sda2.jpg

July 21, 2008

One Chance In A Quadrillionzillion!

What would we do without experts?

State crime lab analyst Kathryn Troyer was running tests on Arizona's DNA database when she stumbled across two felons with remarkably similar genetic profiles.

The men matched at nine of the 13 locations on chromosomes, or loci, commonly used to distinguish people.

The FBI estimated the odds of unrelated people sharing those genetic markers to be as remote as 1 in 113 billion. But the mug shots of the two felons suggested that they were not related: One was black, the other white.

In the years after her 2001 discovery, Troyer found dozens of similar matches -- each seeming to defy impossible odds.

As word spread, these findings by a little-known lab worker raised questions about the accuracy of the FBI's DNA statistics and ignited a legal fight over whether the nation's genetic databases ought to be opened to wider scrutiny.

[...]

Among about 65,000 felons [in the Arizona database], there were 122 pairs that matched at nine of 13 loci. Twenty pairs matched at 10 loci. One matched at 11 and one at 12, though both later proved to belong to relatives.


The American Kennel Club has run into similar issues with their own DNA typing of purebred dogs. Breeders may use now use two different sires on a single litter if they choose, providing they parentage test the offspring. However, problems can arise - sometimes two sires share so many of the same markers that it isn't possible to know which dog fathered what puppy. This should have been forseeable, given that so many breeds are based on a mere handful of founders.

In much the same way, present day humans are thought to be the survivors of genetic bottlenecking events that reduced world populations to mere thousands in the not so distant past. Despite our large modern population, we have relatively low genetic diversity.

Indeed, there is substantially more genetic difference among individuals within chimpanzee troops in West Africa than among all living humans on earth. As shown in Figure 1, this is due to a series of bottlenecks in human evolutionary history. Geneticists studying many different parts of the human genome have concluded that the past effective population size (that is, the number of reproducing females) averaged only 10,000 individuals over the last one million years, and was as low as 5,000 around 70,000 years ago. Compare this to the approximately one billion reproducing females alive today, and it becomes clear just how narrow these bottlenecks were.

Why this was not factored in when probabilities were assigned to DNA profiling is anyone's guess, but the actions of the FBI in attempting to block continued investigations of databases suggests their oft-cited figures arose not in the genetics lab, but in the bureaucracy.

DNA typing will always be a powerful tool in law enforcement. But should it have been touted as infallable in the way that it has? Absolutely not - too many genetic phenomenon break the "rules" for that. See tetragametic chimerism. The bizarre claims of the FBI and other law enforcement agencies about the mathematical rarity of DNA matches, (on samples using as few as 9 loci!) were as unprovable and as ridiculous as asserting that "no two snowflakes are alike". Their attempts to elevate the confidence of the science beyond reasonable limits will likely end up producing the opposite result, giving defense lawyers an opening to undermine the integrity of legitimate DNA evidence in the minds of jurors.

ht

Posted by Kate at July 21, 2008 9:58 AM
Comments

"providing they parentage test the offspring"

Abominable English.

Posted by: Richard Ball at July 21, 2008 10:11 AM

Heckuva post. More on chimeras:

The Chimera Hypothesis: Homosexuality and Plural Pregnancy

Eventually, it was a search into the phenomenon of twinning, which is the main cause of chimerism (excluding fetal cell chimerism in mothers), that led me to the work of Dr. Charles Boklage, a geneticist who is at the forefront of research into chimerism and twinning. Boklage is of the opinion that chimerism is at the root of a many poorly-understood conditions and traits, including homosexuality. His research, although considered controversial, was essential to a better understanding of the complex subject of embryology.
...
However, I see another explanation. Homosexuality is not the result of solely genetic or environmental factors, but rather a combination of the two. This combination results from the complex conditions in the womb during early development, when cells are exchanged in a medium in which the individual is not yet fully defined.

