Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!wuarchive!psuvax1!psuvm!CAOS!JACKL From: JACKL@CAOS.CAOS.KUN.NL (Jack AM Leunissen) Newsgroups: bit.listserv.info-gcg Subject: RE: Divergence Diagrams from distance matrices Message-ID: Date: 7 Feb 90 10:15:00 GMT Sender: "INFO-GCG: Univ. of Wisconsin Genetics Analysis Software Discussion" Reply-To: Jack AM Leunissen Lines: 32 Approved: NETNEWS@PSUVM Gateway X-VMS-To: IN%"INFO-GCG@UTORONTO.BITNET" Comments: Sent using PMDF-822 V3.0, routing is done by KUNRC1 Hi there, I would like to reply to Paul Scheeman's question regarding the contruction of evolutionary trees from DNA (or protein) data. He writes >I have a user who has a set of DNA fragments and an estimate of divergence >from each to each--rather like a distance matrix. Is there a readily >useable program to draw estimated evolutionary trees from this data? The package that I use for the construction of evolutionary (phylogenetic) trees from molecular sequence data is Joe Felsenstein's PHYLIP package. It contains several programs for DNA data. You can 'write' to: joe@genetics.washington.edu or felsenst@uwalocke.bitnnet >The software from other fields (namely SAS, SPSS, etc) will give us the >estimated 2-dimensional map using multidimensional scaling, and we can >then by hand estimate what a tree would be if we know the earliest point >on the map. If there is a method more standard for molecular biologists, >we would like to know before re-inventing a procedure. Programs that construct an evolutionary tree always calculated an UNROOTED tree: this means that the earliest point of divergence in the tree must come from external evidence! Jack Leunissen CAOS/CAMM Center University of Nijmegen jackl@caos.caos.kun.nl P.S. : Joe Felsenstein distributes his package free of charge!