UPGMA

UPGMA, the unweighted pair group method with arithmetic mean, is a simple agglomerative hierarchical clustering method that builds a rooted tree diagram, or dendrogram, from pairwise similarity or distance measurements. The method is generally attributed to Robert Sokal and Charles Michener, who introduced it in 1958 as a way to evaluate systematic relationships. At each step the algorithm finds the two closest clusters and merges them, then recalculates distances to the remaining clusters using a proportional average weighted by cluster size, so that every original distance value contributes equally to the final tree. UPGMA is used in ecology to classify sampling units by species composition, in bioinformatics to build phenetic trees and guide sequence alignment, and in phylogenetics to reconstruct evolutionary relationships, though its assumption of a constant evolutionary rate, the molecular clock hypothesis, and its assumption of an ultrametric tree limit its accuracy when that assumption does not hold. This description is adapted from Wikipedia contributors under CC BY-SA 4.0; changes were made. https://creativecommons.org/licenses/by-sa/4.0/