Modeling DNA mutation and recombination for directed evolution experiments

Directed evolution experiments rely on the cyclical application of mutagene sis, screening and amplification in a test tube. They have led to the creat ion of novel proteins for a wide range of applications. However, directed e volution currently requires an uncertain, typically large, number of labor intensive and expensive experimental cycles before proteins with improved f unction are identified. This paper introduces predictive models for quantif ying the outcome of the experiments aiding in the setup of directed evoluti on for maximizing the chances of obtaining DNA sequences encoding enzymes w ith improved activities. Two methods of DNA manipulation are analysed. erro r-prone PCR and DNA recombination. Error-prone PCR is a DNA replication pro cess that intentionally introduces copying errors by imposing mutagenic rea ction conditions. The proposed model calculates the probability of producin g a specific nucleotide sequence after a number of PCR cycles. DNA recombin ation methods rely on the mixing and concatenation of genetic material from a number of parent sequences. This paper focuses on modeling a specific DN A recombination protocol, DNA shuffling. Three aspects of the DNA shuffling procedure are modeled: the fragment size distribution after random fragmen tation by DNase I, the assembly of DNA fragments, and the probability of as sembling specific sequences or combinations of mutations. Results obtained with the proposed models compare favorably with experimental data. (C) 2000 Academic Press.