Friday, February 27, 2009

NPR CLAIMS Regarding National Academy of Science Report on Forensics

The committee found there isn't enough research behind many forensic techniques to show how accurate and reliable they are.

It also says forensic labs are underfunded and understaffed, and there's no mandatory certification to ensure quality.

It says forensic analysts' court testimony commonly refers to evidence being a "match" or "consistent with" a suspect, even though no forensic method except for DNA analysis "has been rigorously shown to have the capacity to consistently, and with a high degree of certainty, demonstrate a connection between evidence and a specific individual or source."


While autosomal DNA has passed a Daubert test on whether it is scientific, it receives very little needed review that would show it is not the "gold standard" of forensics.

In Illinois they found 903 pairs of profiles of separate individuals matching at nine or more loci in a database of about 220,000.

In Maryland, in a database of fewer than 30,000 profiles, 32 pairs matched at nine or more loci. Three of those pairs were identical at 13 out of 13 loci.


RECOMMENDED READING:
The danger of DNA: It isn't perfect
The verdict is out on DNA profiles
Top lab repeatedly botched DNA tests
Crime labs finding questionable DNA matches:
FBI tries to keep national database away from lawyers

Beware When They Claim a "DNA Match"

Not with any case particularly, but in general, not all DNA tests are equal. They depend on suspect samples which may not be of high quality. If you have no other evidence, they should not be automatically considered as a gold standard in determining guilt beyond a reasonable doubt. Autosomal DNA is dependent on how many loci are found. The FBI tests and records 13 loci. Suspect samples may have 13, 10, 8, 5, or only one loci that show up. The fewer the loci, the more the matches.

Thursday, February 19, 2009

NRA: The Truth About "Ballistic Fingerprinting"

Faking Fingerprints

NAS: Forensic Needs Report


If you think of Forensics being like “CSI,” the report released on February 18, 2009, by the National Academy of Sciences will disabuse you of that notion.

Fingerprinting, ballistics, blood spatter, bite marks, forensic methods are being challanged. The crime labs are just not meeting standards for scientific investigation. Decades of casework are being called into question.How we move forward will have a far-reaching impact on crime labs, courts — and American criminal justice.

There are basic questions being asked about the validity of forensic evidence and its use in court. Academy President of the The American Academy of Forensic Sciences AAFS, Carol Henderson, JD, said, "It is important to recognize that the truth does not belong to a side in litigation and that the access to forensic science evidence should be available to everyone." This is a subject that is in the Supreme Court regarding an inmate who has been refused DNA testing in Alaska.

A 2010 symposium of the NAS will be, "Putting Our Forensic House in Order: Examining Validation and Expelling Incompetence."

At this point, unlike doctors and lawyers, those in forensics do not have professional organizations that license and can sanction practitioners in the U.S. or in the States.

The forensic science structure is fragmented and cannot provide a framework to eliminate or control standards, practice, and education. That raises the question of viability in the current system on the national level.

Tuesday, February 17, 2009

DNA Identification Matching and Mitochondrial DNA

One consumer vendor of DNA testing, Family Tree DNA, when they notify their customers of a match by email state:

When comparing the Hyper Variable Regions 1 and 2 of your mtDNA, a match has been found between you and another person(s) in the Family Tree DNA database. Matches of the mtDNA have more Anthropological significance as the time frame for a common ancestor could go beyond the Genealogical time frame.

Anthropological time for human mitochondrial DNA is essentially from about 200,000 years ago to the present. It is measured in age with relative measurements like carbon 14 testing rather than history.

The Genealogical time frame is the last portion of that since the beginning of the use of surnames. The extreme end of that is about 1000 years ago beginning with tracing lineages in royalty.

The formation of surnames began about 900 years ago and did not become common until about 400 years ago. James the Baker became James Baker. Fred the Blacksmith may have become Fred Black or Fred Smith.

Y Chromosome DNA, is passed on with a typical mutation rate of .02% or two in 10,000. As it is passsed by the male lineage, it tends to represent a paternal lineage that begins with a few mutations that coincide with the formation of surnames.

This is unlike the autosomal DNA used by the FBI which changes with each individual offspring and is a combination of genes from both parents.

Mitochondrial DNA (mtDNA), on the other hand, is passed from mother to daughter in the maternal lineage with a lower rate of mutation than the yDNA. While it is also passed from mother to son, that son does not pass his mtDNA to his offspring.

In one case, mtDNA of a 9,000 old "Cheddar Man" in the United Kingdom was traced down to a current resident of the UK teaching school in the area where Cheddar Man was found.

So, while two individuals may have the same mtDNA, it does not confirm they have a common ancestor within the timeline that surnames and genealogies have existed. If two individuals do not have the same mtDNA, therefore, it does exclude that possibility of a common ancestor.

The "Hyper Variable Regions" of the mtDNA are areas where the most, but not only, change occurs. The mtDNA can be thought as a circle or loop which breaks at the top during the process of reproduction.

The mitochondria are separate from the chromosomes and reproduce independently of them. There can be hundreds of mitochondria within one individual cell. Since there are so many copies of a relatively small chain of DNA, it is more easily tested.

Not all mtDNA testing is alike. For comparison, a full genome of mtDNA in humans is normally 16,569 base pairs of DNA. It is extremely small in comparison with the millions of base pairs in a single chromosome, of course. Consumers can have this tested for less than $500.This level of testing does give a level of resolution that indicates a particular maternal lineage.

