Dear All, I am trying to use next-generation sequencing platforms to profile gene expression patterns during embryogenesis in our fish model (about 36 different stages). The genome of our model is about 2000 Mb. Currently only a draft of the genome with 70% coverage is available. I am considering using the following strategies to carry out our project: 1. Use SMRT of Pacific Biosciences, TMS of Helicos, SOLiD3 of ABI, Genome Analyzer of illumina or other RNA-Seq methods to directly sequence RNA isolated from embryos at each stage. 2. Combine RNAs from embryos at different developmental stages to form an RNA pool. Use DSN treatment to get the normalized cDNA, use illumina to sequence this normalized cDNA and get a high coverage cDNA database. Then use SOLiD 3 SAGE to profile gene expression patterns, which the tags will be mapped onto the cDNA database firstly. I need advice about which strategy is better? And how shall I set up biological and technique replicates? Thanks a lot for any thoughts or advice. Sincerely Yours Ziping _________________________________________________________________ The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with Hotmail. PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_5 [[alternative HTML version deleted]]

On Fri, Jul 23, 2010 at 1:18 PM, ziping zhang <zhangziping@hotmail.com>wrote: > > Dear All, > > I am > trying to use next-generation sequencing platforms to > profile gene expression patterns during embryogenesis in our fish model > (about 36 different stages). The genome of our model is about 2000 Mb. > Currently only a draft of the genome with 70% coverage is available. I > am considering using the following strategies to carry out our project: > 1. Use > SMRT of Pacific Biosciences, TMS of Helicos, SOLiD3 of ABI, Genome > Analyzer of illumina or other RNA-Seq methods to directly sequence RNA > isolated from embryos at each stage. > 2. Combine > RNAs from embryos at different developmental stages to form an RNA > pool. Use DSN treatment to get the normalized cDNA, use illumina to > sequence this normalized cDNA and get a high coverage cDNA database. > Then use SOLiD 3 SAGE to profile gene expression patterns, which the > tags will be mapped onto the cDNA database firstly. > > I need advice about which strategy is better? And how shall I set up > biological and technique replicates? > > Thanks a lot for any thoughts or advice. > > Sincerely Yours > > Ziping > > Hi, Ziping. This is definitely a question that you should be discussing with a collaborator if the answers are not clear to you. What you are proposing is likely to be an expensive and complicated set of experiments that require quite a bit of forethought. I would suggest that you collaborate with someone using one or more of these technologies to perform RNA-seq analysis. Also, don't forget that microarrays are actually very good at this type of thing, so I would not off-hand exclude array technology for such an experiment where the number of samples could easily be over 100. Sean [[alternative HTML version deleted]]

Ziping, This sounds an interesting project, but could be very challenge in some aspects. First, you have an uncompleted genome; Second, your data will be generated from more than one platforms; Third, you are tackling a problem with 36 conditions or time points. In regarding to the third issue, I believe that many of excellent statisticians in this group could help you out. I just want to add some comments for discussion. 1. Reference sequence library Without a reference library, it will be impossible to map the short reads generated from any platform you mentioned, PacBio may be an exception if the reads are long enough and errors are very low. The strategy to construct a cDNA sequencing library may also be problematic because of so many uncertainties, for example the low expressors may not be enriched to the lever that can give you enough coverage. Will it be possible to finish the genome sequencing for the species you are studying? Will the cost be really effective comparing to the genome sequencing? The model you are using doesn't seem to have a big genome (2000mb). If you think that constructing a transcripts library is something you should do, you should consider to use a platform which can generate the reads as long as possible. 2. platform variations When sequencing, it may be a good idea to use same platform for all your samples of a project. The platform variations may affect the downstream data analysis. For mapping or sequencing discovery, it may not be a big deal. But for counting and transcriptome profiling, platform variation will make downstream data analysis a little bit hard. In short, I would like to construct reference library first; second, stick with one platform for a specific project. In terms of the sample replicates issue, I don't think you will need technical replicates unless you have other purposes. People usually use three biological replicates for their microarray studies. I would say that three biological replicates might be minimum number for the study. Statistician may have something to say for this issue. This is just my thought and hope it helps. Thanks, Steve On Fri, Jul 23, 2010 at 3:18 PM, ziping zhang <zhangziping@hotmail.com>wrote: > > Dear All, > > I am > trying to use next-generation sequencing platforms to > profile gene expression patterns during embryogenesis in our fish model > (about 36 different stages). The genome of our model is about 2000 Mb. > Currently only a draft of the genome with 70% coverage is available. I > am considering using the following strategies to carry out our project: > 1. Use > SMRT of Pacific Biosciences, TMS of Helicos, SOLiD3 of ABI, Genome > Analyzer of illumina or other RNA-Seq methods to directly sequence RNA > isolated from embryos at each stage. > 2. Combine > RNAs from embryos at different developmental stages to form an RNA > pool. Use DSN treatment to get the normalized cDNA, use illumina to > sequence this normalized cDNA and get a high coverage cDNA database. > Then use SOLiD 3 SAGE to profile gene expression patterns, which the > tags will be mapped onto the cDNA database firstly. > > I need advice about which strategy is better? And how shall I set up > biological and technique replicates? > > Thanks a lot for any thoughts or advice. > > Sincerely Yours > > Ziping > > _________________________________________________________________ > The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with > Hotmail. > > PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_5 > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > [[alternative HTML version deleted]]