{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

CODENSE - BIOINFORMATICS Vol 21 Suppl 1 2005 pages i213i221...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
BIOINFORMATICS Vol. 21 Suppl. 1 2005, pages i213–i221 doi:10.1093/bioinformatics/bti1049 Mining coherent dense subgraphs across massive biological networks for functional discovery Haiyan Hu 1 , Xifeng Yan 2 , Yu Huang 1 , Jiawei Han 2 and Xianghong Jasmine Zhou 1, 1 Program in Molecular and Computational Biology, University of Southern California, Los Angeles, CA 90089, USA and 2 Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA Received on January 15, 2005; accepted on March 27, 2005 ABSTRACT Motivation: The rapid accumulation of biological network data translates into an urgent need for computational methods for graph pattern mining. One important problem is to identify recurrent patterns across multiple networks to discover bio- logical modules. However, existing algorithms for frequent pattern mining become very costly in time and space as the pattern sizes and network numbers increase. Currently, no effi- cient algorithm is available for mining recurrent patterns across large collections of genome-wide networks. Results: We developed a novel algorithm, CODENSE, to efficiently mine frequent coherent dense subgraphs across large numbers of massive graphs. Compared with previous methods, our approach is scalable in the number and size of the input graphs and adjustable in terms of exact or approxim- ate pattern mining. Applying CODENSE to 39 co-expression networks derived from microarray datasets, we discovered a large number of functionally homogeneous clusters and made functional predictions for 169 uncharacterized yeast genes. Availability: http://zhoulab.usc.edu/CODENSE/ Contact: [email protected] 1 INTRODUCTION The recent development of high-throughput technologies provides a range of opportunities to systematically character- ize diverse types of biological networks. ‘Network Biology’ hasbeenanemergingfieldinbiology. Thevarietyofbiological networks can be classified into two categories: (1) phys- ical networks, which represent physical interactions among molecules, e.g. protein-interaction, protein–DNA interaction and metabolic reactions and (2) conceptual networks, which represent functional associations of molecules derived from genomic data, e.g. co-expression relationships extracted from microarray data and genetic interactions obtained from syn- thetic lethality experiments. While the physical network data To whom correspondence should be addressed. are as yet very limited in size, the large amount of microar- ray data allows us to infer conceptual functional associations of genes under various conditions for many model organisms, thus providing valuable information to study the functions and the dynamics of biological systems. Studying the building principles of biological networks could potentially revolutionize our view of biology and dis- ease pathologies (Barabasi and Oltvai, 2004). The popular clustering approach can draw densely connected modules from biological networks, which are often biologically mean- ingful, e.g. a dense protein interaction subnetwork may
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}