|
Literature
|
|
This page will feature expert opinions and guest columns from a panel of relevant scientists. It will also host the editor's choice of article of the month, which will be a pioneering publication in the field. Web links will be provided to all relevant publications, subject to the editor's discretion. Reader will find this site useful to be up-to-date on the current work in the field and familiarize with the emerging concepts of network biology.
|
Relevant Articles (Editor's Choice)
Inferring malaria parasite population structure from serological networks
Caroline O Buckee,Peter C Bull and Sunetra Gupta1
The malaria parasite Plasmodium falciparum is characterized by high levels of genetic diversity at antigenic loci involved in virulence and immune evasion. Knowledge of the population structure and dynamics of these genes is important for designing control programmes and understanding the acquisition of immunity to malaria .....
Journal:
Proc Biol Sci. 1-9,2008 September 30.
Comprehensive genomic characterization defines human glioblastoma genes and core pathways
The Cancer Genome Atlas Research Network
Human cancer cells typically harbor multiple chromosomal aberrations, nucleotide substitutions and epigenetic modifications that drive malignant transformation. The Cancer Genome Atlas (TCGA) pilot project aims to assess the value of large-scale multi-dimensional analysis of these molecular characteristics in human cancer and to provide the data rapidly to the research community.....
Journal:
Nature 455, 1061-1068,2008 October 23.
Editor's Note
> Background
The sequencing of the human genome has inspired the development of several high-throughput Gene expression analysis using microarrays can simultaneously analyze expression changes in thousands of genes. The genes that show synchronous alterations in mRNA expression can be connected into a biological network that represents a hypothetical hierarchy of molecular events in the cell. Constructing interaction networks between these co-regulated genes provides deeper insights into the molecular mechanisms of the complex biological processes. This mode of analysis is now a natural preference for downstream analysis in gene expression studies.
> Motivation
In order to assist researchers to analyze and explore their genomic data in the context of biological pathways, we have created a unique web based resource - PathwayWorld. This pathway portal hosts thousands of hypothetical molecular interaction maps. Each map represents the molecular relationships between the genes that simultaneously alter expression, in any given experiment. The underlying information that connects two genes is known and procured from published literature and physical interactions databases.
> Analysis
A first-cut analysis of available microarray data sets from Gene Expression Omnibus (GEO), has been performed. We have used an automated version of Strand's proprietary data mining platform, AVADIS™, on which Agilent's GeneSpring software is built. Data sets are downloaded from GEO and separated into groups of samples based on the meta-data associated with each data set. The automation uses common QC criteria for including a sample into the analysis. All replicates within an experimental group of a given data set need to have a correlation coefficient of 0.8 with all others in the same group. Significance analysis uses a unpaired t-test (as in GeneSpring GX) and limits the maximum number of significant genes to 100. However if the minimum p value of the significant genes in the list exceed 0.1, the cut-off is drawn to include only those that fall below this cut-off.
> Pathway Analysis
We have converted thousands of these gene lists into biologically relevant interaction maps. Each map is a collection of genes and their molecular interactions, depicted as nodes and edges, respectively. The depicted interactions between the genes are derived from Strand's proprietary molecular interactions database. This database contains information on molecular interactions between proteins that have been extracted primarily from published scientific literature using Natural Language Processing (NLP). A second source of molecular interactions data is the International Molecular Exchange consortium (IMEX) which forms a repository of high throughput physical interactions experiments.
346 data sets are analyzed including Agilent and Affymetrix arrays. More than 1200 pathways have been created using this automated pipe line from the analysis of human, rat and mouse microarray data. The pathways are tagged with more than 79 disease tags, 37 tissue/organ tags and 12 physiology tags.
Strand's PathwayWorld is an innovative resource for life scientists and clinical researchers. The key feature that distinguishes pathways contained in PathwayWorld is that each network represents a testable hypothesis- where every gene in a given pathway is a potential guide to exploring the mechanistic context of the expression results. Genes differentially expressed in several different experiments could converge under various biological conditions, thus providing novel insights into the mechanism of diseases.
In addition, PathwayWorld presents the genomic research community with an open-source forum for discussions, exchange of ideas and data-sharing.
Last Updated:;
15th December, 2008
|
