated "Electronic Stethoscope". Among the Electronic Stethoscope functionalities, can be cited the volume adjust in order to facilitate the hearing of the heart pulses, the graphic presentation of the heart beat sound waveforms, from a monitor, and the store of the cardiac auscultation data in a memory card for future analysis. It is believed that this additional functions, nonexistent in conventional stethoscopes, can contribute in the identification of pathology anomalies by the doctors and medicine students, growing the diagnoses trusting rate. Beyond that, the stored data in the equipment can be visualized and listened, without the necessity of the patient presence, characteristic that enables the build of a database, for example, of indicative signals of pathologies that could be used in class and practicing. With this equipment it is also possible to make adjusts on amplitude and length of the graphics (zoom), for a better details visualization. Undesirable ambient sounds can also be mitigated by the use of low-pass digital filters of IIR type implemented on the stethoscope software. This equipment has also a graphic resource for the identification of patient's heart rate similar of the gestational ultrasound equipment. Results of interviews with doctors and medicine students shows that the equipment have practical applicability, either in clinic or classroom, and extremely intuitive mode of operation, which would require no prior specific training.