difference between GSEA and Gene Ontology
3
5
Entering edit mode
8.4 years ago

Hi All I need some help regarding RNA basics Can any one tell me the difference between GSEA and Gene Ontology?

gene ontology GSEA RNA-Seq • 26k views
ADD COMMENT
13
Entering edit mode
7.0 years ago

This is an old question, but since I stumbled upon it looking for exactly this, I thought I'd add an answer. First, I'm going to rewrite history. I think the original question really meant: what is the difference between using an enrichment searching approach for GO terms and GSEA. That question makes more sense, to me at least.

There are several enrichment search engines out there, for example DAVID. To use these, you provide a list of upregulated genes that you have already generated. For example, this can be generated by running a differential expression analysis, choosing an (arbitrary) threshold and reducing the results to a list of genes. They are not ranked, it's literally just a list. This is then analysed (using a hypergeometric distribution) for significant enrichment of GO terms.

For GSEA, you need to define groups as you would have done in the differential expression step, but you do not need to set a threshold. The ranked gene expression values are analysed for enrichment in whichever list of pathways you are interested in. The list could be GO terms, but is not limited to that. The approach is very different from a statistical viewpoint - best to read the original 2005 PNAS paper for details.

So, a key difference is that GSEA does not require a cutoff - you use all your genes and the fold change (or similar - this is a parameter in the application) between groups is used to rank them.

ADD COMMENT
3
Entering edit mode
8.4 years ago
russhh 5.7k

Fundamentally, GSEA is an analysis method and the Gene Ontology is a dataset.

There are two different types of entities present in GO: i) genes (or other macromolecules - transcripts, proteins etc); and ii) GO terms.

There are two different datatypes in GO: i) hierarchical (ontological) links between GO terms; and ii) annotational links between a GO term and a gene

They probably have a more formal name for the datatypes but I don't know it. As an example, DNA repair is a GO term, and so is 'Regulation of DNA repair'. Since the latter is a more specific term than the former, and is directly related to the former, there is a hierarchical link between the two. As a second example, the gene TP53 is linked to the GO term for DNA Repair (GO:0006281) since it is involved in this process.

Suppose you've performed an RNA-Seq experiment and want to know whether the DNA repair pathway is affected in a global sense, you could use Gene set enrichment analysis to test this. It would use GO annotations for the genes that are present in your experiment (although it can use other annotations).

HTH

ADD COMMENT
0
Entering edit mode

thanks You means to say GSEA actually finds the group of genes which involved in regulation(like DNA repair) or certain disease.

ADD REPLY
0
Entering edit mode

Yes.

There are several tools that do functional enrichment analysis, but I would strongly urge you to use GSEA.

ADD REPLY
0
Entering edit mode

GSEA is used for RNA seq denovo analysis

ADD REPLY
0
Entering edit mode

Could you explain what you mean in plain English, please

ADD REPLY
1
Entering edit mode
8.4 years ago
Benn 8.3k

GSEA is a method developed by the broad institute http://software.broadinstitute.org/gsea/index.jsp

Gene Ontology is (quote) "the framework for the model of biology", and is maintained by the GO consortium http://geneontology.org/

ADD COMMENT
0
Entering edit mode

Actually i want to use GSEA for Denovo trancriptome analysis.

ADD REPLY
0
Entering edit mode

can i use GSEA for all kind of studies or it is specific for studies like cancer study or some disease related data?

ADD REPLY
0
Entering edit mode

It would help to add more information in your question. Are you working with human RNA? Martian RNA? Why you want to do de novo? What is your research question? What is your design? Etc.

I cannot give a right answer to your question with the information you have provided. Sorry.

ADD REPLY
0
Entering edit mode

I'd strongly recommend you read the GSEA paper, do some gene expression analysis on a published dataset that is related to your project, and then analyse the results of that analysis using the Broad's tool.

ADD REPLY
0
Entering edit mode

Hi, i have genome sequences and i want to perform enrichment analysis but i do not know, how to do it.

ADD REPLY

Login before adding your answer.

Traffic: 1235 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