Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction

Taehyun Hwang, Ze Tian, Rui Kuang, Jean-Pierre Kocher

Research output: Chapter in Book/Report/Conference proceedingConference contribution

23 Citations (Scopus)

Abstract

Building reliable predictive models from multiple complementary genomic data for cancer study is a crucial step towards successful cancer treatment and a full understanding of the underlying biological principles. To tackle this challenging data integration problem, we propose a hypergraph-based learning algorithm called HyperGene to integrate microarray gene expressions and protein-protein interactions for cancer outcome prediction and biomarker identification. HyperGene is a robust two-step iterative method that alternatively finds the optimal outcome prediction and the optimal weighting of the marker genes guided by a protein-protein interaction network. Under the hypothesis that cancer-related genes tend to interact with each other, the HyperGene algorithm uses a protein-protein interaction network as prior knowledge by imposing a consistent weighting of interacting genes. Our experimental results on two large-scale breast cancer gene expression datasets show that HyperGene utilizing a curated roteinprotein interaction network achieves significantly improved cancer outcome prediction. Moreover, HyperGene can also retrieve many known cancer genes as highly weighted marker genes.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE International Conference on Data Mining, ICDM
Pages293-302
Number of pages10
DOIs
StatePublished - 2008
Event8th IEEE International Conference on Data Mining, ICDM 2008 - Pisa, Italy
Duration: Dec 15 2008Dec 19 2008

Other

Other8th IEEE International Conference on Data Mining, ICDM 2008
CountryItaly
CityPisa
Period12/15/0812/19/08

Fingerprint

Gene expression
Proteins
Genes
Oncology
Data integration
Biomarkers
Microarrays
Iterative methods
Learning algorithms

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Hwang, T., Tian, Z., Kuang, R., & Kocher, J-P. (2008). Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction. In Proceedings - IEEE International Conference on Data Mining, ICDM (pp. 293-302). [4781124] https://doi.org/10.1109/ICDM.2008.37

Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction. / Hwang, Taehyun; Tian, Ze; Kuang, Rui; Kocher, Jean-Pierre.

Proceedings - IEEE International Conference on Data Mining, ICDM. 2008. p. 293-302 4781124.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hwang, T, Tian, Z, Kuang, R & Kocher, J-P 2008, Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction. in Proceedings - IEEE International Conference on Data Mining, ICDM., 4781124, pp. 293-302, 8th IEEE International Conference on Data Mining, ICDM 2008, Pisa, Italy, 12/15/08. https://doi.org/10.1109/ICDM.2008.37
Hwang T, Tian Z, Kuang R, Kocher J-P. Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction. In Proceedings - IEEE International Conference on Data Mining, ICDM. 2008. p. 293-302. 4781124 https://doi.org/10.1109/ICDM.2008.37
Hwang, Taehyun ; Tian, Ze ; Kuang, Rui ; Kocher, Jean-Pierre. / Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction. Proceedings - IEEE International Conference on Data Mining, ICDM. 2008. pp. 293-302
@inproceedings{fa5bd59b6e0b424490a98b35c898c719,
title = "Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction",
abstract = "Building reliable predictive models from multiple complementary genomic data for cancer study is a crucial step towards successful cancer treatment and a full understanding of the underlying biological principles. To tackle this challenging data integration problem, we propose a hypergraph-based learning algorithm called HyperGene to integrate microarray gene expressions and protein-protein interactions for cancer outcome prediction and biomarker identification. HyperGene is a robust two-step iterative method that alternatively finds the optimal outcome prediction and the optimal weighting of the marker genes guided by a protein-protein interaction network. Under the hypothesis that cancer-related genes tend to interact with each other, the HyperGene algorithm uses a protein-protein interaction network as prior knowledge by imposing a consistent weighting of interacting genes. Our experimental results on two large-scale breast cancer gene expression datasets show that HyperGene utilizing a curated roteinprotein interaction network achieves significantly improved cancer outcome prediction. Moreover, HyperGene can also retrieve many known cancer genes as highly weighted marker genes.",
author = "Taehyun Hwang and Ze Tian and Rui Kuang and Jean-Pierre Kocher",
year = "2008",
doi = "10.1109/ICDM.2008.37",
language = "English (US)",
isbn = "9780769535029",
pages = "293--302",
booktitle = "Proceedings - IEEE International Conference on Data Mining, ICDM",

}

TY - GEN

T1 - Learning on weighted hypergraphs to integrate protein interactions and gene expressions for cancer outcome prediction

AU - Hwang, Taehyun

AU - Tian, Ze

AU - Kuang, Rui

AU - Kocher, Jean-Pierre

PY - 2008

Y1 - 2008

N2 - Building reliable predictive models from multiple complementary genomic data for cancer study is a crucial step towards successful cancer treatment and a full understanding of the underlying biological principles. To tackle this challenging data integration problem, we propose a hypergraph-based learning algorithm called HyperGene to integrate microarray gene expressions and protein-protein interactions for cancer outcome prediction and biomarker identification. HyperGene is a robust two-step iterative method that alternatively finds the optimal outcome prediction and the optimal weighting of the marker genes guided by a protein-protein interaction network. Under the hypothesis that cancer-related genes tend to interact with each other, the HyperGene algorithm uses a protein-protein interaction network as prior knowledge by imposing a consistent weighting of interacting genes. Our experimental results on two large-scale breast cancer gene expression datasets show that HyperGene utilizing a curated roteinprotein interaction network achieves significantly improved cancer outcome prediction. Moreover, HyperGene can also retrieve many known cancer genes as highly weighted marker genes.

AB - Building reliable predictive models from multiple complementary genomic data for cancer study is a crucial step towards successful cancer treatment and a full understanding of the underlying biological principles. To tackle this challenging data integration problem, we propose a hypergraph-based learning algorithm called HyperGene to integrate microarray gene expressions and protein-protein interactions for cancer outcome prediction and biomarker identification. HyperGene is a robust two-step iterative method that alternatively finds the optimal outcome prediction and the optimal weighting of the marker genes guided by a protein-protein interaction network. Under the hypothesis that cancer-related genes tend to interact with each other, the HyperGene algorithm uses a protein-protein interaction network as prior knowledge by imposing a consistent weighting of interacting genes. Our experimental results on two large-scale breast cancer gene expression datasets show that HyperGene utilizing a curated roteinprotein interaction network achieves significantly improved cancer outcome prediction. Moreover, HyperGene can also retrieve many known cancer genes as highly weighted marker genes.

UR - http://www.scopus.com/inward/record.url?scp=67049160152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67049160152&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2008.37

DO - 10.1109/ICDM.2008.37

M3 - Conference contribution

SN - 9780769535029

SP - 293

EP - 302

BT - Proceedings - IEEE International Conference on Data Mining, ICDM

ER -