miRNA analysis advice
1
0
Entering edit mode
Fiona ▴ 70
@fiona-5790
Last seen 8.1 years ago
United Kingdom
Dear all, I am working with microRNA data and as it is very new to me, I have some general questions. I'm hoping that some people on this list who know more about microarray analysis and miRNA than myself might be willing to offer some opinions/advice. First, when doing quality checks on the data, how useful is it to look at the RNA degradation plots with miRNA, given how short miRNAs are? Second, when analysing differential expression (using lmFit and eBayes in the 'limma' package), I find quite high numbers of probes which are significantly differentially expressed across my samples (in some contrasts as many as ~500 out of 4000 probes are significantly differentially expressed between samples). Is this unusual with miRNA data? Further, although a lot of these probes are from Drosophila species (which I expected since my samples are from D. melanogaster), some are from other insects, and even other phyla (there are some C. elegans and human miRNA probes which are found to be significantly differentially expressed, for example). >From what I have read in the literature so far, there does seem to be some conservation of miRNAs between diverse species, but I can't find any information about to what extent I might expect to find this in my data. I'm a little concerned that this level of homology between species might be unusual, which would suggest that I've made a mistake in the analysis. As I said, I am very new to this, so if anyone is willing to offer advice or just point me in the direction of some useful information I could look at myself, then I would really be very grateful. Many thanks in advance, Fiona Dr Fiona C Ingleby Postdoctoral Research Fellow University of Sussex, UK [[alternative HTML version deleted]]
miRNA Microarray microRNA miRNA Microarray microRNA • 1.1k views
ADD COMMENT
0
Entering edit mode
@james-w-macdonald-5106
Last seen 13 hours ago
United States
Hi Fiona, On 2/25/2013 8:14 AM, Fiona Ingleby wrote: > Dear all, > > I am working with microRNA data and as it is very new to me, I have some general questions. I'm hoping that some people on this list who know more about microarray analysis and miRNA than myself might be willing to offer some opinions/advice. > > First, when doing quality checks on the data, how useful is it to look at the RNA degradation plots with miRNA, given how short miRNAs are? Assuming here that you have an Affy miRNA array of some sort, not useful at all. Note that the Affy probes are all 25-mers, and miRNA transcripts are 21-23 nt long. In the vast majority of cases (with mature miRNA transcripts; this doesn't apply to the scaRNA, snoRNA nor hp-miRNAs), the duplicate probes are all identical. So the underlying premise behind the RNA degradation plot doesn't hold. > > Second, when analysing differential expression (using lmFit and eBayes in the 'limma' package), I find quite high numbers of probes which are significantly differentially expressed across my samples (in some contrasts as many as ~500 out of 4000 probes are significantly differentially expressed between samples). Is this unusual with miRNA data? Further, although a lot of these probes are from Drosophila species (which I expected since my samples are from D. melanogaster), some are from other insects, and even other phyla (there are some C. elegans and human miRNA probes which are found to be significantly differentially expressed, for example). > > From what I have read in the literature so far, there does seem to be some conservation of miRNAs between diverse species, but I can't find any information about to what extent I might expect to find this in my data. I'm a little concerned that this level of homology between species might be unusual, which would suggest that I've made a mistake in the analysis. > > As I said, I am very new to this, so if anyone is willing to offer advice or just point me in the direction of some useful information I could look at myself, then I would really be very grateful. In general I like to see somewhere around 15-30 differentially expressed miRNA transcripts, as anything more can become quite intractable for the people I collaborate with. This is because each miRNA may target > 1000 mRNA species, so too many miRNAs and all of a sudden every gene might be a potential target. That said, I just did an analysis looking at different brain regions in a particular bird species and there were hundreds to thousands of differentially expressed miRNA transcripts, depending on the contrast. So the fact that you have lots of differentially expressed transcripts doesn't necessarily mean you made a mistake. In addition having probes from other species pop up might not be a bad thing. In my experience, if you get the same miRNA from multiple species differentially expressed, it is because the miRNA is highly conserved, and there are little or no differences in the sequence. One thing we commonly do is to start with just those miRNA transcripts that from the species under consideration. This can help limit things to a tractable number of miRNAs. Best, Jim > > Many thanks in advance, > > Fiona > > Dr Fiona C Ingleby > Postdoctoral Research Fellow > University of Sussex, UK > > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099
