 खुतस्धाल धबलीब खॉत्धाल धबलीबおお मतेद फोरी of multiple sequence alignment आर्ँ सामन। मतेद मुल्टिपल सीक्षूblowing जिल्ह में साँआफरी ofR base all ML applications availableँई Tina go' अआर्ँ सचुफी for these multiple sequence alignment purposes. Clustal W is one of the multiple methods available for multiple sequence alignment, global multiple sequence alignment. And through Clustal W, we can align protein sequences with the DNA sequences, globally multiple sequence alignment, we can perform them. This method was initially proposed by Julie Thompson, Toby Gibson, these gentlemen, they were from the EBL Germany. And with the collaboration of other workers, for example, there was Desmond Hitchens and this one was from the EBI UK. So, they collectively, they are proposed and designed this method for multiple sequence alignment, which is popularly known with the name of Clustal W. Aarchkal Clustal W के कापी सारे वेर्यंट्स, multiple sequence alignment को परफाम करनी की लिया अवलिबल है, जिस में मशुर जो है वो Clustal W 2 है, Clustal X है, Clustal Omega है, But in this module, we will only focus on Clustal W. Clustal W has the ability to create multiple sequence alignment. It has the ability to manipulate existing alignments. It has the ability to do profile analysis. It has the ability to create phylogenetic trees. So, it is a multi-dimensional method basically, which has multiple abilities. And when we do multiple sequence alignment through it, then through any other method, if we have done multiple sequence alignment and it gives us multiple sequence alignment in its compatible format, then it can manipulate that alignment, it can do profile analysis of that alignment and it can construct phylogenetic tree along with it. When Clustal W performs multiple sequence alignment, then it uses progressive alignment method. So, basically this Clustal W method is based upon the approach of progressive alignment for multiple sequence alignment and Clustal W's applications are freely available by EMBL or EBI or other sites through which Clustal W is freely available. When Clustal W does progressive alignment, then we have discussed in progressive alignment method that the first two most closely related sequences are aligned with the next closely related sequence with that pairwise alignment and then one by one with this alignment the next closely related sequence is merged with one by one with this alignment and at the end when all the sequences are aligned we get the global multiple sequence alignment when we reach the end sequence. So, this also adopts the philosophy and uses it when it performs multiple sequence alignment. Similarly, one by one progressively pairwise alignment and the first step is to select two sequences with that one by one aligns each sequence and the result is a multiple sequence alignment. Now, if we try to understand its philosophy further then we know the strategy of progressive alignment. This is a strategy based upon the philosophy of heuristic approach and heuristic approach is we have discussed about it that any heuristic approach due to its greediness it never guarantees us that the result we have this is the best one. But still, using heuristics is important because multiple sequence alignment is an NP-hard problem and if we take it towards the optimum then the time complexity and complexity will be so expensive that the solution we have that will not be achievable in the real time because that time will utilize so much that the result we have we will not be able to wait that much. This algorithm like the other progressive approaches first of all this rough distance matrix creates and that distance matrix that will be created from the pair-wise sequences performed with that pair-wise alignment this will generate a rough distance matrix and its scores the distance scores of pair-wise alignment from there the scores of that distance will get from that pair-wise alignment next step when it will have a distance matrix and distance scores it will use the never-gaining method why it will use it the never-gaining method will be used to create the guide tree to get the global multiple sequence alignment and to create the guide tree the never-gaining method will be used and the guide tree will be created and the guide tree will be used further to have that will serve as a rough template for the clades that shares similar insertion and deletion features or generally it provides close to the optimal solution close to the optimal because it will never guarantee you to have the best solution better solution better solution better solution because heuristic approach and the better we selected it and output this is your optimal solution to the best of the rules applied heuristics and the NP-hard nature of the multiple sequence alignment its complexity increases with the length and number of sequences so if it goes towards the best then the complexity will be so much that at that time the researcher cannot wait for a day or a week to reach the best at that time heuristically the first best solution it gives the output and it is our close to the optimal result and the guide tree because the noise is less sensitive so he used a never-getting method to produce the guide tree or if there is a noise or if he is dealing with divergent sequences there is less sensitivity to the noise so there is no difference this diagram is basically showing the whole flow pairwise alignment so it will calculate the distance matrix and then un-rooted neighbor joining tree for the generation of the guide tree and then rooted neighbor joining tree, guide tree and sequence weights alignment following the guide tree so in the first step we have this distance matrix and the joining method will be used to generate this guide tree and this guide tree will be clearly identified which will be more closely related which will share similar insertion and deletion features and then we have close to optimal solution global multiple sequence alignment yield I am saying it again and again that it can be better if we apply another method but in that scenario the output system this is close to more close to the optimal but not the best one