Kinship testing

The application of DNA profiling to kinship analysis is widespread and offers an easy means of establishing biological relationships. Not surprisingly, paternity testing is the most common form of kinship testing, with hundreds of thousands of tests being performed worldwide each year [1]. Since the first DNA based kinship test in 1985 [2], DNA analysis has been applied to larger numbers of kinship tests, to the testing of more complex relationships and to the identification of highly compromised human remains.

Paternity testing

PCR-based STR profiling has now become the standard tool and the PowerPlex

16 (Promega) and AmpF/STR Identifiler (Applied Biosystems) STR kits that can analyse 15 loci simultaneously are routinely used (see Chapter 6). Laboratories that undertake kinship testing often have over 20 genetic markers at their disposal, including STR markers on the X and Y chromosomes [3], that allow for the testing of complex relationships [4]. The sensitivity of STR analysis, while not essential for most forms of paternity testing, allows samples to be routinely collected using buccal swabs [1] and has expanded the possible scenarios where it can be used, for example, the analysis of low amounts of DNA recovered from foetal cells [5, 6] (Figure 11.1).

The methodology used to produce DNA profiles for paternity testing is identical to the analysis of material recovered from crime scenes (see Chapters 4 to 7). The interpretation of results is more complex than when comparing profiles from crime scenes and suspects. If the tested man does not possess the alleles that have been inherited from the biological father we can conclude that he cannot be the biological father. However, because mutations between the father and child could lead to a false exclusion at any given loci [1, 7-10], it is standard practice is to require an exclusion at three or more loci before a test is declared negative.

If we cannot exclude the tested man as being the biological father then we have to assign a value to indicate the significance of non-exclusion. Likelihood ratios (see Chapter 9), which consider two competing, and mutually exclusive hypotheses are

Dp

12D 1ED

TUD

EM

200

2402

120D." 6CO_ 0_

J ill ill -I.I

L

I i J

-Ufl A

B | G:| V |;J£|k: Y: | i

TSf

Figure 11.1 A STR profile generated from foetal cells recovered from amniotic fluid in early pregnancy. In this case it was possible to determine that pregnancy had resulted from a rape, and allowed an informed decision to be made on whether or not to have the foetus aborted. The profile was generated using the AmpF/STR Profiler Plus STR kit (Applied Biosystems) (see plate section for full-colour version of this figure)

Figure 11.1 A STR profile generated from foetal cells recovered from amniotic fluid in early pregnancy. In this case it was possible to determine that pregnancy had resulted from a rape, and allowed an informed decision to be made on whether or not to have the foetus aborted. The profile was generated using the AmpF/STR Profiler Plus STR kit (Applied Biosystems) (see plate section for full-colour version of this figure)

used. The hypotheses are:

The tested man is the biological father = H, p

The tested man is not the biological father = Hd

The symbols Hp and Hd were introduced in Chapter 9 when comparing the proposition or hypothesis put forward by the prosecution (Hp) compared with the hypothesis put forward by the defence (Hd), although in many civil cases the terms prosecution and defence are not appropriate. This likelihood is called a paternity index (PI) and can be assessed using equation (11.1).

where Pr is probability.

To calculate this likelihood ratio, we compare the probability of the child's genotype (Gc) given the mother's (Gm) and tested man's genotype (Gtm), if the tested man is the biological father (Hp) and the probability of the child's genotype given the mother's and tested man's genotype, if the tested man is not the biological father (Hd).

The numerator and denominator are conditional on the genotypes of the mother, child and tested man. They can be derived using a 'Punnet square'.

Punnettsquare

The equations are not difficult to understand, particularly if derived from a Punnett square and converting to text form. Consider the case where the mother is genotype a, b and child is genotype b, c and the alleged father is c, d. If he is the father then the mother must pass on allele b, and the father must pass on allele c. If he is not the father then the mother must still pass on allele b but some other man must pass on allele c. This is given in the Punnett square below.

Alleles from alleged father

c d

Alleles from

a

a, c a, d

mother

b

b, c b, c

If the alleged father is the biological father then this can happen one in four ways, with a probability of 0.25.

