動的ネ トワ ク抽出のための動的ネ トワ ク抽出のための動的ネットワ ク抽出のための動的ネットワ ク抽出のための動的ネットワーク抽出のための動的ネットワーク抽出のための宮野宮野 悟悟 動的ネットワ ク抽出のための動的ネットワ ク抽出のための宮野宮野 悟悟 動的ネットワ ク抽出のための動的ネットワ ク抽出のための宮野宮野 悟悟

イ シ パイプ イ 構築イ シ パイプ イ 構築宮野宮野 悟悟

イン シリコパイプラインの構築イン シリコパイプラインの構築東京大学医科学研究所東京大学医科学研究所 イン・シリコパイプラインの構築イン・シリコパイプラインの構築東京大学医科学研究所東京大学医科学研究所 イン・シリコパイプラインの構築イン・シリコパイプラインの構築東京大学医科学研究所東京大学医科学研究所 イン シリコパイプラインの構築イン シリコパイプラインの構築イン シリ イ ライン 構築イン シリ イ ライン 構築N ki M I t S F jit A S it A Ik d E Shi T Y hi R U K Mi SNagasaki M. , Imoto S., Fujita A., Saito A. , Ikeda E. , Shimamura T., Yamaguchi R., Ueno K., Miyano S.Nagasaki M. , Imoto S., Fujita A., Saito A. , Ikeda E. , Shimamura T., Yamaguchi R., Ueno K., Miyano S.

Human Genome Center Institute of Medical Science University of TokyoHuman Genome Center, Institute of Medical Science, University of Tokyo, , y y

Abstract: Several technologies are currently used for gene expression profiling such asAbstract: Several technologies are currently used for gene expression profiling such asQ tit ti R l Ti RT PCR ( RT PCR) ill i i d C A l i fQuantitative Real Time RT-PCR (qRT-PCR), illumina microarray and Cap Analysis ofB (q ) y p yGene Expression (CAGE) CAGE is a recently developed method for constructingA B

C Gene Expression (CAGE). CAGE is a recently developed method for constructingtranscriptome maps and since its invention it has been successfully applied to analyzing

A C transcriptome maps and since its invention, it has been successfully applied to analyzinggene expressions in diverse biological studies. The principle of CAGE has beeng p g p pdeveloped to address specific issues such as determination of transcriptional starting sitesdeveloped to address specific issues such as determination of transcriptional starting sites,th t d f t i d id tifi ti f t i t H t C llthe study of promoter regions and identification of new transcripts. Here, we present a CellSystem Markup Language (CSML) computational pipeline to analyze high-throughputSystem Markup Language (CSML) computational pipeline to analyze high throughputdata derived from CAGE microarray and qRT PCR This pipeline performs both standarddata derived from CAGE, microarray and qRT-PCR. This pipeline performs both standard

i i l l i d h l bli h d h d f i f lD statistical analysis and the most recently published methods of construction of regulatoryD y y p g ynetworks and data assimilationnetworks and data assimilation.E CSML i li h CSML i li iE CSML pipeline scheme. CSML pipeline receivesas input, an EDF (Expression Data Format) file andas input, an EDF (Expression Data Format) file andperforms several computational and statistical analysisperforms several computational and statistical analysisi l di t ti f l t t k d d tincluding construction of regulatory networks and datag g yassimilation Results may be visualized using Cellassimilation. Results may be visualized using CellIllustratorIllustrator.

Bayesian networkBayesian networkCSML pipeline

yp p

graphical interfacegraphical interface. Th t t

$encoding UTF‐8

The user constructs $vers ion 1$primarykey time$ himself the pipeline, $replicateID repID$typeID type$ i T i

se e p pe e,just dragging and

$experimentType microarray$chipID$ hi V i just dragging and

d i th$chipVersion$chipCompany illumina$description dropping the $description$date 12/2/2008@repID 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

