 Dear students, in this module, I'm going to start working on multiple sequence alignment with you. You're already aware that we can do pairwise sequence alignment and compare two biological sequences and score them using PAM 120 or Blossom 62 matrices, amongst others. So to start with multiple sequence alignment, the pairwise sequence alignment only considered two sequences. So one sequence was to put on top and the second sequence was put on the bottom and you try to compare the nucleotides or amino acids between the two sequences. For the multiple sequence alignment, the same problem is expanded to include three or more sequences. So the need for such a strategy may be that if you want to compare multiple sequences, then MSA can be very useful. So in the pairwise sequence alignment, we had global as well as local sequence alignment besides the repeated matches or the overlapped hanging portions or the trailing portions. In the MSA or multiple sequence alignment, we have mostly global sequence alignment only. Here we have several sequences, five of them that have been aligned and we call them a block and they have been aligned such that if you can see closely that there are some mismatches or substitutions, there are some gaps as well here as well and they are there in the sequence but the five sequences are aligned very nicely. So this is what can be achieved by looking at the MSA or by employing the multiple sequence alignment strategies. So for the pairwise sequence alignment, we use the dynamic programming paradigm in which we had this entire alignment matrix and we computed each position within this matrix step by step. This helped us to compare two sequences but imagine a case where you have to compare multiple sequences and therefore you would have to perform dynamic programming or DP in multiple dimensions. So that will increase the computational cost of the dynamic programming to such a proportion that we will not be able to run the program. In case of MSA, typically we do dynamic programming for no more than 9 or 10 sequences. If you have a lot more sequences, then we move to progressive alignment strategies. So in conclusion, multiple sequence alignment can help you to compare more than two, that is three or even more sequences together and dynamic programming strategy may have to be abandoned in favor of progressive alignment to compare these multiple sequences in a small amount of time. Also the software that is most commonly used and we will see in the later modules is the cluster for multiple sequence alignment.