Forensic DNA Databanks in the USA

Michael T. Risher

DNA databanks are computer databases containing the results of DNA analysis (DNA profiles), used to link evidence from a crime scene to a previously unknown suspect or to another crime scene

CODIS

• Combined• DNA• Index• System

• Local (LDIS)• State (SDIS) • National (NDIS)

CODIS-NDIS

Convicted Offender Index

Arrestee Index

Forensic (Crime Scene) Index

(Missing Person Index)

(Missing Person Reference Index)

(Population Index)

What’s In a CODIS Entry?

DNA Profile – the “genetic fingerprint”

Specimen Identifier

Agency that submitted the profile

Name of lab tech who analyzed sample

• Each DNA profile consists of information taken from 26 sites on the subject’s DNA– 13 paired pieces of information

How is CODIS expanding and Why?

CODIS is Growing Fast

2000 2007 2010 June 2011Offender Profiles 460,365 5,070,473 7,833,009 9,878,881Forensic Profiles 22,484 194,785 302,002 380,702

http://www.fbi.gov/hq/lab/codis/clickmap.htm

Recent California Stats.

CODIS CreepConvicted of sex offenses or homicide (Cal. 1989)Convicted of violent feloniesConvicted of any felony (Prop. 69: Nov. 2004)Convicted of misdemeanorArrested for sex offenses and homicideArrested for violent feloniesArrested for any felony (Prop. 69: Jan. 1, 2009)Detained Immigrants (2009)Arrested for anything (federal 2009)Universal databank?

The Fourth Amendment

“The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.”

• Warrantless searches “are per se unreasonable under the Fourth Amendment-subject only to a few specifically established and well-delineated exceptions.”– Arizona v. Gant, 129 S. Ct. 1710, 1716 (2009)

DNA Sampling is a Search

• Intrudes into the body.

• Intrudes into informational privacy.

• Each separate analysis of sample reveals more information and is a separate search.

Whose DNA is being Collected?

• In 2006, of 534,460 felony arrests reported (includes multiple-offense arrests):

• Violent offenses accounted for 23.6 percent (126,342).

• Property offenses accounted for 27.1 percent (144,781).

• Drug offenses accounted for 28.9 percent (154,468).

• All other offenses accounted for 20.4 percent (108,869).

Challenges to Arrestee Testing(post-conviction testing settled)

• Minn. Ct. of Appeals – struck down

• Va. Supreme Court – upheld

• USA v. Mitchell – just upheld

• USA v. Pool – upheld, rehearing pending

• Haskell v. Brown – appeal pending

Cal. Population and Arrests by Race/Ethnicity(2006/2007)

Group In General Population Persons Arrested for Felony

African American 6.7% 21.5%

Hispanic 36.2% 40.6%

Non-Hisp. Caucasian 42.7% 32.4%

Familial Searching

• Effectively expands databank to include family members – guilt by bloodline

• DNA databank is used to identify people who are by definition not the culprits

• Leads to investigation and potential harassment of previously convicted people who are no longer committing crimes, and their families

Familial Searching– Treat like a wiretap

• Should be governed by public, enforceable rules made by legislature, not informal guidelines developed by law enforcement

• Only in very serious cases where other investigatory techniques have been exhausted

• Court order or warrant• Reporting requirements – how many

applications, orders, to what effect

So What?

A Lifetime of Genetic Surveilance

• Hard to get removed, even if proved innocent

• Being in the databank makes you more likely to be arrested in the future – You’ll get caught for the crimes you do

commit, while people not the databank won’t– Maybe get accused of crimes you didn’t

commit– Familial Searching

• DNA contains personal information

– Ancestry/Family Secrets – Predisposition to physical disease– Predisposition to mental illness?– Predisposition to alcoholism/addiction?– Sexual orientation? – Predictor of personality traits? – Predisposition to criminality?? XYY

Even if these are baseless, DNA analysis may be wrongly used to predict behavior.

DNA Dragnets

• Police seek samples from large numbers of people who live or work in the vicinity of the crime scene or fit a loose profile of the perpetrator.

