Applying Memetic Algorithms to the Analysis of Microarray Data

Genes Related with Alzheimer’s Disease: A Comparison of Evolutionary Search, Statistical and Integer Programming Approaches

“…3. (a) A clustering using the memetic algorithm of [1] of the union of all the genes reported in Fig. 2.…”

“…For 29 of them, we have found that there is a microarray data study in the public domain that contains their gene expression in control and Alzheimer's affected brains. A visual inspection of the relative gene expression of this dataset (containing approximately 2,000 genes), clustered with our memetic algorithm [1], clearly shows a pattern of differentiated gene expression in healthy and Alzheimer's affected brains (see Fig. 1).…”

“…2 and have been obtained using a singular-value decomposition approach [2]. In all figures, we provide a high-quality clustering of the gene expression patterns using the memetic algorithm described in [1].…”

Gene Ordering in Microarray Data Using Parallel Memetic Algorithms

“…Experiments focusing on the effect of the window size indicate that a size of window between s w = 5% and s w = 10% of the instance size is a good tradeoff between solution quality and computational cost [7]. It should be also emphasized that although the use of the 'moving windows' function was a major breakthrough, the correct adjustment of the MA also played an important role.…”

“…We adopted the restriction that recombination can only be made between a leader and one of its supporters within the same cluster. The recombination procedure (the Prune-Delete-Graft operator, check [7]) thus selects any leader uniformly at random and then it chooses -also uniformly at random -one of the three supporters. Indirectly, this recombination is also fitness-biased, since individuals situated at the upper nodes are better than the ones at the lower nodes.…”