ACE

1
ACE RACEannotation of complex/combinatorialexp
ressions
2
Self-introduction
Andrey Zinovyev
M.Sc. in theoretical physics (1997)
Programming, industrial information systems (C,
Delphi)
Ph.D. in computer science (2001), Method of
elastic maps and applications in bioinformatics
Web-services development (Java, JSP)
Senior postdoctoral fellow in IHES, France
http//www.ihes.fr/zinovyev or type zinovyev
in Google
3
Plan of the talk
? ACE framework introduction what we have
? What will be in RACE?
? ACE software C code web-application
? Plans for ACE and RACE
? Computational environment
4
Genome as databaseeverything is annotation
Genomes human, chimp, mouse, rat
ATGCGTGCAAATGCTCTTTGTGTAACGTGTCGACGTACGTGTGTAACGTG
CGACGTACGT
common format for annotation files (binary
p-files)
5
Genome preprocessingcompile once, run everywhere
ATGCGTGCAAATGCTCTTTGTGTAACGTGTCGACGTACGTGTGTAACGTG
CGACGTACGT
ace.annotate
ace.RNAtools ace.annotate
arc
ace.map
r.ace Potential RNA structures, splicing sites
b.ace Potential TF binding sites
m.ace Gene expression data
c.ace Chromatin structure and dynamics
ace.enhance
ace.cluster
ace.display
ace.dyCr ace.stat
6
Structure space the truth is out there
set of annotations
Multidimensional combinatorial space of all
possible structures appearing in a scanning window
7
ace.enhancebe more abstract
Accessing and masking structure space
? view in genome browser (ace.display)

• compare with experiment
• (cross-annotation)
• (ace.dyCr)

• construct more abstract
• space and apply
• ace.enhance further

8
b.ace
Transfac release
Genome release
ace.annotate
b.ace 1.2Tbyte
9
ace.enhance

• Enhance methods
• Fixed spacing of sites
• Fixed order of sites
• Fixed strand orientation of sites
• Multiple copies of site
• Minimal spacing of sites
• Maximal spacing of sites
• Variable, defined spacing between sites
• Minimal p-value for weight matrix
• Maximal p-value for weight matrix
• Bias weight-matrix

M1M2M3M4M5
ace.cluster simplified version of enhance
for detecting clusters of repetitions of one motif
10
Example14 transcription factors, chr14 of
UCSC_HG15
11
Example2clusters of motifs, chr14
ace.enhance expression, window 300bp jfl_im ? 10
copies
Result 51 hits in 5 groups
12
ACE C tools
? aceLib, wraps system-dependent code
? generic programming for code reusability
ace.annotate probability based annotations
and motifs search
ace.enhance accessing (masking) structure
space combinatorial query language
ace.cluster extracting clusters of
repetitions simplified version of enhance
ace.dyCr first step in structure space
analysis dynamic cross-annotation
ace.stat statistical significance analysis
13
ACE web-application (JSP)ace.uit
14
database layout .ace
15
modules layout ace.rte/ace.annotate
16
modules layout ace.rte/ace.enhance
17
data layout my.ace
18
documentation layout ace.doc
19
Plans with ACEprincipal problem
ace.stat statistical model of random noise
maximum entropy principle significance
analysis
20
Plans with ACEvisualizing structure space
creating 2D maps of structure space
data visualization, dimension reduction
21
Plans with ACEintegrating m.ace
ace.map
m.ace
22
Plans with ACEmodel of chromatin structure and
dynamics
chromatin state profiles
arc
c.ace imunoprecipitation experiments
23
Plans with ACEcomparative genomics
genome1
genome2
24
Installation of b.ace in Lillehttp//ace.ibl.fr
1.2 Tbyte PowerVault storage PowerEdge Dell server
25
Installation of RACE in Sherbrooke (golf)
LISA DB
UCSC local
ace
UCSC browser
r.ace DB
G browser
26
Distributed environmentdatabase synchronization
protocol
b.ace Lille France
LISA Sherbrooke Canada
r.ace Sherbrooke Canada
m.ace INSERM Paris
c.ace IHES Paris
public dbs
27
RACE platform for integration
ace.annotate find simple motifs (loops,
hairpins)
ace.RNAtools pluggable algorithms
p-files (r.ace database)
28
ACE team
ace team leader Arndt Benecke, IHES
ace.uit, ace C Andrey Zinovyev, IHES
aceLib, ace C Thomas Bücher, Inst.Neur.
ace.map Sebastian Noth, INSERM
ace.stat Richard Madden, UdSh
arc Graham Smith, IHES