Gene Identification: How Are Genes Identified?

[nesp]

The way I understand it, genes are rather sparse within a strand of DNA. The human genome, for example, has a few thousand genes in a strand of 3.5 million A,C,G,T bases. I also understand that genes code for proteins through RNA but, again, most of the DNA is noncoding. My question is, how do we identify genes from non-genes? I've searched through google and found some useful sites, such as this one http://www.genomenewsnetwork.org/articles/06_00/sequence_primer.shtml

but they are overly complex for my question. I'm impressed that the entire genome can be decoded, but is there a simple explanation for how a geneticist identifies that this particulare section of DNA is a gene and that one is not?

 

[iansmith]

Here a few basic concept, every gene will have 3 base pair (i.e. a codon) that signal the start of protein. There several different start signal and those depend on the organism. There's also stop signal at the end of the gene that signal the last codon

http://en.wikipedia.org/wiki/Start_codon
http://en.wikipedia.org/wiki/Stop_codon#Start.2Fstop_codonsOnce you sequence and assemble a certain area, you can run programs that will predict potential gene and their location. These potential genes are called open reading frame (ORF).

ORFinder is an example
http://www.ncbi.nlm.nih.gov/gorf/gorf.html

Once you id a potential ORF, you can compare, using a computer program, the predicted protein sequence to a database of other predicted and/or sequenced proteins.

BLAST is such a program
http://www.ncbi.nlm.nih.gov/BLAST/

Once you think your prediction is good, you can then look at the sequence before and after the gene to see if there any marker. Non-coding sequence before and after the gene that are required for the gene to be transcribed into RNA. There are also important when you are trying to id a gene.

The last step you be to do experiemental work to confirm that the gene produce the predicted protein and the predicted RNA.

 

[nesp]

Thanks for that explanation and those links, they are very helpful. My field is mathematics, not biology, and if I didn't know what you were referring to, I would think you were describing a computer program, especially the start and stop signals. In essence, DNA is a biological computer. It's amazing the kind of order and logical structures that can be produced through randomness and chaos.