UPGMA

Phylogenetic analysis by distance based method

1. UPGMA Presented By Shreya Gopinath

2. Phylogenetic tree construction 2 methods Distance-based methods – Examples : UPGMA, Neighbor joining, Fitch- Margoliash method, minimum evolution Character-based methods – Input: A ligned sequences Output: Phylogenetic tree Examples : Parsimony , Maximum Likelihood

3. UPGMA UPGMA : Unweighted Pair Group Method with Arithmetic Mean Developed by Sokal and Michener in 1958. It is a Sequential clustering method Type of distance based method for Phylogenetic Tree construction UPGMA is the simplest method for constructing trees.

4. Generates rooted trees Generates ultra metric trees from a distance matrix Uses a simplest algorithm Input: Distance matrix containing pairwise statistical estimation of aligned sequences Output: Phylogenetic tree

5. UPGMA starts with a matrix of pairwise distances.  Each sample is denoted as a 'cluster'. Assigns all clusters to a star-like tree. The algorithm constructs a rooted tree that reflects the structure present in a pairwise similarity matrix. At each step, the nearest two clusters are combined into a higher-level cluster.  It assumes an ultra-metric tree in which the distances from the root to every branch tip are equal. UPGMA Algorithm

6. Steps Find the i and j with the smallest distance Dij . Create a new group ( ij ) which has n ( ij ) = n i + n j members. Connect i and j on the tree to a new node ( ij ). Give the edges connecting i to ( ij ) and j to ( ij ) same length so that the depth of group ( ij ) is D ij /2. Compute the distance between the new group and all other groups except i and j by using = Delete columns and rows corresponding to i and j and add one for ( ij ). If there are two or more groups left, go back to the first step  

7. Computational tools MEGA PHYLIP MVSP MVSP87 SAS SYN-TAX NTSYS DendroUPGMA

8. Advantages simple algorithm Fastest method easy to compute by hand or a variety of software Trees reflect phenotypic similarities by phylogenetic distances Data can be arranged in random order prior to analysis Rooted trees are generated that are easy to analyze

9. Disadvantages It assumes the same evolutionary speed on all lineages It frequently generates wrong tree topologies Re-rooting is not allowed Algorithm does not aim to reflect evolutionary descent It assumes a randomized molecular clock.

10. Applications In ecology, it is one of the most popular methods for the classification of sampling units (such as vegetation plots) on the basis of their pairwise similarities in relevant descriptor variables (such as species composition).[3] In bioinformatics, UPGMA is used for the creation of  phenetic  trees ( phenograms ). UPGMA was initially designed for use in protein electrophoresis studies, but is currently most often used to produce guide trees for more sophisticated algorithms. This algorithm is for example used in sequence alignment procedures, as it proposes one order in which the sequences will be aligned. Indeed, the guide tree aims at grouping the most similar sequences, regardless of their evolutionary rate or phylogenetic affinities, and that is exactly the goal of UPGMA.[4] In phylogenetics, UPGMA assumes a constant rate of evolution (molecular clock hypothesis), and is not a  well­regarded  method for inferring relationships unless this assumption has been tested and justified for the data set being used.

11. Example 1. Calculate the pairwise distance matrix A B C D E F A 1 3 6 7 10 B 1 3 6 7 10 C 3 3 5 6 9 D 6 6 5 1 7 E 7 7 6 1 8 F 10 10 9 7 8

12. 2. Group the 2 most closely related sequences A B C D E F A 1 3 6 7 10 B 1 3 6 7 10 C 3 3 5 6 9 D 6 6 5 1 7 E 7 7 6 1 8 F 10 10 9 7 8 A B 0.5 0.5

13. 3. Recalculate the distance matrix and take the next smallest distance A/B C D E F A/B 3 6 7 10 C 3 5 6 9 D 6 5 1 7 E 7 6 1 8 F 10 9 7 8 A B 0.5 0.5 D E 0.5 0.5

14. 3. Recalculate the distance matrix and take the next smallest distance A B 0.5 0.5 D E 0.5 0.5 A/B C D/E F A/B 3 6.5 10 C 3 5.5 9 D/E 6.5 5.5 7.5 F 10 9 7.5 C 1 1.5

15. 3. Recalculate the distance matrix and take the next smallest distance A B 0.5 0.5 D E 0.5 0.5 C 1 1.5 A/B/C D/E F A/B/C 6 9.5 D/E 6 7.5 F 9.5 7.5 1.5 2.5

16. 3. Recalculate the distance matrix and take the next smallest distance A B 0.5 0.5 D E 0.5 0.5 C 1 1.5 1.5 2.5 A/B/C/D/E F A/B/C/D/E 8.5 F 8.5 F 4.25 1.25