Establishing parameters for objective interpretation of

DNA profile evidence

Forensic Bioinformatics (www.bioforensics.com)

Dan E. Krane, Wright State University, Dayton, OH

Steelman Visiting Scientist Lecture Series, Lenoir-Rhyne University, April 9, 2010

The science of DNA profiling is sound.

But, not all of DNA profiling is science.

The science of DNA profiling is sound.

There is plenty of opportunity for improvement.

The science of DNA profiling is sound.

There is plenty of opportunity for improvement.

I. Interpretation

II. Statistical weighting

Doesn’t someone either match or not?

Opportunities for subjective interpretation?

Who can be excluded?

Suspect D3 vWA FGATom 17, 17 15, 17 25, 25Dick 12, 17 15, 17 20, 25Harry 14, 17 15, 17 20, 25Sally 12, 17 15, 15 20, 22

Signal Measure

μb

μb + 3σb

μb + 10σb

Mean backgroundSignal

Detection limit

Quantification limit

Measu

red

sig

nal (I

n V

olt

s/R

FUS

/etc

)

Saturation

0

Many opportunities to measure baseline

Measurement of baseline in control samples:

• Negative controls: 5,932 data collection points (DCPs) per run ( = 131 DCPs)

• Reagent blanks: 5,946 DCPs per run ( = 87 DCPs)

• Positive controls: 2,415 DCP per run ( = 198 DCPs)

RFU levels at all non-masked data collection points

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 0

R F U

Count

Variation in baseline noise levelsPositive Control μb σb μb + 3σb μb + 10σb

Maximum 6.7 6.9 27.4 75.7 Averag e 5.0 3.7 16.1 42.0 Minimu m 3.7 2.4 10.9 27.7

Negat ive Cont rol μb σb μb + 3σb μb + 10σb

Maximum 13.4 13.2 53.0 145.4 Averag e 5.4 3.9 17.1 44.4 Minimu m 4.0 2.6 11.8 30.0

Reagen t Blank μb σb μb + 3σb μb + 10σb

Maximum 6.5 11.0 39.5 116.5 Averag e 5.3 4.0 17.3 45.3 Minimu m 4.0 2.6 11.8 30.0

All three c ont rols avera g ed μb σb μb + 3σb μb + 10σb

Maximum 7.1 7.3 29.0 80 .1 Average 5.2 3.9 16.9 44.2 Minimu m 3.9 2.5 11.4 28.9

Average (μb) and standard deviation (b) values with corresponding

LODs and LOQs from positive, negative and reagent blank controls in 50 different runs. BatchExtract: ftp://ftp.ncbi.nlm.nih.gov/pub/forensics/

Lines in the sand: a two-person mix?

Two reference samples in a 1:10 ratio (male:female). Three different thresholds are shown: 150 RFU (red); LOQ at 77 RFU (blue); and LOD at 29 RFU (green).

Mixed DNA samples

QuickTime™ and aPhoto - JPEG decompressor

are needed to see this picture.

How many contributors to a mixture if analysts can discard a locus?

How many contributors to a mixture?

Maximum # of alleles observed in a 3-person mixture # of occurrences Percent of cases

2 0 0.00

3 78 0.00

4 4,967,034 3.39

5 93,037,010 63.49

6 48,532,037 33.12

There are 146,536,159 possible different 3-person mixtures of the 959 individuals in the FB I database (Paoletti et al., November 2005 JFS).

3,398

7,274,823

112,469,398

26,788,540

0.00

4.96

76.75

18.28

How many contributors to a mixture if analysts can discard a locus?

How many contributors to a mixture?

Maximum # of alleles observed in a 3-person mixture # of occurrences Percent of cases

2 0 0.00

3 310 0.00

4 2,498,139 5.53

5 29,938,777 66.32

6 12,702,670 28.14

There are 45,139,896 possible different 3-person mixtures of the 648 individuals in the MN BCI database (genotyped at only 12 loci).

8,151

1,526,550

32,078,976

11,526,219

0.02

3.38

71.07

25.53

How many contributors to a mixture?

Maximum # of alleles observed in a 4-person mixture # of occurrences Percent of cases

4 13,480 0.02

5 8,596,320 15.03

6 35,068,040 61.30

7 12,637,101 22.09

8 896,435 1.57

There are 57,211,376 possible different 4-way mixtures of the 194 individuals in the FB I Caucasian database (Paoletti et al., November 2005 JFS). (35,022,142,001 4-person mixtures with 959 individuals.)

The science of DNA profiling is sound.

There is plenty of opportunity for improvement.

I. Interpretation

II. Statistical weighting

What weight should be given to DNA evidence?

Statistics do not lie.

But, you have to pay close attention to the questions they are addressing.

What weight should be given to DNA evidence?

Statistics do not lie.

But, you have to pay close attention to the questions they are addressing.

RMP: The chance that a randomly chosen, unrelated individual from a given population would have the same DNA profile observed in a sample.

What weight should be given to DNA evidence?