• Individuals are asked to “voluntarily” provide a DNA sample.

• 19 known dragnets in U.S.• ~7,000 people

Mistakes have occurred…

Lazaro Soto Lusson faced life in jail and was incarcerated for over a year before it was discovered that the Las Vegas police crime lab had switched his DNA sample with that of his cellmate.

Josiah Sutton spent nearly five years in jail for a rape he did not commit. His conviction

rested almost entirely on a DNA test performed and erroneously interpreted and reported by a Houston Crime Lab analyst.

Possible Exceptions

• Search Incident to Arrest– Must relate to officer/jail security or

preservation of evidence

Possible Exceptions

• Special Needs– Primary Purpose must be something other

than law enforcement. – Still requires balancing of special needs

against privacy interests.

Possible Exceptions

• General Balancing– Never applied to searches of persons other

than persons convicted of felonies.– If police can take the genetic profile of people

who are presumed innocent, simply based on arrest by single officer, what can’t they do?

– Costs and benefits --

Does Expansion to Arrestees

• Encourage pretextual arrests?

• Aggravate racial disparities?

• Aggravate testing backlogs?

• Invade genetic privacy?

• Invade bodily integrity?

• Solve crimes?

• Prevent crimes?

• Affect innocent people?

Does Arrestee Testing Solve Crimes?

• Everybody arrested will soon either be convicted (and give a sample) or not be convicted (in which case what’s the government’s interest in taking DNA from an innocent person?).

The California Experience

1 Total number of CAL-DNA hits (as of 11/30/09) (Supp. Konzak Dec. ¶ 5) 10,893

2 Number of hits involving samples originally submitted from arrestees (includes those later convicted) (Supp. Konzak Dec. ¶ 5)

453

3 Total number of hits involving samples originally submitted from convicted persons (line 1minus line 2)

10,440

4 Annual number of people arrested for a felony in California (2007 statistics) (rounded) (Meier Dec. Ex. C (Cal. Dep’t of Justice, Crime in California 2007at 147, Table 37))

332,000

5 Annual number of California felony arrestees who were eventually convicted of a crime based on this arrest (2007) (rounded) (id.)

231,000

6 Annual number of California felony arrestees who were not convicted of a crime (line 4 minus line 5)

101,000

7 Percentage of felony arrestees who are eventually convicted (line 5 divided by line 4)

69.6%

8 Percentage of felony arrestees not eventually convicted (line 6 divided by line 4, or 100% minus line 7)

30.4%

9 Total number of individual profiles in the CA-DNA database (Supp. Konzak Dec. ¶ 5)

1,480,294

10 Total number of arrestees in database (includes those subsequently convicted and those never convicted) (Supp. Konzak Dec. ¶ 5; see Konzak Dec. ¶ 34 (arrestee samples remain in arrestee database even after conviction)

134,280

11 Estimated number of persons in arrestee database who have not been, and will not be, convicted (line 8 times line 10)

40,821

12 Total number of convicted-person profiles excluding felony arrestees not convicted (line 9 minus line 11)

1,439,473

13 Hit rate for database containing convicted individuals (line 3 divided by line 12)

0.725%

14 Expected number of hits from the arrestee database if hit rate for all arrestee profiles uploaded to database (including profiles from persons later convicted and from those not later convicted) is the same as the hit rate for convicted profiles (line 13 times line 10)

974

15 Expected number of hits from the arrestee database if the hit rate for persons subsequently convicted is the same as hit rate for the convicted-person database (as it should be) and the hit rate for persons arrested but not subsequently convicted is 0% (line 14 times line 7)

678

16 Actual total number of hits obtained from arrestee database (Supp. Konzak Dec. ¶ 5)

453

15 Expected number of hits from the arrestee database if the hit rate for persons subsequently convicted is the same as hit rate for the convicted-person database (as it should be) and the hit rate for persons arrested but not subsequently convicted is 0% (line 14 times line 7)

678

16 Actual total number of hits obtained from arrestee database (Supp. Konzak Dec. ¶ 5)

453