Protocol Online logo
Top : Forum Archives: : Bioinformatics and Biostatistics

How to identify novel gene using bioinfomatics - (May/06/2008 )


I'm working with unknown gene, to study the function. one of the method is by using bioinfomatic. I already have data on the structure (primer, secunder n tertier) and phylogenetic tree. my problem now is, from the phylo tree this gene fall into a group of unknown genes also. I've come to the dead end. Anybody have any suggestion for me how to proceed?


check that your gene doesn't have many copies (southern blot might help), you should use another approach and try to disrupt this gene. hopefully you'll find a phenotype.


thanx for your info. i actually did few wet lab experiment such as RNAi but still i could see any difference on the phenotype. now i'm trying VIGS but still at the early stages. I gues this gene was the multigene famili. what do you think. if my guess is correct what should i do. do u have any suggestion? thank you in advance. smile.gif


Here is a short list: If you want to focus solely on computation, you can do the things in bold.

1. Over-expression in some cell lines (protein microinjection, protein transduction, transient or stable transfection of expression vector)
2. Knock-down (siRNA, inhibitor compound)
3 Above followed by expression microarray, proteomics (2D, protein array)
4. Mutation analysis (truncation, AA modification)
5. Transgenic plant (as you mention VIGS) for your gene, I guess.
6. Interaction analysis may give clue to protein function, if the binding partner is famous (Y2H, Co-Ip, MS/MS and two dozen others)
7. Any functional assays available in your system.
8. bioinformatics for
-Functional domains,
-Motif search,
-Post-translational modification prediction,
-Pathway analysis,
-Promoter analysis for any clues as to what group of TFs bind there, this may lead to a hypothesis for function,
-Genomics, Polymorphisms,
-Gene structure analysis
-literature mining,

Edit: Check out the first presentation here, looks promising: