Mirna Ngs Goterm Analysis

Question

[Sara]

Hi everyone,

I want to do the gene ontology for target gene of the predicted miRNA in my study.

My species dont have compelet genome so i wonder what i should take as a background for doing the analysis?

1-nothing just take M function

2- take an microarray probes Go term from the same species as background

3- take a near species genome as background

many thanks

[Larry_Parnell]

Sara, your title suggests next-gen sequencing but point #2 mentions array probes - confusing. You give no details on what type of data you actually have. How have you assigned microRNAs to GO terms or how have you thought about doing this? "Background" is a confusing term here. Please provide a few more details and you're much more likely to see helpful answers.

Answer 1

[David Quigley]

I think she's asking the question, "I have an organism with a microarray but no finished genome. Given a set of genes that I identified as potential targets of an miRNA, what should I use as the background pool of genes for the hypergeometric test in the GO analysis to see whether my set of genes is enriched for a given term?". I agree with Lars that the question is hard to understand. I would use the set of genes on the microarray, but more important is the question of what ontology you're using. Your background is only relevant so far as the genes in that background have adequate annotation as well. Does your organism have an ontology, and if so what is the state of that ontology? Otherwise, you're probably using another organism as a surrogate. What's the closest well-annotated organism, and can you map gene names to their orthologs? Lots of approximations to make here, but this isn't an exact sort of analysis.

[Sara]

hello David thanks for you reply. sorry for my bad english! Actually, my data comes from miRNA reads data (from NGS).

I found which genes were targeted with these miRNAs . Now, I want to classified them by Biological process, Molecular function and subcellular localization of these target genes.

when, I use David for example it takes the closet species ( as background) for doing this classifications. I wonder if it is correct to do that or not?

[David Quigley]

If your organism doesn't have a well-defined ontology, I would use the closest well-annotated organism as DAVID is apparently doing.