If the alleged father is not the biological father, then the mother must pass on allele b with probability of 0.5 and the chance that a male other than the alleged father is the father is dependent upon the frequency of allele c (pc) in the population. This gives a likelihood ratio of:

The same process can be used for any of the possible combinations. Consider the version where the alleged father is homozygous (b, b) and the mother heterozygous (a, b) and the child is heterozygous (a, b)

Alleles from alleged father

b b

Alleles from

a

a, b a, b

mother

b

b, b b, b

If the alleged father is the biological father then this can happen in two ways, with a probability of 0.5.

If the alleged father is not the biological father then the mother must pass on allele b with probability of 0.5 and the chance that a male other than the alleged father is the father is dependent upon the frequency of allele b ( pb) in the population. This gives a likelihood ratio of:

Consider a case when the mother is a, b, the child is a, b and the alleged father is a, c.

Alleles from alleged father a c

Allele a or b could be passed from the mother. Note that if she passed on allele a then this would be an exclusion and therefore it would need to be allele b that is passed from mother to child if the man is the biological father. Considering the numerator (Hp) genotype a,b occurs in only one of four ways (0.25). Considering the denominator (Hd) the mother passed on either allele a (0.5) or allele b (0.5) and the chance that either event took place, allele a or allele b, is the sum of the probabilities. This results in the equation below:

In Table 11.1 all the potential combinations of alleles from a mother, child and tested man are shown along with the resulting numerator, denominator and PI equation.

Table 11.1 The numerator and denominator that should be used when calculating a paternity index are determined by the genotypes of the child (G C), mother (G M), and tested man (G TM). The alleles are represented by i, j, k and i where i = j = k = i. Reproduced from Lucy, 2006 [11] p. 174, with permission from John Wiley & Sons (originally based on Evett and Weir, 1998 [12] p. 168)

Table 11.1 The numerator and denominator that should be used when calculating a paternity index are determined by the genotypes of the child (G C), mother (G M), and tested man (G TM). The alleles are represented by i, j, k and i where i = j = k = i. Reproduced from Lucy, 2006 [11] p. 174, with permission from John Wiley & Sons (originally based on Evett and Weir, 1998 [12] p. 168)

Gc

g m

gtm

Numerator

Denominator

PI

AiAi

AiAi

AiAi

1

Pi

\

AiAj

/

Pi

/ P

AjAk

0

Pi

0

AiAj

AiAi

/

Pi/2

'/Pi

AiAj

/

Pi/2

/ P i

AiAk

/

Pi/2

/ P

Aj Ak

0

Pi/2

0

AiAj

AiAi

AjAj

1

Pj

11/Pj

AiAj

/

Pj

/2 Pj

AjAk

/

Pj

1/2Pj

AkAi

0

Pj

0

AiAj

AiAi

/

(Pi + PJ)/,

1/( Pi + Pj)

AiAj

v2

(Pi + PJ //

1/ Pi +P j

AiAk

/4

(P i + PJ /

/2( Pi +Pj)

AjAk

74

(Pi + PJ /

/2( Pi +Pj)

AkAi

0

(Pi + PJ //

0

AiAk

AjAj

v2

P/2

1//Pj

AiAj

V4

%

/2 Pj

AjAk

V4

%

/2 Pj

AjAi

V4

%

/2 Pj

AkAi

0

%

0

We can apply the formula in Table 11.1 to the paternity case presented in Table 11.2.

We can apply the formula in Table 11.1 to the paternity case presented in Table 11.2.

Table 11.2 The result of a paternity test using the Powerplex® 16 STR Kit (Promega). The alleles that the child could have inherited from the mother are underlined and the alleles that are from the biological father are shown in bold. The i,j,k, l symbols correspond to symbols in Table 11.1. The allele frequencies were taken from Marino et al. (2006) [13]

Table 11.2 The result of a paternity test using the Powerplex® 16 STR Kit (Promega). The alleles that the child could have inherited from the mother are underlined and the alleles that are from the biological father are shown in bold. The i,j,k, l symbols correspond to symbols in Table 11.1. The allele frequencies were taken from Marino et al. (2006) [13]

Child (GC)

(g tm)

Num

Denom

PI

PI

D3S1358

15'

- 15'

142 -

15'

15' - 19*

/

PI

1/2 p'

0.3239

1.54

VWA

172

- 18'

16* -

18'

172 - 18'

I

P/2

1/2 Pi

0.2715

1.84

D16S359

11'.