ADD COMMENT
0
Entering edit mode
Hi, On Mon, Feb 25, 2013 at 10:25 AM, James W. MacDonald <jmacdon at="" uw.edu=""> wrote: [snip] > In general I like to see somewhere around 15-30 differentially expressed > miRNA transcripts, as anything more can become quite intractable for the > people I collaborate with. This is because each miRNA may target > 1000 mRNA > species, so too many miRNAs and all of a sudden every gene might be a > potential target. ... and don't forget to model competing endogenous RNA (ceRNA) effects! -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
ADD REPLY
0
Entering edit mode
Hi James and Steve, Thanks very much for your advice, you've made things a lot clearer. There's just one more thing (for now!). I can think of two ways of going about a drosophila-specific analysis: I could either filter the data by species and test for differential expression only with drosophila probes, or I could just use the results I have already (for which I used a non-specific filter), but focus only on the drosophila subset of significantly differentially expressed miRNAs (to start with). I think James was suggesting the latter of these two, if I understood correctly, and this makes most sense to me (since the first method would mean completely disregarding a lot of potentially interesting probes), but I just wanted to check that this would be the best way to go about it. Many thanks again, Fiona Dr Fiona C Ingleby Postdoctoral Research Fellow University of Sussex On 25 Feb 2013, at 16:34, Steve Lianoglou <mailinglist.honeypot@gmail.com> wrote: > Hi, > > On Mon, Feb 25, 2013 at 10:25 AM, James W. MacDonald <jmacdon@uw.edu> wrote: > [snip] >> In general I like to see somewhere around 15-30 differentially expressed >> miRNA transcripts, as anything more can become quite intractable for the >> people I collaborate with. This is because each miRNA may target > 1000 mRNA >> species, so too many miRNAs and all of a sudden every gene might be a >> potential target. > > ... and don't forget to model competing endogenous RNA (ceRNA) effects! > > -steve > > -- > Steve Lianoglou > Graduate Student: Computational Systems Biology > | Memorial Sloan-Kettering Cancer Center > | Weill Medical College of Cornell University > Contact Info: http://cbio.mskcc.org/~lianos/contact [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Fiona, On Tue, Feb 26, 2013 at 4:24 AM, Fiona Ingleby <fiona.ingleby at="" gmail.com=""> wrote: > Hi James and Steve, > > Thanks very much for your advice, you've made things a lot clearer. > > There's just one more thing (for now!). I can think of two ways of going > about a drosophila-specific analysis: I could either filter the data by > species and test for differential expression only with drosophila probes, or > I could just use the results I have already (for which I used a non- specific > filter), but focus only on the drosophila subset of significantly > differentially expressed miRNAs (to start with). I think James was > suggesting the latter of these two, if I understood correctly, and this > makes most sense to me (since the first method would mean completely > disregarding a lot of potentially interesting probes), but I just wanted to > check that this would be the best way to go about it. One thing to note is that by removing the probes that do not belong to the species you are interested in prior to running the differential expression analysis, you will have more power to detect differential expression in just the drosophila set of probes since you would be reducing the number of multiple tests you need to correct for. I think you said you had too many "hits" already, though, but perhaps by doing so you can then add a second filter on the log-fold-change of the differential expression, which maybe help to extract the stronger changes in miRNA expression among the ones that are already called significant by their adjusted p value. HTH, -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology | Memorial Sloan-Kettering Cancer Center | Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact
ADD REPLY

Login before adding your answer.

Traffic: 718 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6