Forensic Statistics and DNA-evidence

6 EC

Semester 2, period 4

5274FSDE6Y

Owner Master Forensic Science
Coordinator dr. Maarten Blom
Part of Master Forensic Science, year 1

Course manual 2020/2021

Course content

Forensic Statistics is relevant for all forensic disciplines and the Bayesian paradigm connects them. The area where this is most pronounced and most developed is DNA evidence. For this reason, the focus of the course will be on DNA and biological trace evidence.

This Course has two main topics:

  1. The biology and genetics of DNA-evidence, as well as the current methods for the analysis  and interpretation of DNA traces:
  2. The forensic statistical methods used for the evaluation and combination of (biological) evidence, both at source and activity level.

The Bayesian paradigm for computation of the value of evidence is applied to DNA evidence of increasing complexity, ranging from single source profiles to more complicated DNA evidence such as DNA mixtures and low template DNA profiles.  The issues covered will include: DNA database searches, Kinship analysis and Disaster Victim identification (DVI), Bayesian Networks for Activity level evaluation and the combination of multiple evidence, and Familial Searching.

Study materials

Literature

  • Butler, J. M. (2005) Forensic DNA typing biology, technology, and genetics of STR markers . 2nd ed. London: Elsevier Academic Press.

  • Statistics and the Evaluation of Evidence for Forensic Scientists, C. Aitken and F. Taroni, Wiley, 2020. 

  • Butler, J. (2011) Advanced topics in forensic DNA typing: methodology. Academic Press.

Syllabus

  • Lecture notes and power point presentations

  • Reader Forensic Statistics and DNA-evidence

Software

  • Hugin

  • LRmix / Euroformix

Other

  • the textbooks are available online via the UvA library

Objectives

  • 1. apply knowledge of forensic biology, biological trace examination and DNA-analysis to problems from crime scene, identification and individualization of biological evidence.
  • 2. apply the Bayesian paradigm for Forensic Statistics to DNA-evidence, i.e. to compute match probabilities for standard DNA profiles and, for example, mixtures, relatedness issues and database search,
  • 3. apply some relevant basic statistical procedures to test the validity of DNA-matching procedures and models
  • 4. create Bayesian networks to address e.g. activity-level questions or a combination of evidence
  • 5. Apply a relevant statistical analysis in problems of Kinship, relatedness in pedigrees, DVI and Familial searching
  • 6. Evaluate a comprehensive request for examination of biological trace evidence and to formulate hypotheses and alternative hypotheses at activity level.
  • 7. Evaluate the evidential strength of complex DNA profiles using (semi) continuous statistical tools

Teaching methods

  • Lecture
  • Self-study
  • Presentation/symposium
  • Computer lab session/practical training

Lectures, tutorials consisting of exercises, mock crime scenes, report writing on the basis of mock and real laboratory data (DNA profiles) and group presentations of studied literature.

Learning activities

Activity

Hours

Hoorcollege

40

Presentatie

4

Tentamen

3

Self study

121

Total

168

(6 EC x 28 uur)

Attendance

This programme does not have requirements concerning attendance (OER part B).

Assessment

Item and weight Details

Final grade

50%

Tentamen

Must be ≥ 5.5, Mandatory

10%

Bayesian Framework exercise hand in

Must be ≥ 5.5, Mandatory

15%

Bayesian network

Must be ≥ 5.5, Mandatory

10%

Group Presentations Biological Topics

Must be ≥ 5.5, Mandatory

15%

DVI Assignment

Must be ≥ 5.5, Mandatory

All components will be graded on a scale from 1 to 10, with a maximum of one decimal after the point. These grades are used to calculate the final grade. In order to pass the course, all components and the final grade have to be sufficient, i.e. at least a five and a half. When a student has not fulfilled this requirement, the examiner will register the mark ‘did not fulfill all requirements’ (NAV) whether or not the averaged grade is sufficient.

The components will be weighted as follows:

  1. Hand in exercises Island Problem (10%).
  2. Group assignment on Bayesian networks (15%).
  3. Group assignment mixture analysis (0%; not graded).
  4. Group presentations on biological topics (identification and presumptive testing )(10%).
  5. Individual DVI assignment, to hand in at the final exam (15%)
  6. Final exam (50%)


    Assignments 1 and 2 together cover statistical topics (for 25%). Assignments 4 and 5 cover topics in forensic biology (for 25%). Assignment 3 is not handed in; correct answer is given in tutorial session.

    The exam of this course will be a written examination based on the content covered during the lectures. The exam consists of two parts, a part about the biology in the course and a part about the statistical aspects of DNA analysis. Both count for half of the  score for the exam.

    The final grade will be announced at the latest on 14th of April (= 15 working days after the final course activity). Between the 14th of April to May 12th (= 35 working days after the final course activity) a post-exam discussion or inspection moment will be planned. This will be announced on Canvas and/or via email.


    Table of specification

     

    Exit qualifications

    Learning outcome

    Components (see above)

    1

    2

    3

    4

    5

    6

    7

    8

    9

    10

    1

    4, 6

     

    x

     

     

    x

     

     

     

     

     

    2

    1, 3, 6

    x

     

     

    x

     

     

     

     

     

     

    3

    6

    x

     

     

    x

     

     

    x

     

     

     

    4

    2

    x

     

     

    x

     

     

    x

     

     

     

    5

    5, 6

    x

     

     

    x

     

     

    x

     

     

     

    6

    2, 6

     

     

     

     

    x

     

    x

     

     

     

    7

    6

    x

     

     

     

     

     

    x

     

     

     

    Table 1: Table of specification: the relation between the learning outcomes of the course, the assessment components of the course and the exit qualifications of the Master’s Forensic Science (see the course catalogues for the exit qualifications). 

Assignments

The students will be (randomly) divided in groups of five students. Assignments 2 and 4 will be done in groups and graded as such. Assignments 1 and 5 will be individual and will be so graded. Assignment 3 will be done in a group and will not be graded.

1         Hand in exercise Island Problem

During the course the so called Island problem will be reviewed. In the lecture notes there are some exercises concerning this type of problems. The students will be given one such exercise  a solution of which they have to hand in individually. The solutions will be graded on a scale of one to ten.

2         Group assignment Bayesian Networks

The theory of Bayesian networks will be done during a lecture. For this assignment the students are working in groups of about five. There will be an introduction to the program HUGIN, a program to analyse Bayesian networks. The groups will be given a certain practical situation where a network can be built. The groups have to make their own network in HUGIN and motivate their choices. The solutions will be graded on a scale of one to ten.

3         Group assignment mixture analysis

The theory of analysis of mixtures will be done during a lecture. The assignment consists of performing an analysis of a DNA sample using computer software for the calculation of the evidential strength of DNA mixtures.

4         Group presentation

The students will be divided in groups of about five. Each group is assigigned to a topic on the identification and presumptive testing  of tissular origin if human biological traces. The group selects a relevant scientific publication or case review on this topic and prepares a presentation on their findings. The presentations are graded with the assessment form in the appendix.

5         Individual DVI assignment

This is an individual assignment where the students are given DNA profiles of  unidentified victims in a mass grave and a list of DNA profiles of possible relatives. The assignment consists of a computation based on these profiles to identify the victims.

The calculations and report will be graded on a scale of one to ten.

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

Weeknummer Onderwerpen Studiestof
1

The biology and genetics of DNA-evidence / (Bayesian) statistics

  
2

DNA-typing and Single source DNA evaluation

  
3

From single source to mixed DNA evaluation

  
4

Bayesian Networks and Case examination

  
5

Kinship, DVI and complex DNA-mixtures

  
6

Activity level evaluation and Continuous Models for complex mixtures

  
7

Mixture exercises / Q & A

  
8 Exam  

Timetable

The schedule for this course is published on DataNose.

Last year's course evaluation

In order to provide students some insight how we use the feedback of student evaluations to enhance the quality of education, we decided to include the table below in all course guides.

Forensic Statistics and DNA evidence (6EC) N=23  
Strengths
  • Combination of Biology and Statistics and perspectives from different lecturers
  • Assignments throughout the course, especially the DVI assignment was appreciated
 Notes for improvement
  • One of the lecturers had a soft voice and was difficult to hear.
  • There was limited feedback on the Bayesian network assignment (although some comments indicate students did receive ample feedback). There were examples of Bayesian networks, but students found it difficult to understand how to go from the info they had, to building a network.
  • The exam was too long in total. The statistics part was too long, but it was representative. The DNA part was doable, but was not perceived as representative by the students.
Response lecturer:
  • The course overall is evaluated rather well, but has a low average grade. There is a large spread around the average grade, ranging from 1.0 to 9.0. We suspect the average grade of the course suffered from the postponement in the exam and the sudden switch to a timed take home exam, due to the corona crisis, causing a lot of stress with the students.
  • Previous years we did not get feedback that the teacher was difficult to hear. It turns out we had more square room in the past and now we had rectangular rooms. Part of the students stayed in the back even though there was room in the front. We will try to arrange a microphone coming year. We don’t know how the corona situation will develop. There is a chance that the lectures might be online and no additional microphone is necessary.
  • Also regarding the Bayesian Network assignment the feedback is very varied. Some students indicate they did get feedback. We suspect these are the students who took the initiative to email the teacher with their questions. The teacher will review how to integrate a feedback moment for all students.
  • If next time, we would have to organise a timed take home exam again, we would not use timeslots for different parts of the exam. This will hopefully lower stress levels and as students really need the 3 hrs to make the exam, we expect that the chances for plagiarism are minimal

Contact information

Coordinator

  • dr. Maarten Blom