12 2

11i-

13*

122 - 13*

'k

Pi/2

1/2 Pi

0.2773

1.80

D8S1179

10'

- 13 2

10i-

132

10' - 10'

'k

P' + P|

1/P' + Pi

0.0630

2.73

0.3033

D21S11

30' -

32.22

30i -

31*

27' - 32.22

Xl4

Pi/2

1/2 Pi

0.1245

4.02

D18S51

13i

142

13i-

142

12* - 142

Xl4

P' + Pi/2

1/2( P' + Pi)

0.1326

1.48

0.2063

THO1

9' -

9.32

9i-

9.32

6*- 9.3

74

Pi + Pi/2

1/2( P' + Pi)

0.1407

1.24

0.2624

FGA

18i

232

18i-

25*

23 - 23

72

P /2

1/P

0.1440

6.94

D13S317

8i-

13j

8i-

11*

11*- 13

74

P /2

1/2 Pi

0.1444

3.46

CSF1PO

11i

- 11'

11i-

132

11' - 132

74

P'/2

1/2 P'

0.2916

1.71

D7S820

9i

- 9'

9i-

102

9' - 11*

74

P'/2

1/2 P'

0.0998

5.01

TPOX

8 2

10'

10i-

11*

8 -8

72

P /2

1/P

0.5243

1.91

D5S818

11i

- 122

11i-

122

11' - 122

72

P' + Pi/2

1/P' + Pi

0.3618

1.51

0.2992

Penta D

13

- 15'

12* -

15'

12' - 132

74

P /2

1/2 Pi

0.1726

2.90

Penta E

10'-

182

10i-

10'

16*- 18

72

P

1/2 Pi

0.0304

16.4

Combined PI 2,920,823

Combined PI 2,920,823

The combined PI is calculated by applying the product rule and multiplying the PI from each locus in this case the PI is 2920823. This can be represented by this statement:

Statement of positive paternity

The results of the DNA testing are 2 920 823 times more likely if the tested man is the biological father of the child than if the biological father is another man, unrelated to the tested man.

The significance of likelihood ratios can be difficult for lay people to evaluate and the results are often presented as a probability of paternity, making the results more accessible. To calculate a probability of paternity requires Bayesian analysis and takes into consideration non-genetic evidence: the likelihood ratio (LR) is multiplied by

Table 11.3 The impact of prior probabilities on the probability of paternity is shown with two paternity indexes: one with a value of 1000 and the other taken from the above example, with a value of 2 920 823

Paternity index

Prior Odds 1000 2 920 823

0.0001 0.0010 0.0100 0.1000 0.5000 0.7500 0.9000

0.090 917 356 0.500 250125 0.909 918107 0.991080 278 0.999 000 999 0.999 666 778 0.999 888 901

¡588 329 ) 658 09 966 107 996 919 999 658 999 886 999 962

the prior odds of paternity that are determined by non-genetic evidence, such as the testimony of the woman. It can be calculated using equation (11.2).

Probability of paternity =

r( H„l non-genetic evidence)

LR x Pr(Hp | non-genetic evidence)

LR x Pr(Hp |non-genetic evidence) + [1 — Pr(Hp|non-genetic evidence)]

Taking the above paternity test it is possible to turn the likelihood ratio into a probability of paternity for any prior odds of paternity; for example:

Prior probability = 0.1

2920823 x 0.1

Probability of paternity =-= 0.999996919

When this figure is used to report the results of a test it is often quoted as a percentage, which is more accessible to non-scientists. In this case the probability of paternity would be quoted as 99.9997%.

