Multiple sequence alignment is an important tool for computational analysis of nucleotide or amino acid sequences. It is also a challenging combinatorial optimization problem in computer science. As a large amount of sequence data is becoming available from genome and other large-scale sequencing projects, efficiency, as well as accuracy, is currently required for a multiple sequence alignment program. Several new programs are being developed aiming at improving both efficiency and accuracy. We overview the algorithms and performances of new programs including that by ourselves.