Incomplete Distance Matrices Research Articles

Machine learning based imputation techniques for estimating phylogenetic trees from incomplete distance matrices

BackgroundWith the rapid growth rate of newly sequenced genomes, species tree inference from genes sampled throughout the whole genome has become a basic task in comparative and evolutionary biology. However, substantial challenges remain in leveraging these large scale molecular data. One of the foremost challenges is to develop efficient methods that can handle missing data. Popular distance-based methods, such as NJ (neighbor joining) and UPGMA (unweighted pair group method with arithmetic mean) require complete distance matrices without any missing data.ResultsWe introduce two highly accurate machine learning based distance imputation techniques. These methods are based on matrix factorization and autoencoder based deep learning architectures. We evaluated these two methods on a collection of simulated and biological datasets. Experimental results suggest that our proposed methods match or improve upon the best alternate distance imputation techniques. Moreover, these methods are scalable to large datasets with hundreds of taxa, and can handle a substantial amount of missing data.ConclusionsThis study shows, for the first time, the power and feasibility of applying deep learning techniques for imputing distance matrices. Thus, this study advances the state-of-the-art in phylogenetic tree construction in the presence of missing data. The proposed methods are available in open source form at https://github.com/Ananya-Bhattacharjee/ImputeDistances.

T-REX: a web server for inferring, validating and visualizing phylogenetic trees and networks

T-REX (Tree and reticulogram REConstruction) is a web server dedicated to the reconstruction of phylogenetic trees, reticulation networks and to the inference of horizontal gene transfer (HGT) events. T-REX includes several popular bioinformatics applications such as MUSCLE, MAFFT, Neighbor Joining, NINJA, BioNJ, PhyML, RAxML, random phylogenetic tree generator and some well-known sequence-to-distance transformation models. It also comprises fast and effective methods for inferring phylogenetic trees from complete and incomplete distance matrices as well as for reconstructing reticulograms and HGT networks, including the detection and validation of complete and partial gene transfers, inference of consensus HGT scenarios and interactive HGT identification, developed by the authors. The included methods allows for validating and visualizing phylogenetic trees and networks which can be built from distance or sequence data. The web server is available at: www.trex.uqam.ca.

Ape 3.0: New tools for distance-based phylogenetics and evolutionary analysis in R

Reflecting its continuously increasing versatility and functionality, the popularity of the ape (analysis of phylogenetics and evolution) software package has grown steadily over the years. Among its features, it has a strong distance-based component allowing the user to compute distances from aligned DNA sequences based on most methods from the literature and also build phylogenetic trees from them. However, even data generated with modern genomic approaches can fail to give rise to sufficiently reliable distance estimates. One way to overcome this problem is to exclude such estimates from data analysis giving rise to an incomplete distance data set (as opposed to a complete one). So far their analysis has been out of reach for ape. To remedy this, we have incorporated into ape several methods from the literature for phylogenetic inference from incomplete distance matrices. In addition, we have also extended ape's repertoire for phylogenetic inference from complete distances, added a new object class to efficiently encode sets of splits of taxa, and extended the functionality of some of its existing functions. ape is distributed through the Comprehensive R Archive Network: http://cran.r-project.org/web/packages/ape/index.html Further information may be found at http://ape.mpl.ird.fr/pegas/

Fast NJ-like algorithms to deal with incomplete distance matrices.

BackgroundDistance-based phylogeny inference methods first estimate evolutionary distances between every pair of taxa, then build a tree from the so-obtained distance matrix. These methods are fast and fairly accurate. However, they hardly deal with incomplete distance matrices. Such matrices are frequent with recent multi-gene studies, when two species do not share any gene in analyzed data. The few existing algorithms to infer trees with satisfying accuracy from incomplete distance matrices have time complexity in O(n4) or more, where n is the number of taxa, which precludes large scale studies. Agglomerative distance algorithms (e.g. NJ [1,2]) are much faster, with time complexity in O(n3) which allows huge datasets and heavy bootstrap analyses to be dealt with. These algorithms proceed in three steps: (a) search for the taxon pair to be agglomerated, (b) estimate the lengths of the two so-created branches, (c) reduce the distance matrix and return to (a) until the tree is fully resolved. But available agglomerative algorithms cannot deal with incomplete matrices.ResultsWe propose an adaptation to incomplete matrices of three agglomerative algorithms, namely NJ, BIONJ [3] and MVR [4]. Our adaptation generalizes to incomplete matrices the taxon pair selection criterion of NJ (also used by BIONJ and MVR), and combines this generalized criterion with that of ADDTREE [5]. Steps (b) and (c) are also modified, but O(n3) time complexity is kept. The performance of these new algorithms is studied with large scale simulations, which mimic multi-gene phylogenomic datasets. Our new algorithms – named NJ*, BIONJ* and MVR* – infer phylogenetic trees that are as least as accurate as those inferred by other available methods, but with much faster running times. MVR* presents the best overall performance. This algorithm accounts for the variance of the pairwise evolutionary distance estimates, and is well suited for multi-gene studies where some distances are accurately estimated using numerous genes, whereas others are poorly estimated (or not estimated) due to the low number (absence) of sequenced genes being shared by both species.ConclusionOur distance-based agglomerative algorithms NJ*, BIONJ* and MVR* are fast and accurate, and should be quite useful for large scale phylogenomic studies. When combined with the SDM method [6] to estimate a distance matrix from multiple genes, they offer a relevant alternative to usual supertree techniques [7]. Binaries and all simulated data are downloadable from [8].

