Multiobjective optimization-driven primer design mechanism: towards user-specified parameters of PCR primer.

ABSTRACT

Primers are critical for polymerase chain reaction (PCR) and influence PCR experimental outcomes. Designing numerous combinations of forward and reverse primers involves various primer constraints, posing a computational challenge. Most PCR primer design methods limit parameters because the available algorithms use general fitness functions. This study designed new fitness functions based on user-specified parameters and used the functions in a primer design approach based on the multiobjective particle swarm optimization (MOPSO) algorithm to address the challenge of primer design with user-specified parameters. Multicriteria evaluation was conducted simultaneously based on primer constraints. The fitness functions were evaluated using 7425 DNA sequences and compared with a predominant primer design approach based on optimization algorithms. Each DNA sequence was run 100 times to calculate the difference between the user-specified parameters and primer constraint values. The algorithms based on fitness functions with user-specified parameters outperformed the algorithms based on general fitness functions for 11 primer constraints. Moreover, MOPSO exhibited superior implementation in all experiments. Practical gel electrophoresis was conducted to verify the PCR experiments and established that MOPSO effectively designs primers based on user-specified parameters.