DNA as a Biometric Biometric Consortium Conference 2011 Tampa, FL September 27, 2011 Dr. Peter M. Vallone Biochemical Science Division National Institute of Standards and Technology Gaithersburg, MD 20899
Outline Basics of DNA Typing DNA as a Biometric
General Characteristics of Genomic DNA Each individual has a unique DNA profile with exception of monozygotic siblings Each person's DNA is the same in every cell DNA from skin cells will match DNA from blood cells An individual s DNA profile remains the same throughout life Half of your DNA comes from your mother and half from your father implications for determining kinship
Sources of Biological Evidence Saliva Blood Semen Urine Hair Teeth Bone Tissue Blood Sample Only a very small amount of blood is needed to obtain a DNA profile best results with >100 cells, but DNA profiles can be recovered from fewer cells
Forensic DNA Testing Probe subsets of genetic variation in order to differentiate between individuals (14 to 16 regions in the human genome) DNA typing must be done efficiently and reproducibly (information must hold up in court) Over 10 million profiles in the national FBI database Typically, we are not looking at genes little/no information about race, predisposition to disease, or phenotypic information (eye color, height, hair color) is obtained
What Type of Genetic Variation? Sequence Variation single nucleotide polymorphisms (SNPs) insertions/deletions GCTAGTCGATGCTC[G/A]GCGTATGCTGTAGC Length Variation short tandem repeats (STRs) CTAGTCGT[GATA][GATA][GATA]GCGATCGT
Short Tandem Repeat (STR) Markers An accordion-like DNA sequence that occurs between genes TCCCAAGCTCTTCCTCTTCCCTAGATCAATACAGACAGAAGACA GGTGGATAGATAGATAGATAGATAGATAGATAGATAGATAGATA GATAGATATCATTGAAAGACAAAACAGAGATGGATGATAGATACA TGCTTACAGATGCACAC = 12 GATA repeats ( 12 is all that is reported) 7 repeats 8 repeats 9 repeats 10 repeats 11 repeats 12 repeats 13 repeats Target region [short tandem repeat] The number of consecutive repeat units can vary between people The frequency of these repeats observed in the general population have been sampled and are used for the statistical representation of a DNA profile
Core STR Loci for the United States Position of Forensic STR Markers on Human Chromosomes TPOX 13 Core U.S. STR Loci D3S1358 D5S818 D8S1179 TH01 VWA 1997 FGA CSF1PO D7S820 AMEL Sex-typing D13S317 D16S539 D18S51 D21S11 AMEL
Technology Genetics Biology Steps in Forensic DNA Analysis Usually 1-2 day process (a minimum of ~8 hours) Blood Stain Buccal swab Sample Collection & Storage 1.5 h 1.5 h DNA Extraction DNA Quantitation Statistics Calculated DNA Database search Paternity test Reference sample ~3.5 h Multiplex PCR Amplification DNA separation and sizing Applied Use of Information 1.5 h STR Typing Interpretation of Results
Identifiler [Applied Biosystems] 15 STR Loci Kit Information is tied together with multiplex PCR and data analysis D8S1179 {15,16} D21S11 D8S1179 {29,29} D7S820 {9,11} CSF1PO {10,11} D3S1358 {16,17} TH01 D3S1358 {6,7} TH01 D13S317 {8,12} D16S539 {10,11} D2S1338 {19,19} D19S433 {14,16} D19S433 D5S818 AMEL D5S818 {9,11} VWA VWA {15,17} TPOX {8,12} D18S51 {11,15} Amel {X,Y} FGA {19,22} D21S11 Multiplying the frequency of each D13S317 TPOX FGA genotype D7S820 at CSF1PO each locus gives us the Random Match Probability (RMP) D16S539 of 1.25x10 D2S1338-15 for unrelated individuals The chance of an unrelated individual D18S51 having this exact same profile is 1 in 800 trillion This test contains the 13 FBI core loci
Kinship Testing DNA profiles can also be used to evaluate the probability of a specific familial relationship As a familial relationship becomes more distant, the ability of DNA (using 13-15 STRs) to confirm the likelihood of that relationship decreases 1. Parent-offspring 2. Siblings 3. Half siblings = uncle/nephew = grandchild 4. Cousins
Dad Autosomal Paternity Example Child Brother Sister Mom
DNA as a Biometric
Current Biometrics Some commonly measured features Physical Fingerprints (Palm/hand geometry) Iris, retinal Face Odor/scent DNA Behavioral Gait Voice Vein (IR thermogram) Hand geometry Handwriting
