DOC PREVIEW
Eigengene Network Analysis

This preview shows page 1-2-3-4-5-6 out of 17 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 17 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Eigengene Network Analysis: Four Tissues Of Female MiceR Tutorial Peter Langfelder and Steve HorvathCorrespondence: [email protected], [email protected] is a self-contained R software tutorial that illustrates how to carry out an eigengenenetwork analysis across four corresponding to gene expression measurements in thebrain, muscle, liver and adipose tissues of female mice of an F2 mouse cross. The R codeshows how to perform the analysis reported in Langelder and Horvath (2007). Somefamiliarity with the R software is desirable but the document is fairly self-contained. The methods and biological implications are described in the following refence- Langfelder P, Horvath S (2007) Eigengene networks for studying therelationships between co-expression modules. BMC Systems Biology This tutorial and the data files can be found at http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork/EigengeneNetwork.More material on weighted network analysis can be found at http://www.genetics.ucla.edu/labs/horvath/CoexpressionNetwork.For a detailed description of the mouse intercross, see Wang et al. (2006). A descriptionof the microarray data can be found in Ghazalpour et al (2006). To facilitate comparisonwith the original analysis of Ghazalpour et al (2006), we used the same gene selection inour analysis.Microarray Data The Agilent microarrays measured gene expression profiles in female mice of an F2mouse intercross described in Ghazalpour et al 2006. The F2 mouse intercross data set(referred to as B × H cross) involved 135 female mice derived from the F2 intercrossbetween inbred strains C3H/HeJ and C57BL/6J (Ghazalpour et al. 2006; Wang et al.2006). B×H mice are ApoE null (ApoE −/−) and thus hyperlipidemic and were fed ahigh-fat diet. The B×H mice were sacrificed at 24 weeks.Microarray analysisRNA preparation and array hybridizations were performed at Rosetta Inpharmatics(Seattle, Washington, United States). The custom ink-jet microarrays used in this study(Agilent Technologies [Palo Alto, California, United States], previously described)contain 2,186 control probes and 23,574 non-control oligonucleotides extracted frommouse Unigene clusters and combined with RefSeq sequences and RIKEN full-lengthclones. Mouse livers were homogenized and total RNA extracted using Trizol reagent(Invitrogen, Carlsbad, California, United States) according to manufacturer's protocol.Three μg of total RNA was reverse transcribed and labeled with either Cy3 or Cy5fluorochromes. Purified Cy3 or Cy5 complementary RNA was hybridized to at least twomicroarray slides with fluor reversal for 24 h in a hybridization chamber, washed, andscanned using a laser confocal scanner. Arrays were quantified on the basis of spotintensity relative to background, adjusted for experimental variation between arrays usingaverage intensity over multiple channels, and fit to an error model to determinesignificance (type I error). Gene expression is reported as the ratio of the mean log10intensity (mlratio) relative to the pool derived from 150 mice randomly selected from theF2 population.Microarray data reductionIn order to minimize noise in the gene expression dataset, several data-filtering stepswere taken. First, preliminary evidence showed major differences in gene expressionlevels between sexes among the F2 mice used, and therefore only female mice were usedfor network construction. The construction and comparison of the male network will bereported elsewhere. Only those mice with complete phenotype, genotype, and array datawere used. This gave a final experimental sample of 135 female mice used for networkconstruction. To reduce the computational burden and to possibly enhance the signal inour data, we used only the 8,000 most-varying female liver genes in our preliminarynetwork construction. For module detection, we limited our analysis to the 3,600 most-connected genes because our module construction method and visualization tools cannothandle larger datasets at this point. By definition, module genes are highly connectedwith the genes of their module (i.e., module genes tend to have relatively highconnectivity). Thus, for the purpose of module detection, restricting the analysis to themost-connected genes should not lead to major information loss. Since the network nodesin our analysis correspond to genes as opposed to probesets, we eliminated multipleprobes with similar expression patterns for the same gene. Specifically, the 3,600 geneswere examined, and where appropriate, gene isoforms and genes containing duplicateprobes were excluded by using only those with the highest expression among theredundant transcripts. This final filtering step yielded a count of 3,421 genes for theexperimental network construction.Weighted gene co-expression network construction.Constructing a weighted co-expression network is critical for identifying modules and fordefining the intramodular connectivity. In co-expression networks, nodes correspond togenes, and connection strengths are determined by the pairwise correlations betweenexpression profiles. In contrast to unweighted networks, weighted networks use softthresholding of the Pearson correlation matrix for determining the connection strengthsbetween two genes. Soft thresholding of the Pearson correlation preserves the continuousnature of the gene co-expression information, and leads to results that are highly robustwith respect to the weighted network construction method (Zhang and Horvath 2005).The theory of the network construction algorithm is described in detail elsewhere (Zhangand Horvath, 2005). Briefly, a gene co-expression similarity measure (absolute value ofthe Pearson product moment correlation) is used to relate every pairwise gene–generelationship. An adjacency matrix is then constructed using a “soft” power adjacencyfunction aij = |cor(xi, xj)|β where the absolute value of the Pearson correlation measuresgene is the co-expression similarity, and aij represents the resulting adjacency thatmeasures the connection strengths. The network connectivity (kall) of the i-th gene is thesum of the connection strengths with the other genes. This summation performed over allgenes in a particular


Eigengene Network Analysis

Download Eigengene Network Analysis
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Eigengene Network Analysis and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Eigengene Network Analysis 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?