components@repID 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1@time 0 1 2 3 4 5 6 8 10 12 15 18 21 24 28@type exp exp exp exp exp exp exp exp exp exp exp exp exp exp exp components.@type exp exp exp exp exp exp exp exp exp exp exp exp exp exp exp6960451 54.50 33.31 21.00 33.93 41.85 22.63 60.35 40.71 60.06 87.05 130.68 131.71 111.81 94.13 133.062850504 2 82 2 76 3 63 2 74 2 86 2 94 3 59 3 76 3 07 3 89 3 68 2 88 2 58 3 89 3 232850504 2.82 2.76 3.63 2.74 2.86 2.94 3.59 3.76 3.07 3.89 3.68 2.88 2.58 3.89 3.237560397 3.49 3.33 2.89 2.93 2.63 3.18 4.79 5.68 5.50 3.54 7.78 4.17 3.10 3.52 2.855360747 3.33 2.42 2.50 2.69 3.07 2.38 3.62 2.64 2.15 2.57 2.78 3.16 2.93 3.50 2.795360747 3.33 2.42 2.50 2.69 3.07 2.38 3.62 2.64 2.15 2.57 2.78 3.16 2.93 3.50 2.792600731 195.14 594.52 810.50 548.09 351.30 393.13 361.60 492.52 412.80 375.94 252.54 319.30 248.54 281.30 170.643990110 5.82 6.61 5.12 4.77 6.92 4.49 4.29 4.19 3.82 4.05 3.17 4.94 4.82 4.17 5.89 Database scheme.2120309 1.78 1.85 1.86 2.40 2.11 2.21 2.22 2.79 2.10 2.50 1.27 2.61 1.93 2.83 2.276420474 55.43 33.31 39.72 45.25 52.73 25.88 45.33 26.68 26.29 44.56 57.49 45.68 40.73 48.42 58.12

Database scheme.In the actual version670095 3.05 2.59 2.96 2.51 4.75 3.59 2.17 2.46 2.06 2.13 2.75 2.08 2.33 2.20 2.28

130367 6.74 4.97 2.98 5.68 6.85 3.38 4.63 4.38 4.05 4.85 3.66 4.65 3.82 5.47 5.94In the actual version

.

. (version 2), this

.( ),database containsdatabase contains i f ti l t d tExperiment

Promoter

information related to Experiment

Gene State SpaceInput file The input is an EDF (Expression Data experiment, probe, Probe State Space Input file. The input is an EDF (Expression DataF t) fil hi h t i i f ti b t th

e pe e , p obe,gene promoter regionSNP Model networkFormat) file which contains information about the gene, promoter region,

i l lExpression Model network

technology used in the experiment in the header expression level, tec o ogy used t e e pe e t t e eade(initialized by the “$” symbol) information about the transcription factor(initialized by the $ symbol), information about thet f i t i th tt ib t (i iti li d b th

transcription factor binding sites SNP

TFBS

type of experiment in the attributes (initialized by the binding sites, SNP, “@” symbol) and finally the expression data If a line is exon, proteins, mRNA @ symbol) and finally the expression data. If a line isinitialized with the symbol “#” it comments out the

, p ,and ExonArray Theinitialized with the symbol # , it comments out the and ExonArray. The d t b i ilExonArray

respective line. database is easily ProteinmRNAExonp

extendable , and futureextendable , and future versions aim to includeversions aim to include i f ti b tinformation about gene gontology and proteinontology and protein arraysarrays.

OSMOSM

FOSBFOSB

FOSBFOSBFOSB

Data Assimilation pipeline. Data NR4A216

18

IFNB1

350

400

MYD886000

TNF

90

100

SOCS170

IRAK2Data Assimilation pipeline. Data

Assimilation technology (combination ofNR4A2

10

12

14

16

250

300

350

4000

5000

60

70

80

90

40

50

60 Assimilation technology (combination ofli d i i bi l i l i f i

4

6

8

100

150

200

1000

2000

3000

20

30

40

50

20

30 literature data, a priori biological information0

2

1 2 3 4 5 6 7 8

0

50

1 2 3 4 5 6 7 8

0

1000

1 2 3 4 5 6 7 8

0

10

20

1 2 3 4 5 6 7 8

0

10

1 2 3 4 5 6 7 8

p gand mathematical model) is used to improveand mathematical model) is used to improvethe biological pathway model to be more CAGE expression data CAGEthe biological pathway model to be more CAGE expression data. CAGE

h l i i li kreasonable pathway model by using the technology permits to visualize a networkp y y gobserved data e g mRNA expression data

gy pof transcripts instead of a network of

D i d l Th t t i i li d i C ll Ill t t Additi lobserved data, e.g. mRNA expression data. of transcripts instead of a network of

genesDynamic model. The output is visualized using Cell Illustrator. Additional genes.simulations and analysis will be performed in this platform. The network is

Supported by: Genome Network Project from the Ministry of Education Culturesimulations and analysis will be performed in this platform. The network isthe signal transduction and gene regulatory network model with the LPS Supported by: Genome Network Project from the Ministry of Education, Culture,

S t S i d T h l Jthe signal transduction and gene regulatory network model with the LPS

Sports, Science and Technology, Japan.stimulation to macrophage. p , gy, pp g

aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa