2
$\begingroup$

So I am very new to the area of biology so I am sorry if this is a stupid question.

I have RNA-Seq data carried out over a span of 100 days and I have gene expression data in the following format. Each expression value is the mean of 3 replicates.

           Day 1         Day 10     Day50     Day 100
Gene 1      12             42         35       12
Gene 2      50             53         23       100
.           .              .          .        .   
.           .              .          .        .   
.           .              .          .        .

so on, the above data is just something I came up with, but can you tell me what information I can extract from this type of data ? Such as differential gene expression. Thank you very much

Improve this question
$\endgroup$
1
  • $\begingroup$ Note that there is also a bioinformatics.stackexchange.com site. $\endgroup$
    – bli
    Dec 11 '17 at 10:49
1
$\begingroup$

It can be complicated because RNA-seq data is looking at gene expression, in the form of mRNA. But there are thousands of housekeeping genes that are expressed constantly, as long as an organism is alive. So the real interesting genes in a time course are those that are differentially expressed. So gene 1, in your example is not that interesting. Gene 2 however shoes a 4-fold increase between day 50 and day 100. Still, that is not that huge. You need computer programs to sort all this out. But what you want is to be able to associate a huge change in mRNA (gene expression) that can be distinguished from background.

Improve this answer
$\endgroup$
2
  • $\begingroup$ Hello Karl, thanks for your answer, I need to mention that each expression value I have is the mean rpkm value of 3 replicates. So the change in expression is not due biological noise. Yes, I was thinking of using R to perhaps classify the genes by the time points during which they have maximum expression. $\endgroup$
    – Adi
    Nov 13 '17 at 20:48
  • $\begingroup$ Even though there are replicates that doesn't mean you've eliminated noise -- just reduced one of the many different sources of noise. You should have data for housekeeping genes that you can use to normalize gene numbers against; see for example this question and answer $\endgroup$
    – iayork
    Sep 10 '18 at 12:09
0
$\begingroup$

I have never seen people extracting data from data, maybe you meant which type of information you might want to extract and the answer is "genes differentially expressed". You should check whether your data are normalized or not if they are you can use popular R package like DEseq2 or Limma to detect genes differentially expressed. In case your data have not normalized these tools have the proper way to normalize your matrix.

Here is the link of limma: https://www.bioconductor.org/packages/devel/bioc/vignettes/limma/inst/doc/usersguide.pdf

at page 46 you have an example of how to handle with time-course analysis

Improve this answer
$\endgroup$
3
  • $\begingroup$ Yes, that is exactly what I mean, I want to know what information I can extract. But since each data point is the mean of 3 replicates is it possible to find the differentially expressed genes? Thanks for the lead on DESeq and Limma !! $\endgroup$
    – Adi
    Nov 13 '17 at 21:07
  • $\begingroup$ Yes, of course, you can, it would have been better if you have all the replicates for each sample, but you have to do with what you have unless people in the lab can give you the raw data. Please check the new answer. I have pointed you to which page you have to go to perform the analysis. $\endgroup$
    – fusion.slope
    Nov 13 '17 at 21:09
  • $\begingroup$ if this reply answered your question, mark the question as answered. Best $\endgroup$
    – fusion.slope
    Nov 13 '17 at 21:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.