Costas Maranas and his colleagues at Pennsylvania State University are developing a computational method for predicting the number and locations of crossovers that occur in DNA shuffling, a process that mixes genetic material from different parent sequences in order to identify which genes produce desirable products.
While DNA shuffling is commonly used to generate combinatorial libraries for novel protein design, currently available experimental techniques are labor-intensive and costly. Maranas said that the predictive models he has created to quantify the outcome of these experiments would help researchers identify the best experimental route.
“We used thermodynamics and reaction engineering to evaluate and model this complex reaction network so we can now predict where the DNA from different parent genes will combine,” said Maranas.
The model studied how fragment length, annealing temperature, sequence identity, and the number of shuffled parent sequences affect the number, type, and distribution of crossovers along the length of reassembled sequences. The more similar genes are, the more potential for crossover exists.
“If the sequences are very different and the experiment is done at high temperature, there will be no crossovers at all,” said Maranas. Fragment size also can affect the number of crossovers.
Comparisons with experimental data have shown a high level of agreement. “At minimum, we can predict whether no crossovers, a few crossovers or many crossovers will be generated,” said Maranas.
Maranas said that the University is negotiating licensing agreements with several directed evolution companies interested in using the model as a predictive tool to optimize their protocols.
The researchers are currently studying crossover prediction in other protocols, which unlike DNA shuffling, can be used to recombine sequences with very low sequence identy. Maranas also said the new technique would work in cases where bits and pieces of parent sequences are shuffled rather than the entire length.