Computing translocation distance by a genetic algorithm

Silveira, Lucas Ângelo; Soncco-Álvarez, José Luis; Lima, Thaynara Arielly de; Ayala-Rincón, Maurício

doi:10.1109/clei.2015.7359994

Cited by 8 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Silveira et al [27] proposed HoPIMs for a sequential GA introduced in [28] to solve the unsigned translocation problem. Such PIMs outperformed the accuracy obtained by the GA after careful calibration of the migration and breeding cycle parameters and exploration of a variety of topologies [29], [30].…”

Section: Related Workmentioning

confidence: 99%

Reconfigurable Heterogeneous Parallel Island Models

Silveira¹,

Lima²,

Ayala-Rincón³

2022

Preprint

View full text Add to dashboard Cite

Heterogeneous Parallel Island Models (HePIMs) run different bio-inspired algorithms (BAs) in their islands. From a variety of communication topologies and migration policies finetuned for homogeneous PIMs (HoPIMs), which run the same BA in all their islands, previous work introduced HePIMs that provided competitive quality solutions regarding the best-adapted BA in HoPIMs. This work goes a step forward, maintaining the population diversity provided by HePIMs, and increasing their flexibility, allowing BA reconfiguration on islands during execution: according to their performance, islands may substitute their BAs dynamically during the evolutionary process. Experiments with the introduced architectures (RecHePIMs) were applied to the NP-hard problem of sorting permutations by reversals, using four different BAs, namely, simple Genetic Algorithm (GA), Double-point crossover Genetic Algorithm (GAD), Differential Evolution (DE), and self-adjusting Particle Swarm Optimization (PSO). The results showed that the new reconfigurable heterogeneous models compute better quality solutions than the HePIMs closing the gap with the HoPIM running the best-adapted BA.

show abstract

Section: Related Workmentioning

confidence: 99%

Reconfigurable Heterogeneous Parallel Island Models

Silveira¹,

Lima²,

Ayala-Rincón³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Como contribuição, neste trabalho é apresentada a compilação da prova de que o problema de distância de translocação entre genomas sem sinal (DTS ) é N P-difícil expondo todos os passos necessários para fornecer uma prova detalhada bem como um conjunto apropriado de algoritmos genéticos (AGs) para resolver o problema. O primeiro algoritmo evolucionário denotado como AG (publicado em [42]), proposto para a resolução do problema DTS, tem como enfoque mapear um dado genoma sem sinal de tamanho n fornecido como entrada (com n representado a quantidade de genes) em um subconjunto das 2 n possíveis versões com sinal. Isto é feito assinalando aleatoriamente um sinal positivo ou negativo para cada gene do genoma fornecido na entrada.…”

Section: Motivaçãounclassified

“…Nos primeiros ciclos da pequisa onde foram propostos algoritmos genéticos para lidar com o problema DTS ( [42] e [43]) não nos preocupamos com o tempo de execução, e sim na precisão das soluções fornecidas por estes algoritmos. Esta foi a principal razão para propor a implementação do Algoritmo 12 (AM F ).…”

Section: Algoritmos Paralelosunclassified

See 1 more Smart Citation

Ordenação por translocação de genomas sem sinal utilizando algoritmos genéticos

Silveira¹

View full text Add to dashboard Cite

Translocations are used to measure the evolutionary distance between species. From a biological point of view two types of genomes have received attention: signed and unsigned genomes. When considering signed genomes, the problem can be solved in linear time, while, in the case of unsigned genomes the problem was shown to be N P-hard.Genetic algorithms (GAs) are proposed to solve the translocation distance problem between unsigned genomes. The approach consists in using a population composed of individuals representing signed genomes obtained from a given unsigned genome provided as input. The solution of each individual is also an admissible solution to the given genome. The fitness function used, which is the distance for signed genome, is computed linearly with an algorithm proposed by Bergeron et al.The GA based on this approach has been enhanced with two optimization techniques: memetic and opposition based learning. Also, parallelizations of the GA embedded with memetic were proposed seeking to improve both running time as the accuracy of results.The quality of the results was verified using an implementation of a 1.5+ε-approximation algorithm recently proposed by Cui et al.Experiments were performed taking as input synthetic genomes and genomes generated from biological data. The GAs provided better results than the quality control algorithm. The parallelizations showed improvements both regarding runtime as well as accuracy.A statistical analysis based on the Wilcoxon test was performed to check if the improvements in the solutions provided by enhanced GAs compared to those provided by the basic GA have some significance. This analysis can identify that the GA embedded with the technical memetic provides different (better) results than GA and that the results provided by the GA embedded with opposition based learning presents no significant difference. The test was also performed to compare the solutions of the parallelizations confirming that there are improvements of the results regarding the GA embedded memetic.

show abstract