Article
Computer Science, Software Engineering
Hiroshi Hirai, Yuni Iwamasa
Summary: This paper studies the problem of Quartet Compatibility in phylogenetic trees and introduces two new classes of quartet systems. Polynomial-time algorithms are presented for Quartet Compatibility in these systems.
Article
Biochemical Research Methods
Maryam Rabiee, Siavash Mirarab
Summary: The paper introduces a scalable likelihood-based approach called quartet co-estimation for co-estimation under the multi-species coalescent model. By independently inferring gene tree distributions and computing species tree topology and branch length, the method updates gene tree posterior probabilities based on the resulting species tree. Experimental results on simulations and a biological dataset demonstrate better accuracy compared to traditional methods.
Article
Biodiversity Conservation
Daijiang Li
Summary: Despite the availability of phylogenetic hypotheses for many taxonomic groups, most do not include all species. In phylogenetic ecology, there is a demand for phylogenies that encompass all species in a study. Existing software tools for grafting species to megatrees are limited to specific taxonomic groups. The new R package 'rtrees' aims to provide an easy, flexible, and reliable way to assemble phylogenies from megatrees, facilitating progress in phylogenetic ecology.
Article
Biochemistry & Molecular Biology
Chao Zhang, Siavash Mirarab
Summary: This paper introduces a threshold-free weighting scheme for quartet-based species tree inference, which improves the utility of summary methods and reduces incongruence with gene concatenation.
MOLECULAR BIOLOGY AND EVOLUTION
(2022)
Article
Evolutionary Biology
Xing-Xing Shen, Jacob L. Steenwyk, Antonis Rokas
Summary: Topological conflict is common in phylogenomic data with concatenation and coalescent methods. This study found that around 30-36% of genes in animal and plant phylogenomic studies exhibit inconsistency between likelihood-based and quartet-based signals, leading to a higher likelihood score but lower quartet score in one topology compared to another. Inconsistent genes are more likely to show high levels of gene tree discordance and may not accurately recover either of the conflicting topologies. Removing inconsistent genes can improve accuracy in data sets with low levels of incomlete lineage sorting and gene tree estimation error, but may not always lead to topologically identical species phylogenies in data sets with higher levels of incomlete lineage sorting and gene tree estimation error.
SYSTEMATIC BIOLOGY
(2021)
Article
Computer Science, Software Engineering
Bosheng Li, Jacek Kaluzny, Jonathan Klein, Dominik L. Michels, Wojtek Palubicki, Bedrich Benes, Soren Pirk
Summary: A novel method for reconstructing the 3D geometry of botanical trees from single photographs is introduced, addressing the challenges of faithfully reconstructing a tree from single-view sensor data. The method defines a reconstruction pipeline based on three neural networks, using radial bounding volumes to grow realistic branching for the tree crown and evaluating reconstructed geometries with several metrics.
ACM TRANSACTIONS ON GRAPHICS
(2021)
Article
Mathematics
Sagi Snir, Osnat Weissberg, Raphael Yuster
Summary: Studying the quartet distance in phylogenetic trees under partial information setting is complex and challenging. In this scenario, the average and maximum quartet distances may be influenced by the structure of the tree, and the conjecture proposed by Bandelt and Dress generally does not hold.
JOURNAL OF GRAPH THEORY
(2022)
Article
Biochemical Research Methods
Momoko Hayamizu, Kazuhisa Makino
Summary: Tree-based phylogenetic networks are a powerful model for representing complex data or non-tree-like evolution. However, these networks can have exponentially many support trees, leading to various computational problems. Hayamizu recently proposed a structure theorem and provided linear-time and linear-delay algorithms for different problems. In this paper, we focus on ranking the top-k support trees of a tree-based phylogenetic network based on their likelihood values, and present a linear-delay (optimal) algorithm for this problem.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Biochemical Research Methods
Sharmin Akter Mim, Md Zarif-Ul-Alam, Rezwana Reaz, Shamsuzzoha Bayzid, Mohammad Saifur Rahman
Summary: In this study, we improved the Quartet Fiduccia and Mattheyses (QFM) algorithm to enhance the quality and running time of large-scale phylogenetic trees. Our improved version, QFM-FI, can amalgamate millions of quartets over thousands of taxa into a highly accurate species tree within a shorter time. Comparative analysis shows that QFM-FI outperforms other state-of-the-art phylogeny reconstruction methods in terms of running time and tree quality.
Article
Biology
M. K. Garba, T. M. W. Nye, J. Lueg, S. F. Huckemann
Summary: The paper introduces a new space of phylogenetic trees called wald space, motivated by the need for a space suitable for statistical analysis of phylogenies with a geometry based on biologically principled assumptions. It investigates two related geometries on wald space involving Fisher information metric and continuous-valued Gaussian process. Computational methods are derived to compute geodesics in polynomial time for both geometries and numerical results show their similarity. The canonical and biologically motivated space proposed in the study is shown to be substantially different from the BHV geometry.
JOURNAL OF MATHEMATICAL BIOLOGY
(2021)
Article
Multidisciplinary Sciences
Alessio Caminata, Noah Giansiracusa, Han-Bom Moon, Luca Schaffler
Summary: Researchers successfully demonstrate that weighted dissimilarity vectors form a tropical subvariety of the tropical Grassmannian by replacing the definition of the dissimilarity map, providing a geometric interpretation in terms of configurations of points on rational normal curves.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Forestry
Gabriela Arcos-LeBert, Tamara Aravena-Hidalgo, Javier A. Figueroa, Fabian M. Jaksic, Sergio A. Castro
Summary: This study compared ecosystem services provided by native and exotic tree species in Santiago, Chile, finding no significant differences when not weighted by demand, but native species provided more services when weighted. The spatial distribution of ecosystem services did not correlate with the distribution of native and exotic species.
TREES-STRUCTURE AND FUNCTION
(2021)
Article
Computer Science, Artificial Intelligence
Min Qiu, Lizhi Peng, Ying Pang, Bo Yang, Panpan Li
Summary: The study introduces a new similarity evaluation method for a flexible neural tree (SEFNT) to maintain population diversity and address imbalanced data, significantly improving the classification performance of the FNT model. Comparisons with other algorithms show that SEFNT performs significantly better, demonstrating its effectiveness in practical problems such as internet video traffic identification.
APPLIED SOFT COMPUTING
(2021)
Article
Mathematics, Applied
Vu Dinh, Lam Si Tung
Summary: This paper focuses on the convergence rate of the maximum likelihood supertree method and proposes an analytic approach for analyzing it. By treating each tree as a point in a metric space and proving that the distance between the maximum likelihood supertree and the species tree converges to zero at a polynomial rate under certain conditions, the study contributes to understanding the behavior of supertree reconstruction methods.
Article
Computer Science, Theory & Methods
Jesper Jansson, Konstantinos Mampentzidis, T. P. Sandhya
Summary: We combine two fundamental optimization problems related to the construction of phylogenetic trees into a new problem and develop polynomial-time approximation algorithms. Experimental results show that using trees with fewer nodes typically does not destroy too much branching information.
INFORMATION AND COMPUTATION
(2023)