Region one can be tested for less than $200 and both regions can be tested for less than $300. However, the testing done on both of the Hyper Variable Regions by consumers only covers only 1143 base pairs of that 16,569 base pairs. You could have differences outside those Hyper Variable Regions that occured over 60,000 years ago that are not detected by such testing. This level of testing does not give a level of resolution that indicates a particular maternal lineage.

However, this level of resolution is greater than the 721 base pairs used by the FBI. The mtDNA testing by the FBI is useful for exclusion if suspects, but it cannot be used to identify an individual as the one in an evidence sample.

Even if there is a "match" at that level of resolution, the suspect and the one the evidence sample came from could be not only separate individuals, but not even related for tens of thousands of years.

So, the use of mitochondrial DNA testing has its limits in genealogy and certainly in the courts.

Friday, February 13, 2009

LAPD Fingerprint Forensics Folly

In a study on perception, cognition and expertise, by Thomas A. Busey, Ph.D. Experimental Psychology, of Indianna University and Itiel E. Dror, Ph.D., Senior Lecture of Cognitive Science at the School of Psychology University of Southampton, have done a study on "Special Abilities and Vulnerabilities in Forensic Expertise."

The findings of that study reveal a great flaw in the identification of individuals from suspect samples. When fingerprint comparisons are made, they have the standard of checking each other's work, causing that influence to affect the results. The study says, "The very fact that identifications will be verified (sometimes by more than one verifier) introduces a whole range of issues, from diffusion of responsibility (Darley & Latané, 1968) to conformity, attention, self-fulfilling prophecies wishful thinking. "

Recently, a senior fingerprint analyst for the LAPD was arrested for allegedly sexually assaulted one of his tenants in an apartment in South L.A. Miguel Martinez Rivera, 50, of Montebello was taken into custody Wednesday morning shortly after reporting to work. He has has worked in the LAPD's crime lab for over 20 years. Of the fingerprint staff, he has the highest authority of the civilian workers and reviews the work of the newer and junior staff.

This would not be so serious if the LAPD has not already had the fingerprint examiners who falsely implicated at least two people in crimes have been linked to nearly 1,000 other criminal cases. At least six print analysts with the LAPD latent print section have made critical errors in their work.

The LAPD would do well to pay out the half million dollars needed to reform the unit by hiring an outside firm to review practices and protocols of the 80-person fingerprint unit before the civil actions begin.

Monday, February 9, 2009

Magic Trunks and the DNA "Gold Standard"

For those of you that have followed the Anthony case, what a magical trunk! It has all kinds of vapors and DNA that prove Caylee is dead. Not that that can be scientifically proven.

No DNA sample alone can prove the donor is dead. Nor can any vapor. If it smelled dead, likely an investigator had a bad meal as the same gases can be involved.

As far as googling for ways to create chlorophorm or breaking necks, that could have been done by the kidnapper(s) .

Supposing that a death HAS occured, that eliminates the aggravated murder because the chlorophorm causes the heart to stop while unconcious.

If you mix chlorine (bleach) and acetone (nail polish remover) or isopropal alcohol (rubbing alcohol) this can accidently produce chlorophorm. Obviously, a child can do this - or an adult by accident.

The more I read these articles, the less I trust the AP, the media who copy them, or the prosecutor for eliminating the alternative hypothesis.

The DNA in this case was a "high resolution" test of the mitochondrial DNA. I have had this test. I have over 278 matches at http://www.familytreedna.com/ .

That was part of a full genome sequence (FGS) of my mitochondrial DNA, which I have no matches to in a database of over 80238. I have a copy of the report. Their DNA test is of less than 1319 base pairs. My test was of 16,570 base pairs. (I have one insertion more, which is not rare.) This means my FGS has 12 times more resolution than the FBI test.

I read a Certificate of Analysis from the Virgina Deparment of Forensic Analysis which used 16 markers instead of the 13 that the FBI does for more resolution. The technition said:

This information is provided only as an investigative lead, and any possible connection or involvement of this individual to this case must be determined through futher investigation.

SO, these tests are not the end all in investigations. They add to the proponderence of evidence, but they do not make a case "beyond a reasonable doubt."

Whenever I find that DNA testing is being done, I want to know:

What were the DNA tests that were run?

In the CODIS database of Maryland , of fewer than 30,000 profiles, 32 pairs matched at nine or more loci. Three of those pairs were "perfect" matches, identical at 13 out of 13 loci. Experts say they most likely are duplicates or belong to identical twins or brothers, but they did not establish that.

A study of the Arizona CODIS database carried out in 2005 showed that approximately 1 in every 228 profiles in the database matched another profile in the database at nine or more loci, that approximately 1 in every 1,489 profiles matched at 10 loci, 1 in 16,374 profiles matched at 11 loci, and 1 in 32,747 matched at 12 loci.

In a case against a Murillo-Sosa, a jury was told the match was 5 of 13. They had to say he was not guilty. SO, what kind of matches are they making? How about doing it right the first time and take the time to get a FULL DNA testing of Y chromosome DNA and mitochondrial DNA, as well as autosomal (CODIS) DNA?

Is it a mitochondrial DNA test which matches everyone who descended from the same maternal ancestor in the last 20 generations as THE DEFENDENT(S) ? Is it just a "high resolution test" which the FBI performs or a real full genome sequencing of the mitochondrial DNA?

Is it a Y Chromosome test which matches everyone who has the same paternal lineage for the past 400 years? If so, is it 12 markers, which could be one to fifty percent of the population, or is it a 67 marker test that can pin it down to a surname?

There may be a way to combine all of these that will resolve the identity to one person, but is what they have beyond a reasonable doubt unless they do FULL testing?