The value that is attributed to the prior odds of paternity is, of course, subjective. In civil cases, the value of 0.5 is commonly used, although there is little scientific merit to this value. In criminal cases, probabilities of paternity are often not presented because it is the duty of the jury/judge to assess the prior odds of paternity. If results are presented as a probability of paternity, a range of values calculated using different prior odds is often quoted (Table 11.3).

IDENTIFICATION OF HUMAN REMAINS 111

IDENTIFICATION OF HUMAN REMAINS 111

Figure 11.2 The identification of human remains recovered from an air crash [35]. Blood samples were provided by the mother and father who were missing a son. Alleles in the profile of human remains could have come from the mother and father (indicated by the arrows). The profiles were generated using the AmpF/STR Profiler Plus STR kit (Applied Biosystems) (see plate section for full-colour version of this figure)

Figure 11.2 The identification of human remains recovered from an air crash [35]. Blood samples were provided by the mother and father who were missing a son. Alleles in the profile of human remains could have come from the mother and father (indicated by the arrows). The profiles were generated using the AmpF/STR Profiler Plus STR kit (Applied Biosystems) (see plate section for full-colour version of this figure)

With low paternity indexes the impact of prior odds can be significant. However, with the possibility of analysing a large number of STR loci, the PIs are typically in the millions and the posterior probability of paternity is therefore extremely high, even when the prior odds are very low. In the paternity test presented above, even with the prior odds as low as 0.001, the probability of paternity is still 99.9966 %.

In addition to the standard paternity testing where the mother, child and alleged father are available, testing can also be carried out when the mother is not available [14-17]. More complex relationships can be examined, such as determination of sibship [18] and paternity tests to discriminate between close relatives [12,19-21]. Calculations can also incorporate correction factors to allow for deficiencies in allele frequency databases, in particular, the effects of subpopulations [22,23] (see Chapter 8). Fortunately, computer programs have been developed to deal with both routine and complex scenarios [24-30].

Identification of human remains

The first application of DNA analysis to the identification of human remains was in 1987, when skeletal remains were profiled using single nucleotide polymorphisms in the DQa locus [31, 32]. Unfortunately, this system did not have high powers of discrimination and it was not until the early 1990s that DNA profiling was successfully applied to the identification of human remains [33, 34]. As DNA profiling technology and methodology have evolved to be more robust and powerful, it has been applied to increasingly complex situations including the identification of people killed in air crashes [29, 35-38]; fire [39-42]; terrorist attacks [43-46]; natural disasters [47] and war [48-51]. STRs are the most commonly used tool but mitochondrial DNA (see Chapter 13) and SNPs (see Chapter 12) have also been employed on occasion.

The matching of human remains can be through comparison to personal objects that belonged to the missing person, such as combs and toothbrushes [52], or by comparison to close family members (Figure 11.2).

In cases that involve hundreds of victims, the statistical analysis becomes very complex. Because of the high number of pair-wise comparisons that are made between the victims and relatives, the potential for coincidental matches that result in false positives and ultimately misidentifications is significant [47, 52, 53]. The existence of relatives within the population of victims also complicates the analysis [29, 47] and there are limitations as to what can be achieved.

Further reading

Balding, D.J. (2005) Weight-of-evidence for Forensic DNA Profiles. John Wiley & Sons, Ltd, Chichester, pp. 82-134.

Buckleton, J., Triggs, C.M., and Walsh, S.J. (2005) Forensic DNA Evidence Interpretation. CRC Press, pp. 341-437.

References

1. American Association of Blood Banks (2004) Annual report summary for testing in 2004. http://www.aabb.org/content/Accreditation/Parentage_Testing_Accreditation_Program/ptprog. htm

2. Jeffreys, A.J., et al. (1985) Positive Identification of an immigration test-case using human DNA fingerprints. Nature 317, 818-819.

3. Szibor, R., et al. (2003) Use of X-linked markers for forensic purposes. International Journal of Legal Medicine 117, 67-74.

4. Junge, A., et al. (2006) Mutations or exclusion: an unusual case in paternity testing. International Journal of Legal Medicine 120, 360-363.

5. Makrydimas, G., et al. (2004) Early prenatal diagnosis by celocentesis. Ultrasound in Obstetrics and Gynecology 23, 482-485.

