LargeScale Phylogenetic Analysis Tandy Warnow Associate Professor Department

  • Slides: 46
Download presentation
Large-Scale Phylogenetic Analysis Tandy Warnow Associate Professor Department of Computer Sciences Graduate Program in

Large-Scale Phylogenetic Analysis Tandy Warnow Associate Professor Department of Computer Sciences Graduate Program in Evolution and Ecology Co-Director The Center for Computational Biology and Bioinformatics The University of Texas at Austin

Outline of Talk • Phylogenetic reconstruction from DNA sequences – the problems, and the

Outline of Talk • Phylogenetic reconstruction from DNA sequences – the problems, and the progress • Phylogenetic reconstruction from gene order and content in whole genomes – initial work • The future of large-scale phylogeny, and the possibilities of inferring the “Tree of Life”

I. Molecular Systematics U AGGGCAT V W TAGCCCA X TAGACTT Y TGCACAA X U

I. Molecular Systematics U AGGGCAT V W TAGCCCA X TAGACTT Y TGCACAA X U Y V W TGCGCTT

DNA Sequence Evolution -3 mil yrs AAGACTT AAGGCCT AGGGCAT TAGCCCA -2 mil yrs TGGACTT

DNA Sequence Evolution -3 mil yrs AAGACTT AAGGCCT AGGGCAT TAGCCCA -2 mil yrs TGGACTT TAGACTT AGCACAA AGCGCTT -1 mil yrs today

Major Phylogenetic Reconstruction Methods • Polynomial-time distance-based methods (neighbor joining the most popular) •

Major Phylogenetic Reconstruction Methods • Polynomial-time distance-based methods (neighbor joining the most popular) • NP-hard sequence-based methods – Maximum Parsimony – Maximum Likelihood • Heated debates over the relative performance of these methods

Quantifying Error FN FN: false negative (missing edge) FP: false positive (incorrect edge) 50%

Quantifying Error FN FN: false negative (missing edge) FP: false positive (incorrect edge) 50% error rate FP

Main Result: DCM-Boosting and DCMNJ+ML We have developed the first polynomial time methods that

Main Result: DCM-Boosting and DCMNJ+ML We have developed the first polynomial time methods that improve upon NJ (with respect to topological accuracy) and are never worse than NJ. The method is obtained through DCM-boosting.

Basis of Distance-Based Methods: Additivity • A distance matrix tree and. is additive if

Basis of Distance-Based Methods: Additivity • A distance matrix tree and. is additive if there exists a such that • Waterman et al. (1977) showed that:

Distance-based Phylogenetic Methods

Distance-based Phylogenetic Methods

Statistical Consistency Atteson (1990) showed that is small enough. if Sequence length Hence NJ

Statistical Consistency Atteson (1990) showed that is small enough. if Sequence length Hence NJ is statistically consistent for many models of evolution. But what about performance on finite sequence lengths?

We focus on performance on finite sequence lengths

We focus on performance on finite sequence lengths

Absolute fast convergence vs. exponential convergence

Absolute fast convergence vs. exponential convergence

General Markov (GM) Model • A GM model tree is a pair – –

General Markov (GM) Model • A GM model tree is a pair – – where is a rooted binary tree. , and is a stochastic substitution matrix with. – The sequence at the root of is drawn from a uniform distribution. – the rates of evolution across the sites can be drawn from a fixed distribution • GM contains models like Jukes-Cantor (JC) and Kimura 2 -Parameter (K 2 P) models.

Absolute Fast Convergence • Let. Define parameterize the GM model: . We • A

Absolute Fast Convergence • Let. Define parameterize the GM model: . We • A phylogenetic reconstruction method is absolute fast-converging (AFC) for the GM model if for all positive there is a polynomial such that for all on set of sequences of length at least generated on , we have

Theoretical Comparison of Early AFC Methods to NJ • Theorem 1 [Warnow et al.

Theoretical Comparison of Early AFC Methods to NJ • Theorem 1 [Warnow et al. 2001] DCMNJ+SQS is absolute fast converging for the GM model. • Theorem 2 [Csűrös 2001] HGT+FP is absolute fast converging for the GM model. • Theorem 3 [Atteson 1999] NJ is exponentially converging for the GM model (but is not known to be AFC).

DCM-Boosting [Warnow et al. 2001] • DCM+SQS is a two-phase procedure which reduces the

DCM-Boosting [Warnow et al. 2001] • DCM+SQS is a two-phase procedure which reduces the sequence length requirement of methods. Exponentially converging method DCM SQS Absolute fast converging method • DCMNJ+SQS is the result of DCM-boosting NJ.

