tag:blogger.com,1999:blog-6232819486261696035.post910598373471278501..comments2023-09-25T09:01:44.323-05:00Comments on Getting Genetics Done: Pathway Analysis for High-Throughput Genomics StudiesStephen Turnerhttp://www.blogger.com/profile/06656711316726116187noreply@blogger.comBlogger11125tag:blogger.com,1999:blog-6232819486261696035.post-52369760898549318822015-12-06T22:47:47.566-06:002015-12-06T22:47:47.566-06:00Hi Dr. Turner,
I would like to put in a plug for a...Hi Dr. Turner,<br />I would like to put in a plug for a paper (no connection to my work and no other conflict of interest; I read it for a class project). It's another review of pathway analysis methods, and it's got a nice set of references that I don't think I would have found anywhere else. When I saw on Google Scholar that it only had one citation, I got sad. <br /><br />It focuses on network-topology based methods. Hope you or your readers will enjoy!<br />Eric Kernfeld<br /><br />Full citation<br />Liu, Y., & Chance, M. R. (2013). Pathway analyses and understanding disease associations. Current Genetic Medicine Reports, 1(4), 10.1007/s40142–013–0025–3. http://doi.org/10.1007/s40142-013-0025-3<br /><br />URL<br />http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3851310/Anonymoushttps://www.blogger.com/profile/18282559994474885302noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-51287579977954252872014-08-05T12:56:31.405-05:002014-08-05T12:56:31.405-05:00Thanks Dr. Turner.Thanks Dr. Turner.shrutihttps://www.blogger.com/profile/04056868214123875905noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-15786663498200811512014-08-05T12:34:11.073-05:002014-08-05T12:34:11.073-05:00Hi Shruti. Thanks for the comment, and thanks for ...Hi Shruti. Thanks for the comment, and thanks for introducing me to ROntoTools - I didn't know about this package. I don't know how to solve the problem you're having. You might try SEQanswers.Stephen Turnerhttps://www.blogger.com/profile/06656711316726116187noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-12823108948755603452014-08-05T11:46:17.932-05:002014-08-05T11:46:17.932-05:00Hi Dr. Turner,
Thanks for this useful post (and f...Hi Dr. Turner,<br /><br />Thanks for this useful post (and for many others on Getting Genetics Done). I am trying to use Pathway Express from RontoTools (which as I understand is advanced version of spia) for microarray analysis. I get an error when I use the option "updateCache = TRUE" in keggPathwayGraphs command in order to download the latest KEGG pathways. I was wondering if you have encountered this kind of error before. Am I getting this error because I do not have license for KEGG files?<br /><br />library("KEGGREST")<br />library("ROntoTools")<br />kpg <- keggPathwayGraphs("hsa", updateCache = TRUE, verbose = F)<br /><br />NA element in edges.<br />Error in validObject(.Object) : invalid class “graphNEL” object: FALSE<br /><br />Thanks,<br />Shrutishrutihttps://www.blogger.com/profile/04056868214123875905noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-73391704943970944252014-07-28T06:39:36.301-05:002014-07-28T06:39:36.301-05:00Check out the relevel() function.Check out the relevel() function.Stephen Turnerhttps://www.blogger.com/profile/06656711316726116187noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-10160928854459003872014-07-25T21:15:18.956-05:002014-07-25T21:15:18.956-05:00Hi,
I'm analyzing a microarray data and stuck...Hi,<br /><br />I'm analyzing a microarray data and stuck at the design matrix command. <br /><br />I have few samples, in which I divided them into two groups i.e. "Controls" & "Diseased".<br /><br />I assigned the samples information as a factor with two levels "C" & "D".<br /><br />In what exact order should I assign the levels? If I assign levels as c(C, D) then I get the logFC value as say X and if I do it as c(D, C) I get the logFC value as -X.<br /><br />Should I start with the diseased group (D) or control group (C)?<br /><br />In default, in what exact order the design matrix take the levels?<br /><br /> <br /><br />My code is as follows: <br /><br />info$Group<-factor(info$Group, levels=c("C","D")) <br /><br /># ^^^ Which one should I consider first in the above command? The diseased group or control group? ^^^<br /><br />levels(info$Group)<br /><br />lev<-levels(info$Group)<br /><br />design<-model.matrix(~0+info$Group)<br />colnames(design)<-lev<br />fit<-lmFit(exp, design)<br />contr.str <- c()<br />len<-length(lev)<br />for(i in 1:(len-1))contr.str<-c(contr.str, paste(lev[(i+1):len], lev[i], sep="-"))<br />contr.str<br />contr.mat<-makeContrasts(contrasts=contr.str, levels=lev)<br />fit2<-contrasts.fit(fit, contr.mat)<br />fit2<-eBayes(fit2)<br />top <- topTable(fit2, number=nrow(fit2), adjust.method="fdr")Jeevanhttps://www.blogger.com/profile/06970191571210921746noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-63714650422018367162013-03-27T15:16:43.842-05:002013-03-27T15:16:43.842-05:00Often neglected is the key difference between &quo...Often neglected is the key difference between "competitive" and "self-contained" tests. It's all about how the null hypothesis the test will reject.<br /><br />In brief, <br /><br />competitive gene-set test look for a higher than expected number of differentially expressed genes, or otherwise "highly ranked" or "experimentally positive" genes in your set compared to the gene universe.<br /><br />self-contained tests look for differences at the gene-set level. If you have array data, you have to define a difference index using all genes in the gene-set. If you have rare variants, you will compare the "load" of rare variants in cases to controls (basically doing an "association test" at the gene-set level). <br /><br />When the test is based on random sampling, gene sampling is typically used by competitive tests, phenotype / class-label sampling is typically used by self-contained tests.<br /><br />In my experience, competitive are used a lot for gene expression microarray data, especially when the replicates are reproducible and classes well-separated. Self-contained are prevalent for (rare) genetic variants.<br /><br />Here's one review for more:<br />http://bib.oxfordjournals.org/content/9/3/189.fullDaniele Mericohttp://www.tcag.ca/noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-14044971911164024722012-03-08T09:35:29.775-06:002012-03-08T09:35:29.775-06:00This was a great summary. One of my struggles wor...This was a great summary. One of my struggles working in maize is that many of the ontology and gene pathway information has been worked out in Arabidopsis. What I am seeing is that one copy or some copies are involved in the same pathways, but evolution pushes the duplicates in other directions. And a priori you can't tell which copy is coopted internally for alternate function.BrianGhttps://www.blogger.com/profile/11883278789876067947noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-24097288559504441022012-03-07T09:43:58.062-06:002012-03-07T09:43:58.062-06:00Thank you - very comprehensive!Thank you - very comprehensive!ZChttps://www.blogger.com/profile/13200285398180829713noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-89478868517688266452012-03-07T00:15:26.222-06:002012-03-07T00:15:26.222-06:00Hey,
I just hopped over to your site via Stumbleu...Hey, <br />I just hopped over to your site via Stumbleupon. Not somthing I would normally read, but I liked your thoughts none the less. Thanks for making something worth reading.Crane Failure Analysishttp://www.acplindia.net/noreply@blogger.comtag:blogger.com,1999:blog-6232819486261696035.post-80069903950737810582012-03-06T18:41:02.889-06:002012-03-06T18:41:02.889-06:00Thanks for noticing our paper and giving it such a...Thanks for noticing our paper and giving it such a great plug! -- Atul ButteAtul Buttehttp://med.stanford.edu/profiles/Atul_Buttenoreply@blogger.com