Evaluation of genome similarities using the non-decimated wavelet transform

The wavelets have become increasingly popular in the field of bioinformatics due to their capacity in multiresolution analysis and space-frequency localization; the latter particularity is acquired due to a moving window that runs through the analyzed space. As a feature, they have a better ability to capture hidden components of biological data and an efficient link between biological systems and the mathematical objects used to describe them. The decomposition of signals/sequences at different levels of resolution allows obtaining distinct characteristics in each level. The energy (variance) obtained at each level provides a new set of information that can be used to search similarities between sequences. We show that the behavior of GC-content sequence can be succinctly described regarding the non-decimated wavelet transform, and we indicate how this characterization can be used to improve clustering of the similar strains of the genome of the Mycobacterium tuberculosis, having a very efficient level of detail. The clustering analysis using the energy obtained at each level of the analyzed sequences was essential to verify the dissimilarity of the sequences.

Bayesian forecasting of temporal gene expression by using an autoregressive panel data approach

We propose and evaluate a novel approach for forecasting gene expression over non-observed times in longitudinal trials under a Bayesian viewpoint. One of the aims is to cluster genes that share similar expression patterns over time and then use this similarity to predict relative expression at time points of interest. Expression values of 106 genes expressed during the cell cycle of Saccharomyces cerevisiae were used and genes were partitioned into five distinct clusters of sizes 33, 32, 21, 16, and 4. After removing the last observed time point, the agreements of signals (upregulated or downregulated) considering the predicted expression level were 72.7, 81.3, 76.2, 68.8, and 50.0%, respectively, for each cluster. The percentage of credibility intervals that contained the true values of gene expression for a future time was ~90%. The methodology performed well, providing a valid forecast of gene expression values by fitting an autoregressive panel data model. This approach is easily implemented with other time-series models and when Poisson and negative binomial probability distributions are assumed for the gene expression data.

Previsão Bayesiana de valores genéticos de touros por meio do modelo auto-regressivo para dados em painel

Inferência Bayesiana do modelo auto-regressivo foi aplicado para dados em painel a dados reais de DEP´s de touros da raça Nelore publicadas entre 2000 e 2005. Na análise foram consideradas a função de verossimilhança exata e a obtenção de distribuições preditivas de dados futuros. Verificou-se ser recomendável o agrupamento dos animais em grupos homogêneos de acordo com a acurácia. Constatou-se também, que em média, a eficiência de previsão dos valores de DEP´s para um ano futuro é próxima de 80%.