Statistics do not lie.

But, you have to pay close attention to the questions they are addressing.

RMP: The chance that a randomly chosen, unrelated individual from a given population would have the same DNA profile observed in a sample.

Consider cold hits

CODIS (Combined Offender DNA Index System)

Maintained by the FBI

Contains 7,940,321 profiles as of February, 2010.

Assisted in 109,900 investigations

What weight should be given to DNA evidence?

• Probable Case

– Suspect is first identified by non-DNA evidence

– DNA evidence is used to corroborate traditional police investigation

• Cold Hit Case

– Suspect is first identified by search of DNA database

– Traditional police work is no longer focus

Which is more damning evidence?

What weight should be given to DNA evidence?

• Probable Case

– Suspect is first identified by non-DNA evidence

– DNA evidence is used to corroborate traditional police investigation

– RMP = 1 in 10 million

• Cold Hit Case

– Suspect is first identified by search of DNA database

– Traditional police work is no longer focus

– RMP = 1 in 10 million

Which is more damning evidence?

What weight should be given to DNA evidence?

• Probable Case

– Suspect is first identified by non-DNA evidence

– DNA evidence is used to corroborate traditional police investigation

– RMP = 1 in 10 million

• Cold Hit Case

– Suspect is first identified by search of DNA database

– Traditional police work is no longer focus

– RMP = 1 in 10 million

– DMP = roughly 4 in 5

Which is more damning evidence?

What weight should be given to DNA evidence?

Statistics do not lie.

But, you have to pay close attention to the questions they are addressing.

RMP: The chance that a randomly chosen, unrelated individual from a given population would have the same DNA profile observed in a sample.

What is the relevant population?

1 in 80 quadrillion

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Popular vote in 2008 by county. McCain won red counties, Obama won blue counties.

How would you determine the frequency of Obama supporters in North Carolina?

Obama

N.C. 50.2%

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Popular vote in 2008 by county. McCain won red counties, Obama won blue counties.

How would you determine the frequency of Obama supporters in North Carolina?

Obama

N.C. 50.2%

Region 59.6%

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Popular vote in 2008 by county. McCain won red counties, Obama won blue counties.

How would you determine the frequency of Obama supporters in North Carolina?

Obama

N.C. 50.2%

Region 59.6%

U.S. 52.9%

QuickTime™ and aTIFF (Uncompressed) decompressor

are needed to see this picture.

Popular vote in 2008 by county. McCain won red counties, Obama won blue counties.

How would you determine the frequency of Obama supporters in North Carolina?

Obama

N.C. 50.2%

Region 59.6%

U.S. 52.9%

Utah? 35.5%

?

What is the relevant population?

Errors are multiplicative

What weight should be given to DNA evidence?

Statistics do not lie.

But, you have to pay close attention to the questions they are addressing.

RMP: The chance that a randomly chosen, unrelated individual from a given population would have the same DNA profile observed in a sample.

Combined probability of inclusion (CPI): a cousin of the RMP

• What fraction of a population cannot be excluded as a possible contributor to a given mixture?

Combined probability of inclusion (CPI): a cousin of the RMP

• What fraction of a population cannot be excluded as a possible contributor to a given mixture?

• What fraction of the population cannot be excluded from contributing their DNA to a given mixture if they are allowed to not match at two loci?

Matching profiles in NIST database

Population Sample ID D21 1 D21 2 D7 1 D7 2 CSF 1 CSF 2 D13 1 D13 2 D16 1 D16 2 D2 1 D2 2 D18 1 D18 2 FGA 1 FGA 2 Matching Loci

African American GT37170 28 28 10 11 12 14 11 11 11 11 22 25 12 18 22 23 7

African American OT05896 30 30 10 10 11 12 12 12 11 11 20 22 18 18 19 25 7

African American PT83891 30 32.2 11 11 11 11 12 12 11 11 19 22 12 16 22 23 7

African American PT83912 28 28 10 11 8 11 11 12 9 11 17 25 16 16 22 22 7

African American GT37169 29 32.2 10 10 7 11 12 12 9 11 19 20 14 18 23 26 6

African American MT95356 28 30 11 11 11 11 12 12 9 12 22 22 15 18 21 23 6

African American ZT79617 32.2 33.1 7 12 10 12 12 12 9 11 20 25 12 14 21 21 6

Caucasian GT36877 28 30 9 9 11 11 11 11 9 11 18 22 16 18 20 26 6

Caucasian GT38075 30 30 10 13 12 12 11 11 11 11 19 21 12 14 22 23 6

Caucasian MT94880 28 29 10 11 12 12 11 12 11 12 20 25 16 16 22 23 6

Caucasian WT52480 30 31 8 10 11 12 11 12 9 9 25 25 16 16 21 21 6

n = 257 African Americans, 302 Caucasians(1 in 21 African Americans; 1 in 38 Caucasians)

The science of DNA profiling is sound.

There is plenty of opportunity for improvement.

(particularly in the areas of interpretation and attaching

statistical weights)