""the odds" of a genetic match are 6 trillion to one?" Topic

Statistics, a very boring, very dull, math, that you can use to determine odds, and make scads and scads of money on the stock market if you do it right, and causes others to loose scads and scads of money if you do it wrong.

It is also used by gamblers, playing let's say, poker, to let them figure how likely they are going to be beaten by certain hands they have, letting them know when to hold 'em and when to fold 'em.

Oh, and it can be used in science to determine genetic odds, by basically seeing that DNA can make X amount of combinations, and that since that combination we are looking at looks like this, statistical math tells us that the odds of it happening this way is… 6 trillion to one.

Yes that's Science. A bunch of boring math, comparisons, and testing. Which lets you make atomic bombs, transistors, silicon chips, cancer cures, and a whole lot of money, if you do the science right.

Bah! Science and math are a myth perpetrated by intelligent people on the rest of us.

Out of curiosity, is that 6 trillion to 1 based on any combination of the components that make up DNA, or 6 trillion to 1 odds of a combination that would make a living human being being the same as another living human being? Just because a combination is mathematically possible doesn't mean it is physically possible, much less a combination that produces a living creature of a specific species.

I'm not saying that the odds given are incorrect, I'm just pointing out that there has to be more behind it than just simple random distribution, just as the 26 letters of the alphabet can produce an infinite combination of characters, but not an infinite combination of characters that produce meaningful English (or any other language). At some point it's just gobbledygook.

But how many times do you go somewhere and see people you think you know only to realsie they are complete strangers? I find the genetic makeup of mankind to be scarily limited sometimes!

Very good Parzival, a very pertinent question.

Obviously there is a lot more to the problem then just figuring out a statistical likelihood. Note in my previous post the two words "this way". The scientists are assuming a specific set of circumstances and more importantly, a certain way they believe those circumstances came about. Under those restrictions they believe the odds are 6 Trillian to one. The way that the specific set of circumstances came to be is the important point, and is what is pointed at with the high odds. For instance 6 trillion to one sound pretty convincing odds, nobody is going to bet on those odds.

But let's say I bet you I can find two identical DNA between two people, at 6 trillion to 1 odds. And then I let you met my identical twin brothers. But wait you say, that's cheating, your brothers circumstances are totally different from everyone elses. Exactly, which is the whole point of the 6 trillion to 1 odds in the first place. Those creatures circumstances are way different from normal creatures.

The statistics given are for a specific individual's DNA and a matching sample from someone else. Example:

Person A: John Doe, is accused of a crime and there is DNA evidence left at the crime scene. The evidence is collected and submitted for genetic testing. A search warrant seizes a DNA sample from John Doe, which is then compared with the genetic DNA material left at the crime scene. The statistical probability states that the probability of someone other than John Doe leaving that DNA material at the crime scene is 1 in 6 trillion people. In other words, in order for you to have an identical person with the identical DNA markers – there is a 1 in 6 trillion possibility that the person exists. Put that together with all of the other evidence, and John Doe is pretty guilty of the offense based upon his DNA, and the fact that there is a 1 in 6 trillion possibility that it was someone else who left the DNA material at the crime scene. Pretty much proof beyond a reasonable doubt with some corroborating evidence of the crime. At least that is how it is used in criminal processes. Statistics in genetic/DNA is generally used to exclude other people from having done something.

Another example is this: John Doe was born and the mother, Jane Doe, claims that it was either John Doe Sr., James Doe, Frank Doe, or George Doe, who is the actual father of the child. A DNA sample is taken from John Doe Jr., and provided to a laboratory. John Doe Sr., James Doe, and Frank Doe, all submit samples for testing. George Doe is unable to be located so his sample is not collected. The laboratory conducts a DNA test to determine what the probability is that each person is the father of John Doe Jr. The statistical results indicate that John Doe Sr. is 96.534% likely to be the father, James Doe is 89.421% likely to be the father, and Frank Doe is 99.997% likely to be the father. A court would conclude that Frank Doe is the natural father of John Doe Jr. based upon the statistical probability that he is the father, as the statistical probability that someone OTHER than Frank Doe being the father would be 1 in 2.58 trillion people. Based upon the fact that Jane Doe listed him as a potential father, that is corroborated by the statistical probabilities of the DNA analysis.

In talking with the experts – they will never say that there is a 100% positive identification, because they do not test every single pair, but they test on specific markers throughout the DNA and figure out the statistical probabilities that it could be someone else's DNA.

Yes that's Science.

Does math really count as science? Certainly, it's a tool used by science, but math itself is abstract.

Yes and no: link

Does math really count as science? Certainly, it's a tool used by science, but math itself is abstract.

It's much more than a tool. A microscope and a beaker are tools. Mathematics is a means by which to quantify relationships, as in Jovian1's example above. It is also the only way to describe events beyond our means of experience, such as the diameter of an electron shell, or the diameter of the Universe.

You need to quantify the observations you make in science, one even has to quantify the possibility of error within those observations. Without the math, it's not science.

As I understand it, they are comparing gene *sequences*, not individual genes. I believe that this is what deals with the issue of "linked" gene possibilities. The odds of a fairly large sequence being identical to someone else's sequence becomes very small very quickly. Specific small segments may be the same, but variations across larger sequences are guaranteed. When you start using large numbers of sequences as well as large numbers of genes, that is where you get the vanishingly small odds of finding duplicate.

She blinded me with SCIENCE!