Recommendations on the Implementation of Genetic Algorithms for the Directed Evolution of Enzymes for Industrial Purposes

Abstract

In directed evolution (DE) the assessment of candidate enzymes and their modification is essential. In this study we have investigated genetic algorithms (GAs) in this context and conducted a systematic study of the behavior of GAs on 20 fitness landscapes (FLs) of varying complexity. This has allowed the tuning of the GAs to be explored. On the basis of this study, recommendations for the best GA settings to use for a GA-directed high-throughput experimental program (in which populations and the number of generations is necessarily low) are reported. The FLs were based upon simple linear models and were characterized by the behavior of the GA on the landscape as demonstrated by stall plots and the footprints and adhesion of candidate solutions, which highlighted local optima (LOs). In order to maximize progress of the GA and to reduce the chances of becoming stuck in a LO it was best to use: 1) a large number of generations, 2) high populations, 3) removal of duplicate sequences (clones), 4) double mutation, and 5) high selection pressure (the two best individuals go to the next generation), and 6) to consider using a designed sequence as the starting point of the GA run. We believe that these recommendations might be appropriate starting points for studies employing GAs within DE experiments.

Improved Descriptors for the Quantitative Structure–Activity Relationship Modeling of Peptides and Proteins

“…A polarity scale and the associated hydrophilicity scale were required to provide good fits to both of the properties measured by Feng et al An initial polarity scale based upon Hellberg’s PP(1) was systematically modified in small increments using a simple genetic algorithm (GA) of our own design . To the best of our knowledge no systematic studies have been performed to establish the best GA parameters to use for this type of problem.…”

“…Fitting Hydrophilicity and Polarity Scales to Both the E Values and Conversion Data of Feng et al A polarity scale and the associated hydrophilicity scale were required to provide good fits to both of the properties measured by Feng et al 15 An initial polarity scale based upon Hellberg's PP(1) was systematically modified in small increments using a simple genetic algorithm (GA) of our own design. 22 To the best of our knowledge no systematic studies have been performed to establish the best GA parameters to use for this type of problem. We recently reported on recommended parameters when a GA was used to direct an experimental program, 22 but the results are not applicable to this problem.…”

“…22 To the best of our knowledge no systematic studies have been performed to establish the best GA parameters to use for this type of problem. We recently reported on recommended parameters when a GA was used to direct an experimental program, 22 but the results are not applicable to this problem. For the current GA the size of the search space was limited using a granular polarity scale (smallest increment = 0.01).…”

Acyltransferases as Tools for Polyketide Synthase Engineering

“…Directed evolution [ 214 , 215 , 216 , 217 , 218 ] was successfully applied to generate enzyme libraries. This approach usually uses saturation mutagenesis, error-prone polymerase chain reaction (PCR) or DNA shuffling to generate a library of mutated proteins.…”

Protein engineering of enzymes involved in lipid modification

“…This tool combines energy-and evolution-based approaches with smart filtering to identify additive stabilizing mutations (Bednar et al, 2015). Further algorithms for hot spot identification in enzymes include genetic algorithms (GA), the ProSAR algorithm or the model free ASRA algorithm (Barley, Turner & Goodacre, 2017;Feng et al, 2012). While ProSAR uses a statistical model to identify both positive and negative substitutions in a protein and then creates libraries with enhanced numbers of good mutations and fewer poor mutations, ASRA identifies the underlying regularity of the protein property landscape without any assumptions of linearity, additivity or any form of structure-property relationships.…”