Enrique Bravo

A non-deterministic grammar inference algorithm applied to the cleavage site prediction problem in bioinformatics

We report results on applying the OIL (Order Independent Language) grammar inference algorithm to predict cleavage sites in polyproteins from translation of Potivirus genome. This nondeterministic algorithm is used to generate a group of models which vote to predict the occurrence of the pattern. We built nine models, one for each cleavage site in this kind of virus genome and report sensibility, specificity, accuracy for each model. Our results show that this technique is useful to predict cleavage sites in the given task with accuracy rates higher than 95%.

False Positive Reduction in Automatic Segmentation System

An application has been developed for automatic segmentation of Potyvirus polyproteins through stochastic models of Pattern Recognition. These models usually find the correct location of the cleavage site but also suggest other possible locations called false positives. For reducing the number of false positives, we evaluated three methods. The first is to shrink the search range skipping portions of polyprotein with low probability of containing the cleavage site. In the second and third approach, we use a measure to rank candidate locations in order to maximize the ranking of the correct cleavage site. Here we evaluate probability emitted by Hidden Markov Models (HMM) and Minimum Editing Distance (MED) as measure alternatives. Our results indicate that HMM probability is a better quality measure of a candidate location than MED. This probability is useful to eliminate most of false positive. Besides, it allows to quantify the quality of an automatic segmentation.

Machine Learning Techniques Applied to the Cleavage Site Prediction Problem

The Genome of the Potyviridae virus family is usually expressed as a polyprotein which can be divided into ten proteins through the action of enzymes or proteases which cut the chain in specific places called cleavage sites. Three different techniques were employed to model each cleavage site: Hidden Markov Models (HMM), grammatical inference OIL algorithm (OIL), and Artificial Neural Networks (ANN).

Automatic segmentation of Potyviridae family polyproteins

We describe an automatic segmentation method for polyproteins of the viruses belonging to the Potyviridae family. It uses machine learning techniques in order to predict the cleavage site which define the segments in which said polyproteins are cut in their process of functional maturation. The segmentation application is publicly available for use on a website and it can be accessed through the web service interface too. The prediction models have an average sensitivity of 0.79 and a Matthews correlation coefficient average of 0.23. This method is capable of predicting correctly (coinciding with previously published segmentation) the segmentation of sequences which come from Potyvirus and Rymovirus, genera. However accurate prediction capabilities are affected when faced with either atypical sequences or viruses belonging to less common genera in the Potyviridae family. Future work will focus on establishing greater flexibility in this sense.

Extracción y purificación de la enzima monóxido de carbono deshidrogenasa a partir de la bacteria carboxidótrofa Oligotropha carboxidovorans

This paper reports the extraction and purification of the carbon monoxide dehydrogenase enzyme from the Oligotropha carboxidovorans bacteria, OM 5 strain. The bacteria grew autotrophically in a mineral medium, in carbon monoxide saturated atmosphere, at 30°C with constant agitation. In order to obtain the crude enzymatic extracts, the crops underwent a breakdown by sonication, solvolysis with non-ionic detergent and centrifugation to eliminate the cellular debris. During the purification process, the crude enzymatic extracts were passed through a molecular exclusion column as well as through an ion-exchange one.