Comparative Genomics: International Workshop, RECOMB-CG 2009, Budapest, Hungary, September 27-29, 2009. Proceedings

249 Pages·2009·4.806 MB·English

by  Eric Tannier (auth.), Francesca D. Ciccarelli, István Miklós (eds.)

Lecture Notes in Bioinformatics 5817
EditedbyS.Istrail,P.Pevzner,andM.Waterman
EditorialBoard: A.Apostolico S.Brunak M.Gelfand
T.Lengauer S.Miyano G.Myers M.-F.Sagot D.Sankoff
R.Shamir T.Speed M.Vingron W.Wong
Subseries of Lecture Notes in Computer Science
Francesca D. Ciccarelli
István Miklós (Eds.)
Comparative
Genomics
International Workshop, RECOMB-CG 2009
Budapest, Hungary, September 27-29, 2009
Proceedings
1 3
SeriesEditors
SorinIstrail,BrownUniversity,Providence,RI,USA
PavelPevzner,UniversityofCalifornia,SanDiego,CA,USA
MichaelWaterman,UniversityofSouthernCalifornia,LosAngeles,CA,USA
VolumeEditors
FrancescaD.Ciccarelli
EuropeanInstituteofOncology
IFOM-IEOCampus,ViaAdamello,16,20139Milan,Italy
E-mail:[email protected]
IstvánMiklós
HungarianAcademyofSciences
RényiInstitute
Reáltanodautca13-15,1053Budapest,Hungary
E-mail:[email protected]
LibraryofCongressControlNumber:2009936053
CRSubjectClassification(1998):F.2,G.3,E.1,H.2.8,J.3
LNCSSublibrary:SL8–Bioinformatics
ISSN 0302-9743
ISBN-10 3-642-04743-2SpringerBerlinHeidelbergNewYork
ISBN-13 978-3-642-04743-5SpringerBerlinHeidelbergNewYork
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis
concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting,
reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication
orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965,
initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable
toprosecutionundertheGermanCopyrightLaw.
springer.com
©Springer-VerlagBerlinHeidelberg2009
PrintedinGermany
Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India
Printedonacid-freepaper SPIN:12769131 06/3180 543210
Preface
Asgenome-sequencingcostscontinuetheirdownwardspiral,sequencingofclosely
related organisms has become increasingly affordable. The growing amount of
genomicdata availabledemands for the constantdevelopmentof computational
toolstobeappliedincomparativegenomics.TheRECOMBWorkshoponCom-
parative Genomics (RECOMB-CG) is devoted to bringing together scientists
workingonallaspectsofcomparativegenomics,fromcomputerscientists,math-
ematicians and statisticians working on novel computational approaches for
genome analysis and comparison, to biologists applying these computational
tools to study the structure and the evolution of prokaryotic and eukaryotic
genomes.
Thisvolumecontainsthe19paperspresentedatthe7thAnnualRECOMB-CG
workshopheldduringSeptember27–29,2009attheRenyiInstitute,inBudapest,
Hungary. The papers published in these proceedings were selected for oral pre-
sentationfrom31submissionsfromscientistsaroundthe world.Eachpaperwas
reviewedbyatleastthreemembersoftheProgramCommitteeinastringentand
thoughtfulpeer-reviewprocess.
The conference itself was enlivened by invited keynote presentations from
Richard Durbin (Welcome Trust Sanger Institute), Jotun Hein (Department
of Statistics, University of Oxford, UK), Laurence Hurst (Department of Biol-
ogy and Biochemistry, University of Bath, UK), Csaba Pa´l (Evolutionary Sys-
temsBiologyGroup,BiologicalResearchCenter,Szeged,Hungary),LisaStubbs
(School of Molecular and Cellular Biology, University of Illinois at Urbana-
Champaign) and Jeroen Raes (EMBL Heidelberg). These talks were supple-
mented by both the presentations of the papers in this volume and a poster
session. Together, talks and papers highlighted the state-of-the-art of compar-
ative genomics tools and applications. From the inference of evolution in ge-
neticregulatorynetworks,tothedivergentfatesofgeneandgenomeduplication
events, to the importance of new computational approaches to unraveling the
structuralevolutionof genomes,these presentationsillustrate the crucialrole of
comparative genomics in understanding genome function.
