UVP Doc-It Life Science - Manual (Page 95)

Performing 1D Analysis
90
http://marketing.byu.edu/htmlpages/books/pcmds/WARD.html
Neighbor joining: At each stage of clustering the total branch length is minimized. The distance
between two items is approximately the sum of the branch lengths between them. The trees are
not right-aligned and branches can have negative values. Details of the method can be found at
http://www.icp.ucl.ac.be/~opperd/private/neighbor.html