MATH:7450 (22M:305) Topics in Topology: Scien:fic and Engineering Applica:ons of Algebraic Topology Nov 18, 2013: Viral Evolu:on II and Crickets via cubical homology. Fall 2013 course offered through the University of Iowa Division of Con:nuing Educa:on Isabel K. Darcy, Department of Mathema:cs Applied Mathema:cal and Computa:onal Sciences, University of Iowa hVp://www.math.uiowa.edu/~idarcy/AppliedTopology.html
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
MATH:7450 (22M:305) Topics in Topology: Scien:fic and Engineering Applica:ons of Algebraic Topology
Nov 18, 2013: Viral Evolu:on II and Crickets via cubical homology.
Fall 2013 course offered through the
University of Iowa Division of Con:nuing Educa:on
Isabel K. Darcy, Department of Mathema:cs Applied Mathema:cal and Computa:onal Sciences,
HIV – single segment (so no reassortment) Non-‐trivial Hk
k = 1, 2. Thus horizontal transfer via homologous recombina:on.
TOP = Topological obstruc:on = maximum barcode length in non-‐zero dimensions TOP ≠ 0 è no addi:ve distance tree
TOP is stable
ICR = irreducible cycle rate = average number of the one-‐dimensional
irreducible cycles per unit of :me Simula:ons show that ICR is propor:onal to and provides a lower bound for recombina:on/reassortment rate
Persistent homology Viral evolu:on Filtra:on value ε Gene:c distance (evolu:onary scale) β0 at filtra:on value ε Number of clusters at scale ε Generators of H0 A representa:ve element of
the cluster Hierarchical Hierarchical clustering rela:onship among H0 generators β1 Number of re:culate events
(recombina:on and reassortment)
Persistent homology Viral evolu:on Generators of H1 Re:culate events Generators of H2 Complex horizontal
genomic exchange Hk ≠ 0 for some k > 0 No phylogene:c tree
representa:on No. of Lower bound on rate of higher-‐dimensional re:culate events generators over :me (irreducible cycle rate)
Thursday December 12, 2013 9:00am-‐9:50am Structure of the Afferent Terminals in Terminal Ganglion of a Cricket and Persistent Homology Tomas Gedeon (Montana State University)
2012 dense nerve center
Nerve endings
Terminal Ganglion
distal vs proximal closest 15%
proximal data: long, medium, short 42428, 27442, 29297
distal data: long sparse since harder to obtain (6194).
furthest 30%
Proximal data was filtered: 1.) remove outliers (noise) experimental error using data obtained from many
different crickets 2.) remove (redundant) points in densest regions to improve computa:onal speed Performed comparison using data sampled from a Gaussian Mixture Model.
cubical homology: for reducing memory requirements (bitmap). To calculate persistent homology: Mrozek, Batko and Wanner’s
cubPersistenceMD. To find cycles:
Homcubes in CHomP.
cubical homology To calculate persistent homology: Mrozek, Batko and Wanner’s cubPersistenceMD. www.ii.uj.edu.pl/~mrozek/sooware/homology.html Kaczynski T, Mischaikow K, Mrozek M (2004) Computa:onal Homology. Applied Mathema:cal Sciences. Springer. hVp://chomp.rutgers.edu hVp://www.sagemath.org/doc/reference/homology/sage/interfaces/chomp.html