Abstract—PCR-CTPP (Polymerase chain reaction with confronting two-pair primers) is a time- and cost-effective SNP genotyping method. However, the design of feasible PCR-CTPP primer sets is still challenging. In this study, we propose a PSO (particle swarm optimization) with fuzzy adaptive strategy, named FAPSO, to design feasible PCR-CTPP primer sets. Two hundred and eighty-eight SNPs which exclude the deletion/insertion polymorphism (DIP) and multi-nucleotide polymorphism (MNP) in SLC6A4 gene were tested in silicon by the proposed method. The results shown the proposed method provides more feasible PCR-CTPP primers than a GA (genetic algorithm)-based and a native PSO-based method. In conclusion, the FAPSO-based method is useful to assist the biologists and researchers to design feasible PCR-CTPP primer sets. Index Terms—PCR-CTPP, SNP, GA, PSO, FAPSO I. INTRODUCTION NPs (Single Nucleotide Polymorphisms) are usually used in association studies of diseases and cancers due to their great quantity. Many laboratories have introduced high-throughput platforms of SNP genotyping such as real-time PCR (polymerase chain reaction) [1] and SNP array [2] to validate SNPs or novel mutations, but they are more expensive than the other existing methods. The PCR-restriction fragment length polymorphism (RFLP) genotyping [3-5] is still favorite due to its inexpensive for the small-scale genotyping. However, the chief shortcoming of the PCR-RFLP is usually long digestion time in 2-3 hours for restriction enzymes [6, 7]. Recently, a restriction enzyme-free SNP genotyping technique [8, 9] was developed named PCR-CTPP (PCR with confronting two-pair primers). PCR-CTPP has genotyped many SNPs successfully, such as interleukin-1B (IL-1B) C-31T, interleukin-2 (IL-2) -330G, beta2-adrenergic receptor (beta2-AR) Gln27Glu, aldehyde dehydrogenase 2 (ALDH2) [10], pylori-induced gastric atrophy [11], severe Manuscript received April 30, 2012. This work was supported in part by the National Science Council in Taiwan under grants NSC99-2622-E-151-019-CC3, NSC99-2221-E-151-056-, NSC100-2221-E-151-049-MY3, and NSC100-2221-E-151-051-MY2. Cheng-Hong Yang is with the Department of Electronic Engineering, National Kaohsiung University of Applied Sciences, and Department of Network Systems, Toko University, Chiayi, Taiwan (e-mail: chyang@cc.kuas.edu.tw). Yu-Huei Cheng is with the Department of Network Systems, Toko University, Chiayi, Taiwan (e-mail: yuhuei.cheng@gmail.com). Li-Yeh Chuang is with the Department of Chemical Engineering & Institute of Biotechnology and Chemical Engineering, I-Shou University, Kaohsiung, Taiwan (email: chuang@isu.edu.tw). coronary artery disease [12], and esophageal cancer risk [13]. Although PCR-CTPP is suitable and reliable for most cases of SNPs, the existing computation methods for designing feasible PCR-CTPP primer sets are still insufficient. Many typical primer design constraints need to be considered when designing PCR-CTPP primers, such as primer length, length difference of primer pair, PCR product length, GC proportion, melting temperature (T m ), difference in melting temperature (T m-diff ), GC clamp, the existence of dimers (including cross-dimers and self-dimers), hairpin structure, and specificity. The important factor is especially T m-diff [14]. That makes design the feasible PCR-CTPP primers challenging. In the past, we have introduced a genetic algorithm (GA) to design available PCR-CTPP primer sets [15, 16]. However, the computational results still need to be improved, especially the T m difference. In order to design the more feasible PCR-CTPP primer sets, we applied particle swarm optimization (PSO) [17] to improve the problem [18]. Particle swarm optimization (PSO) simulates the social behavior of organisms, such as birds in a flock or fish in a school, is a population-based stochastic optimization technique developed by Kennedy and Eberhart [17]. In a PSO, each single candidate solution is described as "an individual bird of the flock", that is, a particle in the search space. Each particle finds the better solution using its own memory as well as knowledge gained by the swarm. Each particle has a fitness value evaluated by an optimized fitness function and a velocity directs the movement of the particles. During movement, each particle adjusts its position in terms of its own experience and the experience of a neighbouring particle, thus making the best position encounter. PSO has been successfully applied in many fields, such as function optimization, artificial neural network training, and fuzzy system control. A comprehensive survey of PSO algorithms and their applications can be found in Kennedy et al. [19]. However, a fixed inertia weight or linearly decreasing inertia weight used in PSO simplifies the complex non-linear search process [20, 21]. In order to balance the global and local search ability of PSO, a fuzzy system adapts the inertia weight of PSO dynamically had been implemented [22]. Following, we also introduced the fuzzy adaptive strategy to the PCR-CTPP primer design problem [23]. The PSO with fuzzy adaptive strategy, named FAPSO (Fuzzy Adaptive Particle Swarm Optimization) here, shows the better PCR-CTPP primers designed. In this paper, we will further discuss the results of PCR-CTPP primer design for GA, PSO, and FAPSO. PCR-CTPP design based on Particle Swarm Optimization with Fuzzy Adaptive Strategy Cheng-Hong Yang, Member, IAENG, Yu-Huei Cheng, and Li-Yeh Chuang S Engineering Letters, 20:2, EL_20_2_09 (Advance online publication: 26 May 2012) ______________________________________________________________________________________