Primer on DNA
Harry, RRS and whoever else is interested. i have tried in the past to put together a DNA primer, but it gets out of hand in a hurry, there are all sorts of ways to use DNA.
1. species Identification: for species ID we typically use the mtDNA. mtDNA has several advantages to the researcher, it is more plentiful in a sample, it is easy to isolate, it is a small piece of DNA, and it contains specific genes that have been proven to be unique to each species - We have a number of very large databases of the sequences from mtDNA (BOLD, GenBank for example, hundreds of millions of seqences, hundreds of thousands of species IDed)
Do we have sequence for every species on the planet - nope, we think there are somewhere in the vicinity of 10 million species here on earth, but for mammals, birds, large stuff, and or species of economic or medical interest, we have lots of sequences from lots of individuals of each species. CO1 is the gene we usually use, it is present in every species, but each species has a different form of that gene.
The workflow is pretty simple for species ID, get mtDNA, make millions of copies of it, get a sequence using those copies, compare the sequence of a specific gene CO1 usually, to those in the databases, using powerfull software.
Your results will either say "yup you found Upupa epops" (my favorite scientific name of all time), or it will say - "listen Fred, I can't find an exact match here in my database. I can tell you that it is very similar to species X and Y with XX% divergence from each one"
2. Individual Identification: identifying a specific individual takes a different process. The most common, certainly for humans, but also applicable in other cases, is the use of Short Tandom Repeat STR's. essentially you need to use nuclear DNA for this, which resides inside the nucleus of the cell, and there is only one copy per cell (mitochondrial is outside the nucleus, and you may have a couple of thousand copies per cell).
so scattered throughout you nDNA are regions of your genome that simply repeat a certain pattern of bases. It may repeat that pattern 2 times, or 3 times, or 4 or whatever. Whats interesting about that is everyone has those repeats at the same locations of their genomes, but the number of repeats at each site varies, so site one - you may have 3 repeats of CAA for example, while your neighbor may have 6 repeats of CAA at that site. In location 2 you may have 2 repeats of GCA, while your neighbor would have 3.
So you chose a number of specific sites, and simply quantify how many repeats at each site. From there it is a simple math game. The FBI has standardized 13 sites they use for analysis, and depending on how you do the math, the chances someone has the same number of repeats at the same locations as you is slim - 1 in a billion, or 1 in a trillion, depending on how you chose to do the math and explain it.
So to get individual ID, you of course need either a reference, known sample, or get 2 unknown samples to match up. while this method is great for individual ID, it has no value in species ID, (at least as far as i know)
3. There is also another methodology - SNP's, or single nucleotide polymorphism. essentially when you compare individuals of the same species, there will be a number of locations where the base can vary, and vary randomly, so you can use those SNP's to Id individuals as well (sometimes) , but by far the most common use of SNP is that it indicates certain genetic mutations that make one suceptable for certain diseases.
4. Finally you have whole genome sequencing. In this case you simply record each and every base pair in your nuclear DNA. That has become much easier cheaper, faster and more accurate with the Next Gen systems. This however comes with it's own set of problems. in a field where you used to fight long and hard to pull a few hundred or thousand accurate bases, you now get burried with 3 billion base pairs. it becomes a gigantic data management issue, and the software to smooth that workload is still being developed.
OK, so thats my attempt at summary of the art. If you would like me to apply that to what we know about Melba, let me know!
CWB
.
To generalize, can't DNA testing get all the way down to the individual level of identification no matter the species? And I understand DNA has to be matched against a known quantity to determine exact match, but what about a general match? Isn't the human genome (for example) pretty much exactly the same throughout humankind except for the few genes (relative to the whole) that define the racial makeup and uniqueness of any one individual? That is, say you have two sets of DNA data, one from a human, one from a Capybara (and you don't initially know that), does the fact one is human and one is Capybara become clear/obvious (sooner than later) in the testing, or is there more that has to be done to differentiate them?
?
