Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Gene identification

  1. Dec 19, 2006 #1
    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?
  2. jcsd
  3. Dec 19, 2006 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    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


    Once 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

    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

    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.
  4. Dec 19, 2006 #3
    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 refering 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.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook