China
Africa
Explore chapters and articles related to this topic
Evolutionary Computation
Published in Bogdan M. Wilamowski, J. David Irwin, Intelligent Systems, 2018
The selection, also named as reproduction, is a procedure of choosing given individuals from the population in order to create a new population in the next generation of evolutionary algorithm. The probability of selection of a given individual depends on its fitness value. When the given individual has a higher fitness value, then it possess higher chance to be selected to the new generation. The reproduction process is strictly connected with the two most important factors in evolutionary algorithms: preservation of the diversity of population and selection pressure. These factors are dependent on each other because increase of selection pressure causes decrease of population diversity (and inversely) [1]. Too high value of selection pressure (concentration of the search only on best individuals) leads to premature convergence, which is an undesirable effect in evolutionary algorithms, because the algorithm can stick in local extreme. However, too small value of selection pressure causes that search of solution space has almost random character. The main goal of the selection operators is the preservation of balance between these factors [2]. There exist many selection methods. The oldest one (most popular) is a proportional selection also named as a roulette selection. In this method, the probability of individual selection is proportional to the value of its fitness function [1]. For each individual, the sector size on roulette wheel is equal to the individual relative fitness (rfitness) value, that is, the fitness value divided by the sum of all fitness values GF (global fitness) in the population (see formula (29.6) and (29.7)). In Figure 29.4, an example of roulette wheel with scaled sectors for M = 5 individuals is presented.
IoE-Based Genetic Algorithms and Their Requisition
Published in Suhel Ahmad Khan, Rajeev Kumar, Omprakash Kaiwartya, Mohammad Faisal, Raees Ahmad Khan, Computational Intelligent Security in Wireless Communications, 2022
Neeraj Kumar Rathore, Shubhangi Pande
In regards to the evaluation function, the selection process randomly selects chromosomes within the population. Higher the fitness level, the greater the chance of being chosen as an individual. The extent to which the most appropriate individuals are preferred is termed as the selection pressure. If the selection pressure is high, it means more preference is granted to the best individuals.
Glossary of scientific and technical terms in bioengineering and biological engineering
Published in Megh R. Goyal, Scientific and Technical Terms in Bioengineering and Biological Engineering, 2018
Selection pressure is the intensity of selection acting on a population of organisms or cells in culture. Its effectiveness is measured in terms of differential survival and reproduction, and consequently in change in the frequency of alleles in a population.
Development of capability for genome-scale CRISPR-Cas9 knockout screens in New Zealand
Published in Journal of the Royal Society of New Zealand, 2018
Francis W. Hunter, Peter Tsai, Purvi M. Kakadia, Stefan K. Bohlander, Cristin G. Print, William R. Wilson
The cost-efficiency and technical utility of CRISPR-mediated gene knockout relative to methods involving protein-based targeting have enabled another powerful application of the technology—functional genomic screens (Shalem et al. 2015). This approach entails introducing complex libraries of sgRNAs—which can be designed against virtually every annotated gene using user-programmed or machine-learned algorithms and produced by massive arrayed synthesis—into Cas9-expressing, immortalised cells, typically by transduction with replication-deficient lentiviral vectors for efficient integration of the sgRNA-encoding sequence into the host cell genome. Selective pressure is applied to the resulting cellular library, often in the form of cytostatic/cytotoxic stress conditions or a therapeutic agent, to elicit differential survival of clones according to the phenotype of the mutation they carry. The mutations responsible for differential survival under selection, and thus the implicated genes, can be inferred by the enrichment or depletion of sgRNAs among genomic DNA samples extracted from the selected library relative to a non-selected control library, where sgRNA abundance is assessed using sequential polymerase chain reaction (PCR) amplification, next-generation sequencing and bioinformatic analysis (Figure 4). In this manner, CRISPR-mediated functional genomics allows for powerful, unbiased interrogation of genetic contributions to cell biological/physiological phenotypes, disease states, infection processes and responses to therapies such as drugs or radiation. A collection of sgRNA libraries intended for functional screens using gene knockout, activation or inhibition endpoints, or targeting alternative genetic elements such as ‘long non-coding’ (lnc)RNAs is now readily available to the international research community (Table 1).