Distinguishing Regulators of Biomolecular Pathways Mentor: Dr. Xiwei Wu City of Hope Sean Caonguyen SoCalBSI 8/21/08.

Distinguishing Regulators of Biomolecular Pathways

Mentor:Dr. Xiwei WuCity of Hope

Sean CaonguyenSoCalBSI8/21/08

Expression Pattern Analysis

Microarray technology is a powerful tool for investigating cellular activity at different levels

DNA microarrays can be used to identify genetic ‘‘signatures’’ for disease

http://www.sciencedaily.com/images/2007/09/070912102212.jpg

Pan et al. (2005)

A Traditional Approach to DNA Microarray Analysis

Gene Expression Data

Gene Selected

Biological Interpretation

Threshold

Individual Gene Analysis Two step process

Selects genes from an arbitrarily chosen cut-off

From the selected genes, one infers biological meaning of gene expression data

Jiang Z and Gentlemen R. (2006) and Nam D, et al. (2007)

Emerging Approach to DNA Microarray Analysis

Gene Set Analysis (GSA) Rank all genes based on

their phenotype association

Calculate a maximal enrichment score for each gene set

Rank each gene set score for biological interpretation

Gene Expression Data

Gene Set Database

Biological Interpretation

Assess gene set directly

Jiang Z and Gentlemen R. (2006) and Nam D, et al. (2007)

Biological Significance of Gene Set Analyses

Ability to identify subtle changes in gene expression that are undetectable by traditional approaches

No arbitrary threshold

Generate results that are easier to interpret

Current Problem with GSA

Reduces gene set into a list of names

No difference in up-regulation and down-regulation Directionality is lost

A

B

D E

F

C

P

G

A

D

F

P

E

B

Suggests that the pathway is activatedSuggests a lower probability of pathway activation

HIGHER

up-regulation

down-regulation

up-regulation

Enriched Gene Set Analysis

Gene Set Database

Biological Interpretation

Assess gene set directly

Curated Analysis

Gene Expression Data

Useful Tools for the Pathway Analysis Program

National Cancer Institutes (NCI) Pathway Interaction Database (http://pid.nci.nih.gov/PID/index.shtml)

contains information about molecular interactions and biological processes in signaling pathways

focuses on cancer research in human cells searches for biomolecules, processes, or by

viewing pathways Data format

Graphics: SVG or GIF Texts: XML or BioPax

Segment of the Phosphoinositide 3-Kinases (PI3K) Signaling Pathway

Key to Icons

non-lipid kinase pathway of Class IB PI3K

XML Script

Project Objective

Create a program to distinguish the activators and inhibitors in each signaling pathway Requires extensive use of XML Parser in Python

Approach to Project

1. Identify all the elements in the pathway

2. Record the pairwise interactions

Linking each interaction

3. Determine the role of each molecule

Finding each leaf node Using a traceback

method

A

B

D E

F

C

P

G

1) Identify the Elements in the Pathway

Properly assign each ID to reference a “preferred symbol”

Locate each interaction ID

2) Record the Pairwise Interactions

How to can we store each interaction? Memory efficient Easy extraction of

data

A

B

D E

F

C

P

GSparse Matrix!

Sparsing Matrix Initialization

A

B

D E

F

C

P

G

  A B C D E F G P

A 0 0 0 0 0 0 0 0

B 1 0 0 0 0 0 0 0

C 0 0 0 0 0 0 0 0

D 0 1 1 0 0 0 0 0

E 0 -1 0 0 0 0 0 0

F 0 0 0 1 0 0 0 0

G 0 0 0 0 -1 0 0 0

P 0 0 0 0 0 1 0 0

Sparse Matrix

Regulators

Ou

tpu

t

1

1

11

1

-1

-1

3) Determine the Role of Each Molecule

A

B

D E

F

C

P

G

  A B C D E F G P

A 0 0 0 0 0 0 0 0

B 1 0 0 0 0 0 0 0

C 0 0 0 0 0 0 0 0

D 0 1 1 0 0 0 0 0

E 0 -1 0 0 0 0 0 0

F 0 0 0 1 0 0 0 0

G 0 0 0 0 -1 0 0 0

P 0 0 0 0 0 1 0 0

Regulators

Ou

tpu

t

Traceback each leaf node

1

1

11

1

-1

-1

Leaf Node P

Role Activator Inhibitor

Protein A,B,C,D,F

Identify each leaf node

Leaf Node G

Activator Inhibitor

A,B E

Locate Activated Pathways for Better Biological Interpretation

Gene Expression Data Up-regulation of B and D Down-regulation of E

Enriched Gene Set Analysis

Leaf Node P Leaf Node G

Role Activator Inhibitor Activator Inhibitor

Protein A,B,C,D,F A,B E

Possible activation of Pathway

A

B

D E

F

C

P

G

D E

B down-regulationup-regulation

Results

For each pathway menu, one can: find a list of proteins with associated roles for

each node look at each protein in an interaction find a list of all interactions in a pathway

Percentage of Inhibitors

Number of Pathwa

ys Percentage

0% 55 46.6%

0-5% 38 32.2%

>=5% 25 21.2%

>=10% 8 6.8%

>=20% 2 1.7%

Total 118 100%

Conclusion

Successfully parse XML files Pathway analysis program works ~50% of pathways include inhibitors 20% of the pathways contains >=5% of inhibitors

Average total molecules = 60

Future Directions

Improvements to Software Ambiguous roles Proteins in different Complex

may have different roles Fine tune the overall role of

proteins in each pathway

Run program with real expression data set

Improve prognoses and drugs for diseases

A

B

D E

F

C

P

G

References Pan KH, Lih Cj, Cohen SN. Effects of threshold choice on biological conclusions

reached during analysis of gene expression by DNA microarrays. Proc Natl Acad Sci 2005, 102:8961-5.

Subramanian A, Tamayo P, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci 2005, 102:15545-50.

Nam D, Kim SY. Gene-set approach for expression pattern analysis. Brief Bioinform 2008, 9:189-97.

Dupuy A, Simon RM. Critical review of published microarray studies for cancer outcome and guidelines on statistical analysis and reporting. J Natl Cancer Inst 2007, 99:147-57.

Jiang Z, Gentleman R. Extensions to gene set enrichment. Bioinformatics 2007,23:306-13.

Dinu I, Potter JD, et al. Improving gene set analysis of microarray data by SAM-GS. BMC Bioinformatics 2007, 8:242.

Liu Q, Dinu I, et al. Comparative evaluation of gene-set analysis methods. BMC Bioinformatics 2007,8:431.

Acknowledgements Mentor

Xiwei Wu

SoCalBSI Faculty and Staff Jamil Momand Sandy Sharp Nancy Warter-Perez Wendie Johnston

Funding for SoCalBSI: DOE and NASA LA / Orange County

Biotechnology Center NSF, NIH, and Economic &

Workforce Development

Funding at City of Hope: National Cancer Institute National Institute of Health