RECOMB-CG 2009 would not have been possible without the participation
of the many scientists who contributed their time and effort to making the con-
ference a success. We thank the scientists who submitted their work for presen-
tation at the conference, those members of the Program Committee who made
everyefforttoensurefairandbalancedreviewofthemanypaperssubmittedfor
consideration,the members of the localOrganizingCommittee for arrangingall
themyriaddetailsoftheorganizationalaspectsoftheevent,andthemembersof
the Steering Committee for their ongoing dedication and guidance. RECOMB-
CG 2009 is also deeply indebted to its sponsors including the National Office
VI Preface
for Research and Technology, the Hungarian Academy of Sciences, the Renyi
Institute and the Hungarian Society for Bioinformatics.
It is the continued support and dedication of this community that allows
RECOMB-CG to bring together comparative genomics researchers from across
the globe to exchange ideas and information and focus the force of comparative
genomics on improving our understanding of genome evolution and function.
August 2009 Francesca Ciccarelli
Istvan Mikl´os
Table of Contents
Yeast Ancestral Genome Reconstructions: The Possibilities of
Computational Methods .......................................... 1
Eric Tannier
Natural Parameter Values for Generalized Gene Adjacency............ 13
Zhenyu Yang and David Sankoff
Parking Functions, Labeled Trees and DCJ Sorting Scenarios.......... 24
A¨ıda Ouangraoua and Anne Bergeron
Counting All DCJ Sorting Scenarios................................ 36
Mar´ılia D.V. Braga and Jens Stoye
Minimal Conflicting Sets for the Consecutive Ones Property in
Ancestral Genome Reconstruction.................................. 48
Cedric Chauve, Utz-Uwe Haus, Tamon Stephen, and Vivija P. You
Finding Nested Common Intervals Efficiently ........................ 59
Guillaume Blin and Jens Stoye
DCJ Median Problems on Linear Multichromosomal Genomes: Graph
Representation and Fast Exact Solutions............................ 70
Andrew Wei Xu
RearrangementModels and Single-Cut Operations ................... 84
Paul Medvedev and Jens Stoye
Aligning Two Genomic Sequences That Contain Duplications.......... 98
Minmei Hou, Cathy Riemer, Piotr Berman, Ross C. Hardison, and
Webb Miller
Inferring the Recent Duplication History of a Gene Cluster............ 111
Giltae Song, Louxin Zhang, Toma´ˇs Vinaˇr, and Webb Miller
Pseudo Boolean Programming for Partially Ordered Genomes ......... 126
S´ebastien Angibaud, Guillaume Fertin, Annelyse Th´evenin, and
St´ephane Vialette
Computing the Summed Adjacency Disruption Number between Two
Genomes with Duplicate Genes Using Pseudo-BooleanOptimization.... 138
Jo˜ao Delgado, Inˆes Lynce, and Vasco Manquinho
Reconstructing Histories of Complex Gene Clusters on a Phylogeny .... 150
Toma´ˇs Vinaˇr, Bronˇa Brejova´, Giltae Song, and Adam Siepel
VIII Table of Contents
Co-evolutionary Models for Reconstructing Ancestral Genomic
Sequences: Computational Issues and Biological Examples ............ 164
Tamir Tuller, Hadas Birin, Martin Kupiec, and Eytan Ruppin
Whole-Genome Analysis of Gene Conversion Events.................. 181
Chih-Hao Hsu, Yu Zhang, Ross Hardison, and Webb Miller
A Statistically Fair Comparison of Ancestral Genome Reconstructions,
Based on Breakpoint and RearrangementDistances .................. 193
Zaky Adam and David Sankoff
Comparative Genomics and Extensive Recombinations in Phage
Communities .................................................... 205
Guylaine Poisson, Mahdi Belcaid, and Anne Bergeron
Properties of Sequence Conservation in Upstream Regulatory and
Protein Coding Sequences among Paralogs in Arabidopsis thaliana ..... 217
Dale N. Richardson and Thomas Wiehe
Transcription Factor Binding Probabilities in Orthologous Promoters:
An Alignment-Free Approachto the Inference ofFunctional Regulatory
Targets ......................................................... 229
Xiao Liu and Neil D. Clarke
Author Index.................................................. 241
Yeast Ancestral Genome Reconstructions:
The Possibilities of Computational Methods
Eric Tannier
INRIARhˆone-Alpes, Universit´edeLyon, F-69000, Universit´e Lyon 1, CNRS,UMR
5558, Laboratoire deBiom´etrie et Biologie E´volutive, F-69622, Villeurbanne, France
Abstract. In 2006, a debate has risen on the question of theefficiency
ofbioinformaticsmethodstoreconstructmammalianancestralgenomes.
Threeyearslater,Gordonetal.(PLoSGenetics,5(5),2009)chosenotto
useautomaticmethodstobuildupthegenomeofa100million yearold
Saccharomyces cerevisiae ancestor. Their manually constructed ances-
tor provides a reference genome to test whether automatic methods are
indeed unable to approach confident reconstructions. Adapting several
methodological frameworks to thesame yeast gene order data, I discuss
the possibilities, differences and similarities of the available algorithms
forancestralgenomereconstructions.Themethodscanbeclassifiedinto
two types: local and global. Studying the properties of both helps to
clarifywhatwecanexpectfromtheirusage.Bothmethodsproposecon-
tiguous ancestral regions that come very close (> 95% identity) to the
manually predicted ancestral yeast chromosomes, with a good coverage
of theextant genomes.
1 Introduction
The reconstruction of ancestral karyotypes and gene orders from homologies
between extant species is a long-standing problem [7]. It helps to understand
the large-scale evolutionary mutations that differentiate the present genomes.
Computational methods to handle gene order and propose ancestral genome
architectures have a shorter [17], but prolific [8] history. However, despite the
numerous efforts of the computational biology community, two recent rounds
of publications have put a doubt on their efficiency. In 2006, comparing ances-
tralboreoeutheriangenomereconstructionsmade bycytogeneticistsonone side
and bioinformaticians on the other, Froenicke et al. [9] found that the man-
ually constructed and expertized cytogenetics one was not acceptably recov-
ered by computational biology. This provoked many comments [2,15] published
the same year. Since then, several bioinformatics teams have tried to approach
the manually obtained boreoeutherian ancestor [13,4,1]. In may 2009, Gordon
et al. [10] published the refined configuration of the genome of an ancestor of
Saccharomyces cerevisiae (whichis approximatelyas ancientas the boreoeuthe-
rianancestor),andchosenottouseanycomputationalframework,arguingthat
thosearestillindevelopmentandyetcannothandletheavailabledata.Itseems
tohavebeenconfirmedbythepublication,afewweekslater,bytheGenolevures
F.D.CiccarelliandI.Mikl´os(Eds.):RECOMB-CG2009,LNBI5817,pp.1–12,2009.
(cid:2)c Springer-VerlagBerlinHeidelberg2009
2 E. Tannier
consortium [18], of a wide yeast genome comparative study including a recon-
structionofanancestralconfigurationofsomenonduplicatedspecies,withayet
unpublished method [11]. Thus the ancestor of duplicated species constructed
by Gordon et al. [10] is still inaccessible by this approach. Sankoff [16] even-
tually wrote a comment detailing the deficiencies of computational approaches,
yet with an optimistic conclusion for the future.
But right now there still seems to be a gap between formal methods and the
application they are made for. This paper intends to fill this gap, and presents
efficient implementations of automatic methods on yeast genomes.
In [4], a new computational framework to reconstruct ancestral genomes was
presented, generalizing a method by Ma et al. [13]. Whereas the earlier meth-
odswereallbasedonadistanceminimizationprinciplethatisneverusedinthe
manualreconstructions,these onestendtomimic the principlesusedbycytoge-
neticists on mammalian data, which appear to be very close the the ones used
by Gordon et al. [10] on yeast genomic data. This framework can handle the
presence of a whole genome duplication in the history of the available species,
as it was already remarkedin [14], a study using teleost fish genomes.
It was shown in [4] that implementing this method on mammalian genomes
resulted in a boreoeutherian ancestor which is very close to the one proposed
by cytogeneticists. It shows in particular more similarity than the results of
the distance based methods. Here I applied two different frameworks (the one
describedin[4]andthedistancebased“GuidedGenomeHalving”ofZhengetal.
[20])onthesameyeastdatausedbyGordonetal. [10]andcomparedtheresults
with the manually constructed ancestral configuration. I used genomic markers
that cover a large part of the extant genomes, and which allow to apply both
methodsonthe sameinstance.Analyzingthedifferencesbetweenthe three(one
manually and two computationally) constructed ancestors helps to understand
the behavior of automatic methods and show what can be expected from them
at the present time.
2 The Two Kinds of Methods
In this section, I describe the concepts of the two kinds of methods, and why
they often give different results. These reasons, to my mind, are an important
part of the terms of the debates that have sometimes opposed biologists and
bioinformaticians[9,2,15]. More details on the local method canbe found in [4],
and on global methods in [8].
2.1 The Local Method
The local method consists in three steps:
– identifying ancestral markers and their homologs in extant genomes;
– searching for ancestral syntenies, i.e. sets of ancestral markers that are be-
lieved to be contiguous in the ancestral genome, and weighting these an-
cestral syntenies according to the confidence put in their presence in the
ancestral genome, guided by its phylogenetic signal;