As the science of genetics has advanced in recent years, the existence of human chimeras has surprised a number of researchers. There are varying degrees of chimeric expression, and the effects span a fairly wide spectrum. In most cases, they are the result of twinning or blood exchange between mother and fetus. The most extreme form of chimerism, called tetragametic chimerism, is the merging of two genetically distinct embryos to form one organism.

Posted by: Merlot at July 21, 2008 10:15 AM

Weren't we just talking about this on the other thread? Black death resistance, malarial resistance, even white skin for increased vitamin D production, these are all due mostly to lots of people with the wrong genetic structure DYING, rather than the more benign business of being out-produced by those with a better genetic structure.

Mother Nature isn't a nice person. That's why I like technology, gives me a chance to return fire on the old b1tch.

Posted by: The Phantom at July 21, 2008 10:16 AM

On the subject of questionable law enforcement techniques and practices, the FBI is not at all alone in using scary court tactics.

Recent cases by the ATF (formerly Bureau of Alcohol, Tobacco and Firearms, responsible for both Ruby Ridge AND Waco Texas) in the USA involving semi-automatic firearms reveal that the lab weenies at the ATF have been very... creative... in their testing methods. Basically they seize some poor bastard's gun, then they fiddle with it until it will fire more than one round with one trigger pull.

This can include using the wrong ammunition in the gun, exchanging parts from other guns, applying lubricant until the thing is practically swimming in it, and actually machining parts to make them malfunction. Once they manage to make the mechanism fail and fire off more than one bullet, they simply put it back the way it was and charge the victim with owning an unlicensed machine gun.

Just to make it clear, these are all semi-automatic rifles we are talking about here. Designed and manufactured to fire one round at a time, not converted select -fire automatic rifles. As in, they have to seriously alter the gun, break it basically, to make it fire in full-auto. They've been known to do this kind of thing with knives too, making a normal lock-back into a flick knife by fiddling with the springs and the pivot joint.

Here's a little radio program about it. www.jpfo.org/media-sound/len-savage-2008-07-16.mp3

Its also all over the Internet, just google "ATF abuses" and you'll get a million hits.

Here we have the opposite problem. The RCMP don't test evidence, they just let the bad guys walk. Budget constraints, eh?

Posted by: The Phantom at July 21, 2008 10:41 AM

Tell you what, Richard. I'll just shut down the
blog. Over 9000 posts in fewer than 5 years, and yet, my grammar remains imperfect. There's no point in doing this anymore.


Posted by: Kate at July 21, 2008 10:41 AM

Practice makes imperfect! ~:D

Posted by: The Phantom at July 21, 2008 10:44 AM

Another interesting post. It simply reinforces the fact that there's no substitute for sound, emotionally detached, agenda-free science and hard-nosed, pavement-pounding, evidence-based police work.

Posted by: DrDave at July 21, 2008 10:44 AM

I didn't mention it in this post, but there is also the huge metallurgy issue in which assumptions that no two batches of ammunition would have the same "fingerprint" were proven false. No telling how many people were in jail over worthless "scientific" testimony.

Posted by: Kate at July 21, 2008 10:44 AM

Probably an English teacher and we all know how significant they are in the overall scheme of things.

Posted by: Western Canadian at July 21, 2008 10:51 AM

My favorite idiocy is the "tagged" ammunition idea. They propose to put minuscule bits of plastic in the powder, each one printed with a barcode. Idea being when the round is fired some of these little tags will survive and spew out all over the area. Super Sleuth agents will then swab up these tags and identify them under a microscope, thereby telling them Whodunnit.

Manufacturers (and wholesalers, and jobbers, and retailers, and end users) will naturally maintain 100% accurate records of who bought what lot of ammunition. Tags will never get mixed up between lots. Numbers encoded on microscopic bits of burnt plastic will never be read wrong. Tags from different ammunition will never float on the breezes and end up at a crime scene in sufficient numbers to screw up the counting and maybe get an innocent guy arrested. Most important, the introduction of hydrocarbon based plastics into gunpowder will not affect the characteristics or performance of it in any way.

No, none of these things will ever, ever happen. Maybe. Hey come on, if it just saves one life...

My second favorite are the geniuses who think they can label every single bullet and case with a unique identifying number, then keep track of it all the way until it ends up sticking out of a murder victim. Vitruvius should love that one, he's a computer software designer.

Meanwhile, our government rocket scientists can't even make a list of gun owners for under a billion dollars. Or make a -complete- list for any amount of money, for that matter.

Posted by: The Phantom at July 21, 2008 11:02 AM

too many genetic phenomenon break the "rules"

This is not a criticism, it is a question. Should the word used above been phenomena rather than phenomenon? Or am I mistaken in this case?

I believe it is a plural being referred to. I find that those two words are misused about nine of ten times that I read or hear them.

Anyone grammar teachers out there?

More to the point. Has anyone convicted on DNA evidence continued to plead not guilty? Or do they accept that they have been caught?

Posted by: John V at July 21, 2008 11:08 AM

Since we're on the broader subject of forensics now, even fingerprinting is of questionable scientific value:

"Despite the absence of objective standards, scientific validation, and adequate statistical studies, a natural question to ask is how well fingerprint examiners actually perform. Proficiency tests do not validate a procedure per se, but they can provide some insight into error rates. In 1995, the Collaborative Testing Service (CTS) administered a proficiency test that, for the first time, was “designed, assembled, and reviewed” by the International Association for Identification (IAI).The results were disappointing. Four suspect cards with prints of all ten fingers were provided together with seven latents. Of 156 people taking the test, only 68 (44%) correctly classified all seven latents. Overall, the tests contained a total of 48 incorrect identifications. David Grieve, the editor of the Journal of Forensic Identification, describes the reaction of the forensic community to the results of the CTS test as ranging from “shock to disbelief,” and added:

Errors of this magnitude within a discipline singularly admired and respected for its touted absolute certainty as an identification process have produced chilling and mind- numbing realities. Thirty-four participants, an incredible 22% of those involved, substituted presumed but false certainty for truth. By any measure, this represents a profile of practice that is unacceptable and thus demands positive action by the entire community.

What is striking about these comments is that they do not come from a critic of the fingerprint community, but from the editor of one of its premier publications."

It's no secret in Canada that the RCMP fingerprint division employees, who are civilians, are often the sons and daughters of RCMP, this nepotism racket has been well documented. This raises a huge conflict of interest: imagine if a cop gets his own son/daughter, or his best buddy's son or daughter, to "confirm" the prints of a case he is working on. Is the kid going to give the cop what he wants, which is a positive ID? In virtually all cases, I'd say yes.

Posted by: Merlot at July 21, 2008 11:09 AM

"Dick Testicle", the little children used to yell at his back as they laughed.

Those days are OVER Richard, OVER!
Come here, who needs a hug?
There there, let it out , that's it; Let it ALL out!

Posted by: richfisher at July 21, 2008 11:20 AM

Does this mean O.J. may be innocent after all?

Posted by: Joe Molnar at July 21, 2008 11:33 AM

Guess being a Conservative in PEI would lead to something like appointing oneself language policeman of the blogs. Right Richard? You must be pretty lonely out there.

Do you also do punctuation and paragraphing?

Please don't take offense, this is all in jest!

Posted by: Liz J at July 21, 2008 11:42 AM

I think us beat upped on Richard enough.

Please return to topic.

Posted by: Kate at July 21, 2008 11:56 AM

Joe Molner,

I was thinking something along the same lines. But given the rest of the evidence, I still think that he did it.

Posted by: Half Canadian at July 21, 2008 12:04 PM

Damn. I wanted to be the first to crack an OJ joke. Joe Molnar ruined it by not even being funny about it.

Posted by: Brad in Waterloo at July 21, 2008 12:18 PM

Gee Phantom, you happen to get a visit in the middle of the night by the ATF? Last time I heard they were on the good guy's side.

As for the DNA, in order to be 100% accurate don't all the markers have to match? Also, since a lot of breeds of our four pawed friends has been regulated in that for the most part sires and bitches are chosen, would it be more likely to go back to the Adam and Eve of most breeds? Just how diverse are certain breeds? I know a lot of breeds went through a "popular and trendy" stage where questionable bloodlines were being bred in puppy mills. That ended up producing a lot of family pets that were plumb loco. Dalmations, St Bernards and Cocker Spaniels all come to mind.

Posted by: Texas Canuck at July 21, 2008 12:33 PM

Amazing, how when reading quickly and seeing SDA printed here so often, I "saw" SDA when reading DNA. It happened a few time with this post.

But, perhaps it is not a perception thing. After all, DNA is the intelligence in all things living and SDA is, compared to MSM, the intelligence to all news information, heh !!

Posted by: ron in kelowna at July 21, 2008 1:03 PM

Hey folks, I am really with Bill O'Reilly in believing O.J. guilty as sin regarding Nicole, despite the "Quadrillionzillion" DNA scenario post!

Posted by: Joe Molnar at July 21, 2008 2:46 PM

Obama is a fraudulent candidate with a trumped-up Resume. Just what exactly is a 'community organizer' anyway? The NYT is protecting Obama from McCain exposing him for what he really is. That's okay, just wait until after the conventions - McCain will whip Obama and the American people, who aren't stupid, will vote for the Republicans, because at the end of the day, post 9-11 they want to be safe.

Posted by: Lori at July 21, 2008 3:19 PM

If there were only 5,000 reproducing females alive around 70,000 years ago, were they all members of a common racial/ethnic stock? Is 70,000 years (say, 2,800 generations) a long enough timeframe to produce such radically different strains as Norwegians, Nigerians and Koreans? Or did different ethno-racial groupings exist among the 5,000 reproducing females before the population bottleneck of 70,000 years ago?

Posted by: JP at July 21, 2008 4:38 PM

Lies ... damned lies ... statistics.

What's worse?

IMO - the willful misrepresentation of facts and/or drawing unsubstantiated conclusions by leaping from a simple fact to the assumption that it 's mere existence "proves" something nefarious.

In my experience Machiavellian conspiracy theories are ALWAYS BS. The answer to why things are as they are always comes down to the fact that human beings are essentially to lazy to do most things right.

Posted by: OMMAG at July 21, 2008 5:30 PM

*cough* kill the death penalty ;) *cough*

Posted by: barjebus at July 21, 2008 7:01 PM

[quote]The answer to why things are as they are always comes down to the fact that human beings are essentially to lazy to do most things right.[/quote]

OMMAG,
That is the Truth... I would challenge anyone to write a 1950's "Job Description" for someone to manually record “Temperature” for some unknown future use. Maybe a Robot!

Its been my experience that Engineers will reject BS tasks as unworthy of their time, and some would even fake the records...

Posted by: Phillip G. Shaw at July 21, 2008 7:36 PM

If the Chimerism doesn't fit, you must acquit.

Posted by: J Cochrane at July 21, 2008 8:38 PM

Texas Canuck, you live in Texas and you've forgotten the Branch Davidians already? That beauty cluster f- was ATF all the way.

The ATF isn't on the "good guy's" side. They are on their own side. They do not seek justice, they seek advancement in the federal bureaucracy food chain and damn anybody who gets in their way. They -make stuff up- so as to get in the newspapers. That's evil, in my book.

I would go so far as to say the only US government entity more fundamentally broken and more dangerous to the US public is the Immigration department. ATF is meaner, but Immigration has them beat on pure, uncaring bureaucratic sloth. The only way to get work out of an Immigration employee is to render their lard ass for oil and burn it in a diesel engine.

Posted by: The Phantom at July 21, 2008 10:58 PM

To accompany this thread, I wonder if Vitruvius can find "Sweet Gene Vincent" by Ian Dury and the Blockheads? Yes, I know it's not that kind of "sweet gene", but still...

Posted by: ebt at July 22, 2008 1:44 PM
Site
Meter