Incomplete distance matrices, supertrees and bat phylogeny

In this paper, we evaluate the relative performance of competing approaches for estimating phylogenies from incomplete distance matrices. The direct approach proceeds with phylogenetic reconstruction while ignoring missing cells, whereas the indirect approach proceeds by estimating the missing distances prior to phylogenetic analysis. Two distinct indirect procedures based on the ultrametric inequality and the four-point condition are further compared. Using simulations, we show that more reliable results are obtained when such indirect methods are used. Expectedly, the phylogenies become less accurate as the percentage of missing cells increases, but combining different estimation methods greatly improves the accuracy. An application to bat phylogeny confirms the results obtained in the simulation study and illustrates the effect of missing distances in the construction of supertrees.

T-REX: reconstructing and visualizing phylogenetic trees and reticulation networks.

T-REX (tree and reticulogram reconstruction) is an application to reconstruct phylogenetic trees and reticulation networks from distance matrices. The application includes a number of tree fitting methods like NJ, UNJ or ADDTREE which have been very popular in phylogenetic analysis. At the same time, the software comprises several new methods of phylogenetic analysis such as: tree reconstruction using weights, tree inference from incomplete distance matrices or modeling a reticulation network for a collection of objects or species. T-REX also allows the user to visualize obtained tree or network structures using Hierarchical, Radial or Axial types of tree drawing and manipulate them interactively. T-REX is a freeware package available online at: http://www.fas.umontreal.ca/biol/casgrain/en/labo/t-rex

The triangles method to buildX-trees from incomplete distance matrices

On decrit une methode pour la reconstruction des X-arbres (arbres values admettant X comme ensemble de feuilles) a partir de tableaux de distances incomplets (ou certaines valeurs sont incertaines ou inconnues). Elle permet de construire un arbre non oriente a partir de 2n - 3 valeurs de distance entre les n elements de X, sous des conditions qui sont explicitees. Cette construction est basee sur une relation entre X-arbres et 2-arbres values generalises d'ensemble de sommets X.

Reweaving the Tapestry: What Can We Learn from Sibley and Ahlquist (1990)?

ABsTRAcr.-Sibley and Ahlquist's phylogeny of the birds (The Tapestry) has drawn both praise and criticism. Two major criticisms are that trees were based on incomplete distance matrices and that their tree-building algorithm (UPGMA) was inappropriate. Their 1990 book answered critics by including several complete matrices analyzed by the Fitch-Margoliash algorithm. Matrices were constructed by combining species into composite taxa, which requires additional (possibly defensible) assumptions and introduces additional (probably random) error. Three problems remain: (1) The algorithm used does not always find the bestfit topology, depending on taxon order. (2) The error variance of the data does not fit the assumptions of the Fitch-Margoliash algorithm. The assumptions of the algorithms of Fitch and Margoliash and of Cavalli-Sforza and Edwards are limiting cases that bracket the truth. (3) Even a matrix with no phylogenetic content has a best-fit topology; some test of the strength of support for branches on a tree is required. To address these problems, I analyzed: (1) each matrix several times with different orderings of taxa and with user-defined trees; (2) each matrix with both algorithms; (3) upper-right and lower-left halves of each matrix separately, performed a complete set of single-taxon jackknifings, and created a jackknife strict consensus of all best-fit trees using both algorithms. Based on these analyses, 97 of 173 interior branches (56%) on the FITCH trees published by Sibley and Ahlquist (1990) were still present in my consensus trees. Many trees remained nearly intact; others collapsed into polytomies. I found that 11 of 97 remaining branches (11%) contradict the Tapestry. I conclude that the data in Sibley and Ahlquist (1990), properly analyzed, have a strong phylogenetic signal.