Experimental Comparison of Early AFC Methods to NJ • rbc. L 500 -taxon tree

Experimental Comparison of Early AFC Methods to NJ • rbc. L 500 -taxon tree • Jukes-Cantor model • Avg. branch length = 0. 264

Improving upon early AFC methods • These early AFC methods outperform NJ only on

Improving upon early AFC methods • These early AFC methods outperform NJ only on long enough sequences and on large enough trees with high enough rates of evolution. • Hence we need new fast converging methods which improve upon NJ on more of the parameter space, and are never worse than NJ. • We modify the second phase to improve the empirical performance, replacing SQS with ML (maximum likelihood) or MP (maximum parsimony).

DCMNJ+ML vs. other methods on a fixed tree • 500 -taxon rbc. L tree

DCMNJ+ML vs. other methods on a fixed tree • 500 -taxon rbc. L tree • K 2 P+ model ( =2, =1) • Avg. branch length = 0. 278 • Typical performance

Comparison of methods on random trees as a function of the number of taxa

Comparison of methods on random trees as a function of the number of taxa • Random tree topologies • K 2 P+ model ( =2, =1) • Avg. branch length = 0. 05 • Seq. length = 1000

Summary • These are the first polynomial time methods that improve upon NJ (with

Summary • These are the first polynomial time methods that improve upon NJ (with respect to topological accuracy) and are never worse than NJ. • The advantage obtained with DCMNJ+MP and DCMNJ+ML increases with number of taxa. • In practice these new methods are slower than NJ (minutes vs. seconds), but still much faster than MP and ML (which can take days). • Conjecture: DCMNJ+ML is AFC.

II. Whole-Genome Phylogeny A A C D X B E Y E Z C

II. Whole-Genome Phylogeny A A C D X B E Y E Z C F B D W F

Genomes As Signed Permutations 1 – 5 3 4 -2 -6 or 6 2

Genomes As Signed Permutations 1 – 5 3 4 -2 -6 or 6 2 -4 – 3 5 – 1 etc.

Genomes Evolve by Rearrangements 1 2 3 4 5 6 7 9 10 1

Genomes Evolve by Rearrangements 1 2 3 4 5 6 7 9 10 1 2 3 – 8 9 – 7 -8 4 – 6 – 7 5 – 6 6 -4 – 5 7 – 4 8 9 10 • Inversion (Reversal) • Transposition • Inverted Transposition 8

Genome Rearrangement Has A Huge State Space • DNA sequences : 4 states per

Genome Rearrangement Has A Huge State Space • DNA sequences : 4 states per site • Signed circular genomes with n genes: states, 1 site • Circular genomes (1 site) – with 37 genes: states – with 120 genes: states

Distance-based Phylogenetic Methods for Genomes

Distance-based Phylogenetic Methods for Genomes

Genomic Distance Estimators • Standard: – Breakpoint distance – (Minimum) Inversion distance • Our

Genomic Distance Estimators • Standard: – Breakpoint distance – (Minimum) Inversion distance • Our estimators: We attempt to estimate the actual number of events (the ``true evolutionary distance”): – EDE [Moret et al, ISMB’ 01] – Approx-IEBP [Wang and Warnow, STOC’ 01] – Exact-IEBP [Wang, WABI’ 01]

Breakpoint Distance • Breakpoint distance=5 1 2 3 4 5 6 7 8 9

Breakpoint Distance • Breakpoint distance=5 1 2 3 4 5 6 7 8 9 10 1 – 3 – 2 4 5 9 6 7 8 10

Minimum Inversion Distance • Inversion distance=3 1 2 3 4 8 9 10 1

Minimum Inversion Distance • Inversion distance=3 1 2 3 4 8 9 10 1 2 3 – 8 – 7 – 6 – 5 – 4 9 10 1 8 – 3 – 2 – 7 – 6 – 5 – 4 9 10 1 8 – 3 9 10 7 5 6 7 2 – 6 – 5 – 4

Measured Distance vs. Actual Number of Events Breakpoint Distance Inversion Distance 120 genes, inversion-only

Measured Distance vs. Actual Number of Events Breakpoint Distance Inversion Distance 120 genes, inversion-only evolution

Generalized Nadeau-Taylor Model • Three types of events: – Inversions – Transpositions – Inverted

Generalized Nadeau-Taylor Model • Three types of events: – Inversions – Transpositions – Inverted Transpositions • Events of the same type are equiprobable • Probability of the three types have fixed ratio: Inv : Trp : Inv. Trp = (1 -a-b): a: b

Estimating True Evolutionary Distances for Genomes Given fixed probabilities for each type of event,

Estimating True Evolutionary Distances for Genomes Given fixed probabilities for each type of event, we estimate the expected breakpoint distance after k random events: • Approx-IEBP [Wang, Warnow 2001] – Polynomial-time closed-form approximation to the expected breakpoint distance – Proven error bound • Exact-IEBP [Wang 2001] – Exact, recursive solution for the expected breakpoint distance – Polynomial-time but slower than Approx-IEBP

Estimating True Evolutionary Distances for Genomes (cont. ) Estimating the expected Inversion distance: EDE

Estimating True Evolutionary Distances for Genomes (cont. ) Estimating the expected Inversion distance: EDE [Moret, Wang, Warnow, Wyman 2001] – Closed-formula based upon an empirical estimation of the expected inversion distance after k random events (based upon 120 genes and inversion only, but robust to errors in the model). – Polynomial time, fastest of the three.

