Hi everyone, I am currently working with gene expression analysis. I have downloaded a dataset GSE17537 from GEO database. In this dataset, it shows 19 and 36 patients with cancer recurrence and without recurrence respectively. ( no treatment at all)
how can I find the differentially expressed genes? What basic tool can I use? I find many use Limma but it is too difficult for me!! And how can I calculate the fold change of each gene from this dataset since I don’t know the reference gene ?
thanks for all HELPING hands!!
Hi,
probably the the easiest way is to use GEO2R as Bhaskar already mentioned.
At the bottom of the description page for the dataset you'll find a button labeled "Analyze with GEO2R". Clicking the button will lead you to a new page where you find the samples from the study. Next you have to click on the link labeled "Define groups". Just enter 2 group names, e.g. "recurrence" and "no recurrence". After this, for each sample you want to include in the analysis, you have to select the sample and click on the group this sample belongs to. When you have assigned all the samples to their group you can click on either the button "Top 250" or "save all results" in the GEO2R tab at the bottom of the page. At the Options tab you can change same parameters of the analysis but I think the default settings are appropriate in most cases.
In microarrays genes are represented by short oligonucleotide probes. For some genes more than one probe exist on a particular chip. That's why you will find more than one ID for some genes.
I don't know if there is some "gold standard" for coping with these muliple probes, but I would choose one value as representative for all the probes of a certain gene.
You may check the sequences of the probes and BLAST them to check which are the most specific, since, as far as I know, one of the biggest issues with microarrays is cross hybridization due to the short length of the probes.
If you want only differentially expressed genes from a straightforward analysis, GEO2R can be enough. However, GEO2R may be not flexible for other purposes that R can do, like plotting and data filtering.
For a more complete affymetrix-specific analysis, you'll have to install Affymetrix's Transcriptome Analysis Console (TAC), although data preparation may be difficult for some users. I don't know if they changed something in the recent updates.
Now, if you really want limma in a user-friendly interface, I particularly recommend Gene Expression Analysis Platform (GEAP). It has a intuitive interface where you just put the GSE code and the program gets the entire data for you, then you distribute samples to comparison groups to obtain the differentially expressed genes in a very straightforward way. This program does not need installation (it's portable and can fit in a USB flash drive), and uses R and limma on background to obtain the differentially expressed genes. I did the analysis in less than ten minutes (attached images) using GEAP. It's Windows-only, tough.
Links for each program:
- GEO2R analysis for GSE17537:
https://www.ncbi.nlm.nih.gov/geo/geo2r/?acc=GSE17537
- Affymetrix Transcriptome Analysis Console:
https://www.thermofisher.com/br/en/home/life-science/microarray-analysis/microarray-analysis-instruments-software-services/microarray-analysis-software/affymetrix-transcriptome-analysis-console-software.html
- Gene Expression Analysis Platform (GEAP):
http://inf.ufrgs.br/geap/
i tried GEO2R, but i shows " Your request could not be completed. Below is the R output which might give you a hint about possible errors. You may also consult documentation on GEO2R Limitations and caveats. " angbody know why? thank you
- Badges
- Science topic