Statistical methodology plays a key role in ensuring that DNA evidence is collected, interpreted, analyzed and presented correctly. With the recent advances in computer technology, this methodology is more complex than ever before. There are a growing number of books in the area but none are devoted to the computational analysis of evidence. This book presents the methodology of statistical DNA forensics with an emphasis on the use of computational techniques to analyze and interpret forensic evidence.
Preface. List of figures. List of tables.
1. Introduction. 1.1
Statistics, forensic science and the law. 1.2 The use of statistics in
forensic DNA. 1.3 Genetic basis of DNA profiling and typing technology. 1.3.1
Genetic basis. 1.3.2 Typing technology. 1.4 About the book.
2. Probability
and statistics. 2.1 Probability. 2.2 Dependent events and conditional
probability. 2.3 Law of total probability. 2.4 Bayes' Theorem. 2.5 Binomial
probability distribution. 2.6 Multinomial distribution. 2.7 Poisson
distribution. 2.8 Normal distribution. 2.9 Likelihood ratio. 2.10 Statistical
inference. 2.10.1 Test of hypothesis. 2.10.2 Estimation and testing. 2.11
Problems.
3. Population genetics. 3.1 Hardy-Weinberg equilibrium. 3.2 Test
for Hardy-Weinberg equilibrium. 3.2.1 Observed and expected heterozygosities.
3.2.2 Chi-square test. 3.2.3 Fisher's exact test. 3.2.4 Computer software.
3.3 Other statistics for analysis of a population database. 3.3.1 Linkage
equilibrium. 3.3.2 Power of discrimination. 3.4 DNA profiling. 3.5
Subpopulation models. 3.6 Relatives. 3.7 Problems.
4. Parentage testing. 4.1
Standard trio. 4.1.1 Paternity index. 4.1.2 An example. 4.1.3 Posterior odds
and probability of paternity. 4.2 Paternity computer software. 4.2.1 Steps in
running the software. 4.2.2 The software to deal with an incest case. 4.3 A
relative of the alleged father is the true father. 4.4 Alleged father
unavailable but his relative is. 4.5 Motherless case. 4.5.1 Paternity index.
4.5.2 Computer software and example. 4.6 Motherless case: relatives involved.
4.6.1 A relative of the alleged father is the true father. 4.6.2 Alleged
father unavailable but his relative is. 4.6.3 Computer software and example.
4.7 Determination of both parents. 4.8 Probability of excluding a random man
from paternity. 4.9 Power of exclusion. 4.9.1 A random man case. 4.9.2 A
relative case. 4.9.3 An elder brother case: mother available. 4.10 Other
issues. 4.10.1 Reverse parentage. 4.10.2 Mutation. 4.11 Problems.
5. Testing
for kinship. 5.1 Kinship testing of any two persons: HWE. 5.2 Computer
software. 5.3 Kinship testing of two persons: subdivided populations. 5.3.1
Joint genotype probability. 5.3.2 Relatives involved. 5.4 Examples with
software. 5.5 Three persons situation: HWE. 5.6 Computer software and
example. 5.7 Three persons situation: subdivided populations. 5.7.1 Standard
trio. 5.7.2 A relative of the alleged father is the true father. 5.7.3
Alleged father unavailable but his relative is. 5.7.4 Example. 5.7.5 General
method and computer software. 5.8 Complex kinship determinations: method and
software. 5.8.1 EasyPA-In-1-Minute software and the method. 5.8.2
EasyPAnt-In-1-Minute. 5.8.3 EasyIN-In-1-Minute. 5.8.4 EasyMISS-In-1-Minute.
5.8.5 Other considerations: probability of paternity and mutation. 5.9
Problems.
6. Interpreting mixtures. 6.1 An illustrative example. 6.2 Some
common cases and a case example. 6.2.1 One victim, one suspect and one
unknown. 6.2.2 One suspect and two unknowns. 6.2.3 Two suspects and two
unknowns. 6.2.4 Case example. 6.2.5 Exclusion probability. 6.3 A general
approach. 6.4 Population in Hardy-Weinberg equilibrium. 6.5 Population with
multiple ethnic groups. 6.6 Subdivided population. 6.6.1 Single ethnic group:
simple cases. 6.6.2 Single ethnic group: general situations. 6.6.3 Multiple
ethnic groups. 6.7 Computer software and example. 6.8 NRC II Recommendation
4.1. 6.8.1 Single ethnic group. 6.8.2 Multiple ethnic groups. 6.9 Proofs.
6.9.1 The proof of Equation (6.6). 6.9.2 The proof of Equation (6.8). 6.9.3
The proof of Equation (6.9). 6.9.4 The proofs of Equations (6.11) and (6.12).
6.9.5 The proofs of Equations (6.14) and (6.15). 6.10 Problems.
7.
Interpreting mixtures in the presence of relatives. 7.1 One pair of
relatives: HWE. 7.1.1 Motivating example. 7.1.2 A probability formula. 7.1.3
Tested suspect with an unknown relative. 7.1.4 Unknown suspect with a tested
relative. 7.1.5 Two related persons were unknown contributors. 7.1.6 An
application. 7.2 Two pairs of relatives: HWE. 7.2.1 Two unknowns related
respectively to two typed persons. 7.2.2 One unknown is related to a typed
person and two other. unknowns are related. 7.2.3 Two pairs of related
unknowns. 7.2.4 Examples. 7.2.5 Extension. 7.3 Related people from the same
subdivided population. 7.3.1 Introductory example. 7.3.2 A simple case with
one victim, one suspect and one relative. 7.3.3 General formulas. 7.3.4 An
example analyzed by the software. 7.4 Proofs. 7.4.1 Preliminary. 7.4.2 The
proof of Equation (7.5). 7.4.3 The proof of Equation (7.7). 7.4.4 The proof
of Equation (7.9). 7.4.5 The proof of Equation (7.11). 7.4.6 The proof of
Equation (7.13). 7.4.7 The proofs of Equations (7.18) and (7.20). 7.5
Problems.
8. Other issues. 8.1 Lineage markers. 8.2 Haplotypic genetic
markers for mixture. 8.3 Bayesian network. 8.4 Peak information. 8.5 Mass
disaster. 8.6 Database search. Solutions to Problems. Appendix A: The
standard normal distribution. Appendix B: Upper 1% and 5% points of w2
distributions. Bibliography. Index.
Wing Kam Fung - Department of Statistics and Actuarial Science, University of Hong Kong With 20 years of lecturing and research experience, Professor Fung has been invited many times to give talks at workshops and international conferences. He has written over 130 papers in statistics, DNA profiling and forensic science, and is currently an Associate Editor for 4 statistical journals. Yue-Qing Hu - Department of Mathematics, Southeast University An Associate Professor, Yue-Qing Hu has published over 30 refereed papers, a number of these with Wing Kam Fung.
