Quote:
Originally Posted by
Matthew3590 
I always get confused when they say we share 98% of DNA. Does that just mean that we have 98% of the same possible genes?
Most of our DNA is not genes. The old-school definition of a gene is a segment of DNA that gets transcribed and translated to produce a protein. In terms of our total DNA, that's in the single-digits percentage. The rest is what used to be called "junk DNA" -- because, until recently, if DNA didn't make a protein, it was considered "junk" or merely "structural." We now know that there is a lot more to our DNA than just "genes" and "not-genes." Some stretches of DNA may not code for a protein themselves but affect the rate that other genes are expressed. And within a gene, only some parts may actually be translated into a protein -- there are many genes that stretch on and on and on, but only a minor portion of the gene actually makes it to the ribosome to direct amino acid sequence to build a protein. It's a more complicated system than originally thought.
Mutations are errors that occur during DNA replication at mitosis and meiosis. When a mutation occurs in a section of DNA that does not code for a protein or affect another section that codes for a protein, there is no negative result in the organism -- a "harmless" or "neutral" mutation. As such, these will vary more between lineages, since there isn't a selective pressure to maintain a non-mutated version.
Genes, however, tend to vary less down a lineage, so differences in genes between related species are more significant than differences in DNA overall between related species. Those differences found in the genes will have effects that are related to the differences seen in the related species when we look at them.
Differences found in the rest of the DNA is what's used in "genetic clocks" to measure divergence between species since the differences are mutations which are calculated to occur at regular intervals. When you read research that states how long ago two different species shared a common ancestor based upon DNA studies, it is often based on measuring differences at particular regions on chromosomes that have a steady mutation rate, and then doing some math to determine how long it's been since the differences diverged.
See below for a more in-depth explanation:
http://www.talkorigins.org/faqs/molgen/
And, if you read someone's attempt at dismissing the work above, the author posted a response which I (in my geekiness) find somewhat amusing.
http://www.talkorigins.org/faqs/molgen/plaisted.html
ETA -- oh, and another response
http://www.talkorigins.org/faqs/molgen/wieland.html
Edited by AquaEyes - 3/8/12 at 7:24pm