6. Makrydimas, G., et al. (2004) Prenatal paternity testing using DNA extracted from coelomic cells. Fetal Diagnosis and Therapy 19, 75-77.

7. Brinkmann, B., et al. (1998) Mutation rate in human microsatellites: influence of the structure and length of the tandem repeat. American Journal of Human Genetics 62, 1408-1415.

8. Gunn, P.R., et al. (1997) DNA analysis in disputed parentage: the occurrence of two apparently false exclusions of paternity, both at short tandem repeat (STR) loci, in the one child. Electrophoresis 18, 1650-1652.

9. Leopoldino, A.M., andPena, S.D.J. (2003) The mutational spectrum of human autosomal tetranu-cleotide microsatellites. Human Mutation 21, 71-79.

10. Chakraborty, R., and Stivers, D.N. (1996) Paternity exclusion by DNA markers: effects of paternal mutations. Journal of Forensic Sciences 41, 671-677.

11. Lucy, D. (2005) Introduction to Statistics for Forensic Scientists. John Wiley and Sons, Ltd, Chichester.

12. Evett, I.W., and Weir, B.S. (1998) Interpreting DNA Evidence. Sinauer Associates, Inc.

13. Marino, M., et al. (2006) Population genetic analysis of 15 autosomal STRs loci in the central region of Argentina. Forensic Science International 161, 72-77.

14. Chakraborty, R., et al. (1994) Paternity evaluation in cases lacking a mother and nondetectable alleles. International Journal of Legal Medicine 107, 127-131.

REFERENCES

15. Poetsch, M., et al. (2006) The problem of single parent/child paternity analysis - Practical results involving 336 children and 348 unrelated men. Forensic Science International 159, 98-103.

16. Thomson, J.A., et al. (2001) Analysis of disputed single-parent/child and sibling relationships using 16 STR loci. International Journal of Legal Medicine 115, 128-134.

17. Brenner, C.H. (1993) A note on paternity computation in cases lacking a mother. Transfusion 33, 51-54.

18. Wenk, R.E., et al. (1996) Determination of sibship in any two persons. Transfusion 36, 259-262.

19. Ayres, K.L., and Balding, D.J. (2005) Paternity index calculations when some individuals share common ancestry. Forensic Science International 151, 101-103.

20. Fung, W.K., et al. (2002) Power of exclusion revisited: probability of excluding relatives of the true father from paternity. International Journal of Legal Medicine 116, 64-67.

21. Wenk, R.E., and Chiafari, F.A. (2000) Distinguishing full siblings from half-siblings in limited pedigrees. Transfusion 40, 44-47.

22. Ayres, K.L. (2000) Relatedness testing in subdivided populations. Forensic Science International 114, 107-115.

23. Fung, W.K., et al. (2003) Testing for kinship in a subdivided population. Forensic Science International 135, 105-109.

24. Fung, W.K. (2003) User-friendly programs for easy calculations in paternity testing and kinship determinations. Forensic Science International 136, 22-34.

25. Riancho, J.A., and Zarrabeitia, M.T. (2003) A Windows-based software for common paternity and sibling analyses. Forensic Science International 135, 232-234.

26. Brenner, C.H. (1997) Symbolic kinship program. Genetics 145, 535-542.

27. Fung, W.K., et al. (2006) On statistical analysis of forensic DNA: theory, methods and computer programs. Forensic Science International 162, 17-23.

28. Brenner, C.H. DNA.VIEW. (http://dna-view.com/)

29. Leclair, B., et al. (2004) Enhanced kinship analysis and STR-based DNA typing for human identification in mass fatality incidents: The Swissair Flight 111 disaster. Journal of Forensic Sciences 49, 939-953.

30. Buckleton, J., et al. (2005) Forensic DNA Evidence Interpretation. CRC Press.

31. Stoneking, M., et al. (1991) Population variation of human mtDNA control region sequences detected by enzymatic amplification and sequence-specific oligonucleotide probes. American Journal of Human Genetics 48, 370-382.

32. Blake, E., et al. (1992) Polymerase chain-reaction (Pcr) amplification and human-leukocyte antigen (Hla)-Dq-Alpha oligonucleotide typing on biological evidence samples - casework experience. Journal of Forensic Sciences 37, 700-726.

33. Hagelberg, E., et al. (1991) Identification of the skeletal remains of a murder victim by DNA analysis. Nature 352, 427-429.

34. Sajantila, A., et al. (1991) The polymerase chain-reaction and postmortem forensic identity testing - application of amplified D1S80 and HLA-DQ-Alpha loci to the identification of fire victims. Forensic Science International 51, 23-34.

35. Goodwin, W., et al. (1999) The use of mitochondrial DNA and short tandem repeat typing in the identification of air crash victims. Electrophoresis 20, 1707-1711.

36. Olaisen, B., et al. (1997) Identification by DNA analysis of the victims of the August 1996Spits-bergen civil aircraft disaster. Nature Genetics. 15, 402-405.

37. Hsu, C.M., et al. (1999) Identification of victims of the 1998 Taoyuan Airbus crash accident using DNA analysis. International Journal of Legal Medicine 113, 43-46.

38. Ballantyne, J. (1997) Mass disaster genetics. Nature Genetics. 15, 329-331.

39. Clayton, T.M., et al. (1995) Further validation of a quadruplex STR DNA typing system: A collaborative effort to identify victims of a mass disaster. Forensic Science International 76, 17-25.

40. Clayton, T.M., et al. (1995) Identification of bodies from the scene of a mass disaster using DNA amplification of short tandem repeat (STR) loci. Forensic Science International 76, 7-15.

41. Meyer, H.J. (2003) The Kaprun cable car fire disaster - aspects of forensic organisation following a mass fatality with 155 victims. Forensic Science International 138, 1-7.

42. Calacal, G.C., et al. (2005) Identification of exhumed remains of fire tragedy victims using conventional methods and autosomal/Y-chromosomal short tandem repeat DNA profiling. American Journal of Forensic Medicine and Pathology 26, 285-291.

43. Kahana, T., et al. (1997) Suicidal terrorist bombings in Israel - identification of human remains. Journal of Forensic Sciences 42, 260-264.

44. Budimlija, Z.M., et al. (2003) World trade center human identification project: experiences with individual body identification cases. Croatian Medical Journal 44, 259-263.

45. Holland, M.M., et al. (2003) Development of a quality, high throughput DNA analysis procedure for skeletal samples to assist with the identification of victims from the world trade center attacks. Croatian Medical Journal 44, 264-272.

46. Budowle, B., et al. (2005) Forensic aspects of mass disasters: strategic considerations for DNA-based human identification. Legal Medicine 7, 230-243.

47. Brenner, C.H. (2006) Some mathematical problems in the DNA identification of victims in the 2004 tsunami and similar mass fatalities. Forensic Science International 157, 172-180.

48. Huffine, E., et al. (2001) Mass identification of persons missing from the break-up of the former Yugoslavia: structure, function, and role of the International Commission on Missing Persons. Croatian Medical Journal 42, 271-275.

49. Holland, M.M., et al. (1993) Mitochondrial-DNA sequence-analysis of human skeletal remains -identification of remains from the Vietnam War. Journal of Forensic Sciences 38, 542-553.

50. Boles, T.C., et al. (1995) Forensic DNA testing on skeletal remains from mass graves - a pilot project in Guatemala. Journal of Forensic Sciences 40, 349-355.

51. Lleonart, R., etal. (2000) Forensic identification of skeletal remains from members of Ernesto Che Guevara's guerrillas in Bolivia based an DNA typing. International Journal of Legal Medicine 113, 98-101.

52. Brenner, C.H., and Weir, B.S. (2003) Issues and strategies in the DNA identification of World Trade Center victims. Theoretical Population Biology 63, 173-178.

53. Gornik, I., et al. (2002) The identification of war victims by reverse paternity is associated with significant risks of false inclusion. International Journal of Legal Medicine 116, 255-257.

Was this article helpful?

0 0

Post a comment