Characteristics of a Biometric Universality each person should have the characteristic Uniqueness is how well the biometric separates individuals from another Permanence measures how well a biometric resists aging and variance over time Collectability ease of acquisition for measurement Jain, A. K.; Ross, Arun; Prabhakar, Salil [January 2004], "An introduction to biometric recognition", IEEE Transactions on Circuits and Systems for Video Technology 14th [1]: 4 20
Characteristics of a Biometric (practical considerations) Performance accuracy, speed, and robustness of technology used Acceptability degree of approval of a technology Circumvention/Spoofing ease of use of a substitute Jain, A. K.; Ross, Arun; Prabhakar, Salil [January 2004], "An introduction to biometric recognition", IEEE Transactions on Circuits and Systems for Video Technology 14th [1]: 4 20
DNA Typing as a Biometric Advantages High level of accuracy (Gold Standard) Solid scientific foundation of Forensic DNA Testing (pop stats, molecular biology, court acceptance, protocols, training, education) Kinship determination (unique to DNA) Potential use for: Phenotype (traits; eye/hair color) Biogeographical Ancestry (but not with STR markers) Expensive Challenges Time consuming Sample collection (invasive, stability issues) Technical expertise required for analysis Policy/Privacy/Ethical issues
Interest in Rapid DNA Typing DoD (field testing, rapid intelligence, mass fatalities) DHS (kinship determination, border security, immigration) DoJ (law enforcement, arrestees, initial information) Industry (security, authentication) Each customer will have specific requirements sample input information output degrees of accuracy Time required for generating STR profiles will have to be reduced to less than 2 h. Does the application warrant the time and expense?
Goals for Rapid DNA Typing Systems Develop a fully integrated system capable of performing DNA testing in less than 1-2 hours Little user interaction (or experience) Rugged Swab in answer out Robust Simple data interpretation 4-16 samples per run Disposable chips (with reagents on board)
Rapid DNA Typing Systems Under Development Systems are currently under development These are STR-based and use similar genetic marker systems as law enforcement (CODIS-FBI NDIS) Network Biosystems (Woburn, MA) http://www.netbio.com ZyGEM and Lockheed Martin (Charlottesville,VA) http://www.zygem.com IntegenX (Pleasanton, CA) http://www.integenx.com Forensic Science Service (UK) and Univ of AZ http://www.forensic.gov.uk/ Tomorrow: Special Rapid DNA Session 9:00 AM Session 2 - Rooms 15/16
Questions about the limitations of DNA Identical (monozygotic) twins Occurrence ~1 in 285 births Standard forensic DNA tests can not distinguish between identical twins Fingerprints will be different Random Match Probability does not apply to related individuals Combine biometric modes (DNA + fingerprints) If possible, question when enrolling an individual
Questions about the limitations of DNA Chimeras (different DNA types within the same person) Example: Blood may exhibit one DNA type, but saliva another, OR a combination of both Inherited, acquired from transplant or transfusion Occurrence??? If the DNA profile indicates a mixture, repeat DNA typing to confirm Question when enrolling an individual Combine biometric modes
Birthday Problem 365 possible birthdays Assume no leap year, that all birthdays are equivalent, no bias In a room of 23 people what is the probability that two of them will share a birthday???? 1/365 = 0.27% Answer: There is a ~50% probability of two people sharing a birthday in a room of 23 people. Relevant to multiple occurrences of any birthday http://en.wikipedia.org/wiki/birthday_problem Weir, B. The Rarity of DNA Profiles The Annals of Applied Statistics (2007) 1: 358 370.
These are different questions One to one Many to many The estimated frequency at which a particular STR profile would be expected in a population The estimated frequency at which two STR profiles will match in a set of n profiles Kaye, David H., Trawling DNA Databases for Partial Matches: What is the FBI Afraid of? (February 1, 2009). Cornell Journal of Law and Public Policy, Vol. 19, No. 1, 2009; Penn State Legal Studies Research Paper No. 8-2010.
Thank you for your attention! Questions? Peter.Vallone@nist.gov 301-975-4872 Acknowledgements Erica Butts Outside funding agencies: FBI - Evaluation of Forensic DNA Typing as a Biometric Tool NIJ Interagency Agreement with the Office of Law Enforcement Standards