Hi everyone, We had this discussion off-list about Affy's E. coli chips and how they are different in many ways from eukaryotic Affy chips, particularly the effects of random priming and end-labeling of the samples (no polyA tails). Because there is so little about E. coli on the mailing list, we thought we should put this in the record... Cheers, Jenny Date: Thu, 11 May 2006 23:06:51 +0200 From: lemerle@embl.de To: Jenny Drnevich <drnevich at="" uiuc.edu=""> Subject: Re: coli annotation files sure. there's so little to read about coli chips that anything is helpful. cheers and thanks, caroline Quoting Jenny Drnevich <drnevich at="" uiuc.edu="">: >Hi Caroline, > >We pretty much decided that the bimodality in the density plots was due to >RNA degredation... Because of the end-labeling of prokaryotic samples, >partially degraded samples have higher amounts of label and hence results >in higher signals, which is opposite of what happens in eukaryotic >samples. Additionally, 3':5' ratios aren't as useful for QC because of the >end-labeling, and neither are the AffyRNAdeg plots - ours all had slopes >of 0, even for a sample that was known to be almost completely degraded! >If you don't mind, I'd like to post our exchange on Bioconductor's mailing >list as an FYI - maybe someone else will have an E. coli annotation package. > >Cheers, >Jenny > >At 11:39 AM 5/11/2006, you wrote: >>hi jenny, >>well, there's quite a lot you can do without using bioc, but those r packages >>really help save a lot of time, so i was a bit disappointed to see that >>it was >>so much more easy to do certain things with other organisms/other chips. >>i hadn't realized coli was so out of fashion in labs that afford many chips. >>about the bimodality of the density plots you posted one day, mine look >>nothing >>like yours. single mode. sorry, no help, only to say that it isn't >>necessarily >>so with coli. >>i wasn't aware of how the labeling methods impact qc results. actually qc in >>simpleaffy doesn't work with coli yet, but the maintainer of the package >>promised to update the info about my platform in a couple days. kind of feels >>like i'm doing things in reverse order, but well, i'm learning. >>in any case, thanks for the answer >> >>caroline >> >>Quoting Jenny Drnevich <drnevich at="" uiuc.edu="">: >> >>>Hi Caroline, >>> >>>Sorry for not replying to you sooner - your e-mail got buried in my >>>inbox. I only did the DEG analysis for a client with E. coli data, and >>>they were going to handle the annotation of the gene lists, but not >>>using Bioconductor. I got the feeling that not very many people were >>>using the Affy E. coli chips, because I couldn't find that much >>>information on them and I got very few responses to my posts on the list >>>server. The labeling methods are so different than for Eukaryotic >>>species that some of the Affy quality control functions don't really >>>work for E. coli, so I was surprised that more people weren't asking questions. >>> >>>Sorry I couldn't help you, >>>Jenny >>> >>>At 07:04 AM 4/28/2006, you wrote: >>>>hi jenny, >>>>i saw some posts of your on the bioconductor mailing list, that said >>>>you were >>>>working with affy coli chips. is it the genome2.0 type or the asv2 type >>>>or both >>>>that you were analysing? >>>>i ask you because i need annotation files for both (especially GO). >>>>would you >>>>happen to have build an R package for these? otherwise, do you know >>>>where to >>>>get GO annotations for coli genes? >>>>thanks for any help, >>>>caroline >>>> >>>>-- >>>>Caroline Lemerle >>>>PhD student, lab of Luis Serrano >>>>Structural Biology and Biocomputing Dept >>>>tel 00-49-6221-387-8335 >>>>-- >>> >>>Jenny Drnevich, Ph.D. >>> >>>Functional Genomics Bioinformatics Specialist >>>W.M. Keck Center for Comparative and Functional Genomics >>>Roy J. Carver Biotechnology Center >>>University of Illinois, Urbana-Champaign >>> >>>330 ERML >>>1201 W. Gregory Dr. >>>Urbana, IL 61801 >>>USA >>> >>>ph: 217-244-7355 >>>fax: 217-265-5066 >>>e-mail: drnevich at uiuc.edu >>