Goodness of fit for Approx-IEBP • 120 genes • Inversion-only evolution (similar performance under

Goodness of fit for Approx-IEBP • 120 genes • Inversion-only evolution (similar performance under other models) • EDE and Exact-IEBP have similar performance Approx-

Absolute Difference • 120 genes • Inversion only evolution (Similar relative performance under other

Absolute Difference • 120 genes • Inversion only evolution (Similar relative performance under other models)

Accuracy of Neighbor Joining Using Distance Estimators • 120 genes • Inversion-only evolution •

Accuracy of Neighbor Joining Using Distance Estimators • 120 genes • Inversion-only evolution • 10, 20, 40, 80, and 160 genomes • Similar relative performance under other models

Accuracy of Neighbor Joining Using Distance Estimators • 120 genes • All three event

Accuracy of Neighbor Joining Using Distance Estimators • 120 genes • All three event types equiprobable • 10, 20, 40, 80, and 160 genomes • Similar relative performance under other models

Summary of Genomic Distance Estimators • Statistically based estimation of genomic distances improves NJ

Summary of Genomic Distance Estimators • Statistically based estimation of genomic distances improves NJ analyses • Our IEBP estimators assume knowledge of the probabilities of each type of event, but are robust to model violations • NJ(EDE) outperforms NJ on other estimators, under all models studied • Accuracy is very good, except when very close to saturation

Maximum Parsimony on Rearranged Genomes (MPRG) • The leaves are rearranged genomes. • Find

Maximum Parsimony on Rearranged Genomes (MPRG) • The leaves are rearranged genomes. • Find the tree that minimizes the total number of rearrangement events A A 3 B E C 2 B D 6 C 3 4 Total length = 18 F D

GRAPPA [Bader et al. , PSB’ 01] (Genome Rearrangements Analysis under Parsimony and other

GRAPPA [Bader et al. , PSB’ 01] (Genome Rearrangements Analysis under Parsimony and other Phylogenetic Algorithms) Reimplementation of BPAnalysis [Blanchette et al. 1997] for the Breakpoint Phylogeny problem. • Uses algorithm engineering to improve performance. • Improves the algorithm by reducing the number of tree length evaluations. (Evaluating the length of a fixed tree is NP-hard)

Campanulaceae

Campanulaceae

Analysis of Campanulaceae • 12 genomes + 1 outgroup (Tobacco) • 105 gene segments

Analysis of Campanulaceae • 12 genomes + 1 outgroup (Tobacco) • 105 gene segments • BPAnalysis [Blanchette et al. 1997] over 200 years [Cosner et al. 2000] Using GRAPPA v 1. 1 on the 512 -processor Los Lobos Supercluster machine: 2 minutes = 100 million-fold speedup (200, 000 -fold speedup per processor)

Consensus of 216 MP Trees Strict Consensus of 216 trees; 6 out of 10

Consensus of 216 MP Trees Strict Consensus of 216 trees; 6 out of 10 internal edges recovered. Trachelium Campanula Adenophora Symphandra Legousia Asyneuma Triodanus Wahlenbergia Merciera Codonopsis Cyananthus Platycodon Tobacco

Future Work • New focus on Rare Genomic Changes – New data – New

Future Work • New focus on Rare Genomic Changes – New data – New models – New methods • New techniques for large scale analyses – Divide-and-conquer methods – Non-tree models – Visualization of large trees and large sets of trees

Acknowledgements • Funding: The David and Lucile Packard Foundation, The National Science Foundation, and

Acknowledgements • Funding: The David and Lucile Packard Foundation, The National Science Foundation, and Paul Angello • Collaborators: Robert Jansen (U. Texas) Bernard Moret, David Bader, Mi-Yan (U. New Mexico) Daniel Huson (Celera) Katherine St. John (CUNY) Linda Raubeson (Central Washington U. ) Luay Nakhleh, Usman Roshan, Jerry Sun, Li-San Wang, Stacia Wyman (Phylolab, U. Texas)

Phylolab, U. Texas Please visit us at http: //www. cs. utexas. edu/users/phylo/

Phylolab, U. Texas Please visit us at http: //www. cs. utexas. edu/users/phylo/