Abstract: Sequence alignment has become a fundamental process in computational biology as it helps in finding similarity regions between biological sequences that may indicate the common properties across the sequences. Global Multiple Sequence Alignment (MSA) is a way to align the entire group of biological sequences. The exact solution of the global MSA is an NP-complete problem and iterative sequence alignment is among the main heuristic methods used for solving this computationally expensive problem. This approach realigns and evaluates initial biological sequences repeatedly through a number of iterations. The iterative approach is mostly used in conjunction with other computational optimization approaches and the total number of iterations as well as the iteration time mainly affects the overall alignment time. In this study, a parallel Particle Swarm Optimization (PSO) algorithm is presented for solving the global MSA problem based on iterative sequence alignment. The algorithm has been implemented using the massage-passing interface (MPI) library and tested over a Linux cluster and over the EUMed Grid. Experimental results are presented that demonstrate the performance of the proposed algorithm using different proteins from the SABmark and BAliBASE benchmark databases, different substitution matrices and different gap penalty models.