Files in this item

FilesDescriptionFormat

application/pdf

application/pdfPARK-THESIS-2021.pdf (370kB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Disjoint Tree Mergers for large-scale maximum-likelihood tree estimation
Author(s):Park, Minhyuk
Advisor(s):Warnow, Tandy J
Department / Program:Computer Science
Discipline:Computer Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:M.S.
Genre:Thesis
Subject(s):phylogeny estimation
maximum likelihood
RAxML
IQ-TREE
FastTree
cox1
heterotachy
disjoint tree mergers
Tree of Life
Abstract:Gene tree estimation is a biological problem that garners a lot of interest due to its ability to uncover evolutionary relationships in different genes which provides valuable insight into the hidden mechanisms of evolution. However, large-scale gene tree estimation has largely been unexplored, partially due to the limited available methods that can run on large datasets. Instead, a lot of effort has been focused on developing methods that are accurate on small to medium-sized datasets. We present a re-evaluation of divide-and-conquer pipelines on a variety of model conditions, including fragmentary sequences and heterogeneous evolution patterns, and show that our design of divide-and-conquer pipelines can consistently match or outperform FastTree and IQ-TREE with little overhead in runtime while matching or outperforming RAxML except on small datasets with very challenging model conditions. Furthermore, our divide-and-conquer pipeline is able to run on datasets that are too large for IQ-TREE or RAxML to handle.
Issue Date:2021-04-23
Type:Thesis
URI:http://hdl.handle.net/2142/110713
Rights Information:Copyright 2021 Minhyuk Park
Date Available in IDEALS:2021-09-17
Date Deposited:2021-05


This item appears in the following Collection(s)

Item Statistics