On September 17th, the seminar participants divided into 4 groups. Each
group is undertaking a project in a specific area and will present their
results in the seminar. Below are descriptions of each project. If your
project has progress to report, please feel free to post to the mailing
list by sending email to [EMAIL PROTECTED] Please
let me know if you would like something posted on the webpage.
*** The Polyhedra & Experiments project will present on October 29th. ***
This project's general theme is section 7.7, "Perturbing Tree Metrics", of
_Phylogenetics_. The group is using various software packages for
Serkan Hosten ([EMAIL PROTECTED]) is in charge.
Josephine Yu ([EMAIL PROTECTED]) is using Polymake.
Ian Sammis ([EMAIL PROTECTED]) is using Phylip.
Alex Han ([EMAIL PROTECTED]) is using SplitsTree.
Francisco Santos ([EMAIL PROTECTED]) and
Federico Ardila ([EMAIL PROTECTED]) are also participating.
Josephine Yu has created a picture of a higher-dimensional tree using
Polymake. You can view it and rotate it at
(this page is also linked from the class webpage).
*** The Algebraic Invariants project will present on November 12th. ***
This project's general theme is section 7.6, "Group-Valued
Dissimilarities", and section 8.10, "Algebraic Dissimilarities", in
_Phylogenetics_. Related papers are Hagedorn, "Determining the Number and
Structure of Phylogenetic Invariants", and Evans/Speed, "Invariants of Some
Probability Modes Used in Phylogenetic Inference".
Kristian Ranestad ([EMAIL PROTECTED]) is in charge.
Radu Mihoescu ([EMAIL PROTECTED]),
Lauren Williams ([EMAIL PROTECTED]),
Nick Eriksson ([EMAIL PROTECTED]), and
Nicolas Bray ([EMAIL PROTECTED]) are participating.
Seth Sullivant ([EMAIL PROTECTED]) and perhaps
Dave Speyer ([EMAIL PROTECTED]) are semi-participating.
*** The Statistics project will present on November 19th. ***
This project's general theme is section 8.7, "The Felsenstein Zone", of
_Phylogenetics_, and Felsenstein's book, _Inferring Phylogenies_.
In their paper "Comparative Genomics via Phylogenetic Invariants for
Jukes-Cantor Semigroups", Blanchette/Sankoff presented a different
notion of distance between sequences, using signed permutations. The
models of Thorne-Kishino-Felsenstein, Graeme Mitchison, and Ian Holmes take
into account insertions & deletions in sequences.
Lior Pachter ([EMAIL PROTECTED]) is in charge.
David Yuchno ([EMAIL PROTECTED]) may write an undergraduate honors
thesis on Thorne-Kishino-Felsenstein.
Inga Hallgrimsdottir ([EMAIL PROTECTED]) and
Galen Huntington ([EMAIL PROTECTED]) are participating.
Samantha Reisenfeld ([EMAIL PROTECTED]) may be participating.
Peter Ralph ([EMAIL PROTECTED]) may be participating.
*** The Applications & Computation project will present on December 3rd. ***
This project's theme is implementation of algorithms.
One paper to look at is Atteson, "The Performance of Neighbor-Joining
Methods of Phylogenetic Reconstruction". An interesting mathematical
question arises: every algorithm for tree reconstruction defines an
equivalence relation on data. So how can we describe this equivalence
relation, i.e., which data will produce the same tree under this algorithm?
Another paper to look at is Pachter/Speyer, "Reconstructing Trees from
Subtree Weights", using m-dissimilarity maps. An interesting question is,
can we *decide* whether the supposed subtree weights actually come from a
Another paper to look at is Billera/Holmes/Vogtmann, "Geometry of the
Space of Phylogenetic Trees", on geodesics between trees. The consensus at
the AIM conference in Palo Alto was that the main open problem is finding a
good way to compute the distance between trees. There is a very
inefficient algorithm in the paper, which should be implemented.
Francis Su ([EMAIL PROTECTED]) is in charge.
Nina Amenta ([EMAIL PROTECTED]),
Dan Levy ([EMAIL PROTECTED]),
Steve Canon ([EMAIL PROTECTED]), and
Rudy Yoshida ([EMAIL PROTECTED]) are participating.
Ruchira Datta ([EMAIL PROTECTED]) may be participating.
Amelia Nissenbaum ([EMAIL PROTECTED]) may be participating.
Susan Holmes has a package, which she will send, to compute the distance
between trees. Half of the Applications & Computation Group will work
on implementing the Pachter-Speyer algorithm, and the other half will
work on computing convex hulls in the space of trees.
Lior is providing sequence data (from fugu fish, chickens, fungi, etc.) for
all of us to try out.
Visit the Mathematics of Phylogenetic Trees webpage: