MAKING SENSE OF

FORENSIC
GENETICS

What can DNA tell

you about a crime?
Published in 2017
ABOUT THIS GUIDE ACKNOWLEDGEMENTS

Working in Partnership Acknowledgements

Sense about Science’s public engagement team helps researchers to discuss We’d like to thank people who have answered specific questions, provided feedback
and present research information, guided by the public and people who will use it. over email or at user testing as we developed the guide: David Balding, David Ballard,
We draw from our extensive public networks and over a decade of working with David Bentley, Duncan Brown, Zoë Chapman, Philip Dawid, Martin Evison,
the public on some of the trickiest issues of evidence. Our ethos is public led, Dugald Foster, Nigel Hawkes, Chris Hughes, Debbie Kennett, Benedetta La Corte,
expert fed, which means engaging early and directly addressing what people are Chris Lawless, Emma Lawrence, Adrian Muller, Georgina Meakin, Nick Ross,
thinking. These partnership projects are only available for socially or scientifically Jonathan Smith, David Spiegelhalter and Mark G. Thomas.
difficult issues where researchers make a convincing case that it is a matter
of public interest and evidence is neglected, conflicting or misunderstood. In 2016, All our guides are date stamped and reflect the scientific findings and knowledge
we were approached by the European Forensic Genetics Network of Excellence available at the time of publication.
(EUROFORGEN-NoE) to help develop their guide to forensic genetics.
This project was financially supported from the European Union Seventh Framework
Programme (FP7/2007-2013) under grant agreement n° 285487 (EUROFORGEN-NoE).
A word from EUROFORGEN
This guide is designed to introduce professional and public audiences to the use
of DNA in criminal investigations; to understand what DNA can and can’t tell us
about a crime, and what the current and future uses of DNA analysis in the criminal
justice system might be.
Forensic DNA analysis is a complex area open to misinterpretation. So we set out
to provide a straightforward guide for the police, judiciary, lawyers, jurors, journalists
and those intrigued by criminal casework — in other words, anyone with an interest
in the use of DNA for crime investigation. To help us, we approached Sense about
Science and formed a public engagement partnership; they connected us with
a wide public audience who gave us invaluable feedback on the guide.
Making Sense of Forensic Genetics is the final output of our European Union
Seventh Framework Programme funded research and networking project, which
has spanned five years and ranged in expertise from forensic geneticists and social
scientists to representatives of the judiciary. The European Forensic Genetics
Network of Excellence will continue to exist independently from EC funding
to provide information and training both to the scientific community and to the
interested public.

The contributors’ disclosure of interests are available at: senseaboutscience.org/

activities/making-sense-of-forensic-genetics

2 ABOUT THIS GUIDE 3 ACKNOWLEDGEMENTS

CONTRIBUTORS

Christopher Phillips
Researcher in Forensic Genetics
at University of Santiago de
Compostela, Spain
EUROFORGEN Consortium member

Peter Schneider
Tracey Brown Professor of Forensic Molecular
Director, Genetics, Institute of Legal Medicine,
Sense about Science, UK University of Cologne, Germany
EUROFORGEN Consortium member

Linda Geddes Denise Syndercombe Court

Reader in Forensic Genetics,
Freelance science
King’s College London, UK
and medical writer, UK
EUROFORGEN Consortium member

Peter Gill Joanne Thomas

Professor of Forensic Genetics,
Projects and Events Coordinator,
University of Oslo, Norway
Sense about Science, UK
EUROFORGEN Consortium member

Emily Jesper-Mir Matthias Wienroth

Head of Partnerships Research Fellow in Social Science,
and Governance, King’s College London, UK
Sense about Science, UK EUROFORGEN Consortium member

Manfred Kayser Robin Williams

Professor of Forensic Molecular Professor of Forensic
Biology, Erasmus MC University Science Studies, Northumbria
Medical Centre Rotterdam, NL University, Newcastle-upon-Tyne, UK
EUROFORGEN Consortium member EUROFORGEN Consortium member

4 CONTRIBUTORS 5 CONTRIBUTORS
INTRODUCTION

DNA is present in most cells of our body. It is unique to each of us, In October 2011, Adam Scott was arrested
and we leave a trail of it everywhere we go. Forensic investigators and charged with raping a woman
take advantage of this, using our DNA to draw conclusions about in Manchester, UK.
where we’ve been and who we’ve interacted with. The Adam Scott case
Swabs of the woman’s genitals revealed traces of sperm, is a good example
In popular television dramas such as The Killing and Midsomer Murders, the science
and one of these swabs yielded a DNA profile that matched of confirmation bias
of forensic DNA profiling often helps identify suspects when other lines of evidence
have gone cold. Of course DNA analysis has revolutionised forensic science in real Mr Scott’s. This was the only evidence against him. — where inconvenient
life too; helping to catch prolific murderers such as the Green River Killer (see page The forensic scientist who processed the sample said: information to the
“It is estimated that the chance of obtaining matching DNA
15); enabling the remains of those killed by mass disasters and atrocities such as prosecution is ignored
the Srebrenica Massacre to be repatriated to their loved ones; and shining a light components if the DNA came from someone else unrelated
to Adam Scott is approximately 1 in a billion”2 . But Mr Scott or dismissed. The scientist
on miscarriages of justice that have seen innocent people wrongfully convicted
of serious crimes. claimed he was in his home town of Plymouth, UK (more than assumed that because
200 miles away) at the time of the attack, and had never been sperm was recovered,
Such is the power of DNA to identify, convict, and exonerate, that many to Manchester in his life. all of the male DNA must
perceive it to be infallible. Yet DNA evidence has a number of limitations:
it might be undetectable, overlooked, or found in such minute traces as to make When challenged, the scientist claimed the DNA evidence have come from the sperm
interpretation difficult. Its analysis is subject to error and bias. Additionally, DNA provided: “strong scientific support for the view that Adam (when in fact Mr Scott’s
Scott had sexual intercourse with [the victim] rather than
profiles can be misinterpreted, and their importance exaggerated, as illustrated DNA was a spit sample).
by the wrongful arrest of a British man, Adam Scott (see page 7). Even if DNA he did not.” However, this was an error. By itself, a DNA profile
is detected at a crime scene, this doesn’t establish guilt. Accordingly, DNA needs can’t provide any information about the body fluid it came Peter Gill
to be viewed within a framework of other evidence, rather than as a standalone from, or lead to the inference that sexual intercourse took
place. Two months after his arrest, mobile phone records Professor of Forensic
answer to solving crimes1. Genetics, University of Oslo
came to light that corroborated Mr Scott’s version of events;
Forensic scientists take great care to minimise errors, by ensuring their methods revealing that his mobile phone had been used in Plymouth
EUROFORGEN Consortium
have been thoroughly tested, and that they are performed by competent people member
a few hours after the reported rape. Finally, after five months
using properly calibrated equipment and following well-controlled procedures in custody, he was released.
to prevent contamination. Even so, mistakes can happen.
A subsequent investigation revealed that Mr Scott had
The purpose of this guide is to inform readers about what’s currently possible become implicated as a result of accidental contamination
with DNA testing in forensic applications, what its limitations are, and what might of samples within the lab. The day before processing samples
be possible in future. It will explain how DNA profiles are generated, what they’re from the alleged rape victim, the lab had handled a DNA
used for, and how they can be misconstrued. It will also describe cases where sample from Mr Scott, following a ‘spitting incident’ in Exeter,
DNA has been a game changer, and turned an investigation around. We hope UK. Unfortunately the disposable plastic plate used to analyse 2
it will be a useful resource to anyone who works with, or crosses paths with DNA this sample had been inadvertently reused in the rape case, R
 eport into the circumstances of a complaint
received from the Greater Manchester Police on 7
evidence in the criminal justice system. resulting in the misidentification. The true perpetrator was March 2012 regarding DNA evidence provided by
never found. LGC Forensics https://www.gov.uk/government/
uploads/system/uploads/attachment_data/
This incident highlights two important points for courts: file/118941/dna-contam-report.pdf
3
R
 oth A (2010) Safety in Numbers - Deciding When
a) DNA should not be used as the sole evidence DNA Alone is Enough to Convict. New York University
in a criminal case3,4 Law Review 85: 1130
4
T
 he Crown Prosecution Service (2014) Guidance
b) There is a considerable danger if the importance on Expert Evidence, p14
of the DNA evidence is inappropriately afforded 5
G
 ill P (2014) Misleading DNA evidence: Reasons
greater weight than other evidence5. for Miscarriages of Justice. 1st ed. Academic Press.

1
W
 ienroth M, Morling N, Williams R (2014) Technological Innovations in Forensic Genetics:
Social, Legal and Ethical Aspects. Recent Advances in DNA and Gene Sequences 8, 98-103

6 INTRODUCTION 7 INTRODUCTION
CONTENTS

01 What can we detect? 10

it’s only one part of the puzzle

DNA isn’t everything,
DNA can come from almost all types of biological sources and is analysed using a variety
of techniques. Which technique investigators choose depends on the amount of DNA available
and the questions they are trying to answer. As forensic DNA techniques have developed over time,
their ability to detect smaller and smaller amounts of DNA has increased. This has brought justice

INVESTIGATION
to the perpetrators of unsolved crimes, but it also raises the risk of wrongful acquittals and convictions
if appropriate safeguards are not in place.

ARREST
02 Where can we detect DNA? 16
 ur DNA is everywhere. We’re constantly shedding it, passing it to other people, and moving
O
it around. This means that sometimes DNA detected at a crime scene has nothing to do with

and we don’t always get a match

Samples aren’t always perfect
the crime. Because of this, investigators need to consider when and how DNA might have

15 17 6 7 14 18 19 22 15 15

15 15
been deposited onto a surface or object.

03 Context is key 19

7 14 18 19
DNA doesn’t solve crimes in isolation. DNA profiling is an effective investigative tool
to be used within the wider context of all other evidence in a case.

MATCH?
MATCH!
04 What are DNA databases for? 23

15
Matching DNA profiles from crime scene material with those stored in DNA databases

30 SECONDS LATER,

RESULTS COME IN OVER

NEXT FEW HOURS/DAYS
has been one of the most significant innovations in crime fighting in recent history, providing

RESULTS ARRIVE!

Results aren’t instantaneous

and resources aren’t infinite
vital intelligence and saving police forces time and money. However, the use of DNA databases
has also raised concerns about privacy, data security, and fairness.

05 The meaning of a match 27

Not all DNA matches are equally informative. Just because DNA from a crime scene matches
a suspect’s DNA, this doesn’t necessarily mean they contributed it. Crime stain DNA is often

CRIME FICTION vs REALITY

missing some of the markers needed to generate a full DNA profile; in such cases several people
may be a ‘match’, but none may be the contributor. For this reason forensic scientists often employ

LAB

LAB
statistics to convey the meaning of the strength of the evidence.

One person does not do everything!

06 Predicting appearance and biogeographic ancestry from DNA 30
The latest advances in forensic genetics enable externally visible characteristics such as hair
or eye colour to be predicted from someone’s DNA. This could be a powerful investigative tool,
but the possibilities of what is currently achievable have sometimes been exaggerated.

THE MEANING OF A MATCH

07 Delving deeper 39

CRIME SCENE

CRIME SCENE
More information and sources.

If you see a term that’s green and bold it’s defined in our Useful Terms to Know on page 37

If you see a term that’s blue and bold it’s defined in our Techniques on page 12 CRIME FICTION REALITY

9
8 CONTENTS
01 WHAT CAN WE DETECT?

(see section 6 for more details).

to improve ability to deal
to narrow down suspect

This is still under development

than one person’s DNA.

to provide investigative leads.

Very small invisible

to predict eye and hair colour

with mixtures of more

Forensic DNA phenotyping

traces, typically

Age predictive tests

under development,

and rarely used in forensic

touch traces

development, aims
sequencing under

Starting to be used
Next-generation

investigations currently
is beginning to be used
pool by age.
DNA is a molecule that contains genetic instructions.

DEVELOPMENT OF FORENSIC DNA TECHNIQUES THROUGH TIME...

It is a major factor in determining the way we look — although 99.9% of our DNA

2008
is identical to that of other humans. It is the remaining 0.1% that marks us out as
individuals, and is therefore of primary interest to forensic geneticists. They can use
it to generate a DNA profile from human biological material at a crime scene. This

African, Eurasian
ancestry testing
can be compared with reference DNA from a named suspect, and a probability that

Biogeographic

and East Asian

limited to large
gives ancestry

database to match a DNA sample from

the suspect contributed it can be calculated.

information

geographic

individuals.
regions eg

the crime scene to relatives, leading

investigators to potential suspects.
The same DNA sequence is present in every cell of your body (apart from red

2005

Familial searching uses a DNA

blood cells), and because you’re constantly shedding cells into your environment,

In limited use
this means you leave a trail of DNA behind you. DNA is present in your house dust;
in the residue you leave on a glass; and in the root of the hairs stuck to your jumper.

Single Nucleotide
Everywhere you go, and everything you touch could contain traces of your DNA.

be used on very
Polymorphism

degraded DNA
eg old bones.
analysis can
Autosomal
Until around 2000, forensic geneticists would not have been able to generate
a DNA profile from such tiny samples of biological material. But as forensic DNA

2003
techniques have developed over time, their sensitivity — or their ability to detect

2001
smaller and smaller amounts of DNA — has increased. In the early days, you would
have needed a reasonably fresh sample of blood or semen about the size of a
British 5 pence piece or European 1 cent coin to generate a DNA profile6; today

2000
Low template analysis is the
most sensitive technique

sample of DNA.
so can be used on the smallest
a profile can be generated from just 50 picograms of DNA (the amount contained See diagram p11

assault cases to give information

in cells in more copies than other
types of DNA so analysing it can
in roughly 8 human cells). Such traces are invisible to the naked eye.

Mitochondrial DNA is present

have been damaged by heat,

be useful in mass disasters

about the male component

(male only) useful in sexual
situations where DNA may
(eg a plane crash) or other
DNA can be detected and analysed using a number of different forensic techniques,

Y chromosome analysis
each of which target different parts of DNA. Some, such as STR profiling (the most

Small invisible
common sort of DNA profiling — see the techniques table), target the nuclear DNA

light and humidity.
traces
in our chromosomes; others target the small circles of DNA found in cellular energy

of a sample.
factories called mitochondria.
The ease with which DNA profiles can be extracted from different body tissues

1996

Widely used
also varies. It is relatively easy to generate a DNA profile from blood, saliva and
semen, but extracting DNA from touched objects when often only a small number
of skin cells are present is more challenging. Full STR profiles can be generated

database 
First DNA
1995

established.
from hairs, but only if they contain a root (which has intact cells attached to it). If no

1991
Autosomal Short
Tandem Repeat profiling
can generate a DNA
profile in a few hours
compared to a few

used technique currently).
days (most commonly
root is present — as is often the case with hairs recovered from crime scenes —

traces such as small blood

Small (mm size) visible
generating a full STR profile can be difficult, but you might still be able to generate

stain on the carpet

a mitochondrial DNA profile. Mitochondrial DNA is less efficient at identifying
individuals, but can still provide very useful evidence eg in the identification
of burned or badly decomposed human remains. It can also provide useful evidence
to eliminate someone from an inquiry if a mismatch is found.
WHAT CAN WE DETECT?

WHAT CAN WE DETECT?

1985
DNA fingerprinting
Colin Pitchfork
case — the first
conviction that relied
on DNA evidence
(sentenced in 1988).
DNA can now be analysed

No longer in use
Large (2cm)
blood stain

blood groups
and proteins.
Analysis of
6
Silverman M, Thompson T (2014) Written in blood. 1st ed. Bantam Press, p297

1900

11
10 WHAT CAN WE DETECT? Size of DNA sample Technique
TECHNIQUES WHAT CAN WE DETECT?

DNA profiling Familial searching

The method by which patterns within an Involves searching a DNA database for profiles Because so much of our DNA is virtually identical, forensic scientists don’t analyse
individual’s DNA can be used to create a DNA that partly match the crime scene DNA profile. all of it (this would also be very expensive). Instead, they usually concentrate
profile. This is often visualised as a graph with If a database profile matches to more markers than on short, highly variable regions of repetitive DNA called short tandem repeats
peaks (see page 14). The DNA profile from a crime is expected by chance, it may belong to a relative or STRs. These differ in length between individuals and can be used as genetic
scene is compared with one from a suspect, of the suspect. This technique can therefore markers to generate a DNA profile that is extremely rare in a population
and the strength of evidence supporting the generate leads when a full match can’t be found. of unrelated individuals).
identification can be calculated. DNA profiling
Typically, markers are examined at a minimum of 16 locations plus a sex marker,
methods vary according to the type and quantity of Low template DNA analysis
DNA available, and the question investigators are or loci, in an individual’s DNA. These are visualised as a series of peaks on a graph,
A number of techniques (eg low copy number the position of which corresponds to the length of the STR, and is recorded as a
trying to answer. analysis) used to produce a DNA profile from number. Each location has two STRs (because we inherit one version from our
very small quantities of crime scene DNA. See diagram p14
STR profiling mother and one from our father), which means that an individual’s genetic profile
The standard STR profiling technique is modified can be represented as a series of digits, eg 12/13, 13/15, 9/9, 5/8 — a bit like a
The most commonly applied method of DNA to enable forensic scientists to look at more lottery number. Each pair of digits always corresponds to a specific location on a
profiling that makes use of highly repetitive of the DNA present. chromosome.
regions of DNA called short tandem repeats found
throughout human DNA. Biogeographic ancestry testing Once a DNA profile has been generated from a crime scene sample, it can then
A technique that enables an individual’s broad be compared to other profiles, such as DNA from other crime scenes (eg taken
Y chromosome analysis geographic origins (eg Africa, Western Eurasia, with a swab from a biological trace), the suspect’s DNA (eg taken at the police
Makes use of genetic information on the Y East Asia, South Asia) to be estimated based station with a swab), the victim’s DNA, or to DNA profiles within a national DNA
chromosome (which only males have). It is on genetic differences in their DNA. This method database. If the two profiles are identical, this is called a full match; if parts of the
especially useful in cases of sexual assault where uses DNA markers that are more or less common profiles match, this is a partial match. Once a match has been declared, the strength
male and female DNA are mixed, as only the male in different parts of the world, and can help narrow of evidence supporting the identification of a named individual can be calculated.
pattern will show up during analysis. The Y down a pool of suspects when no match in a So what is the chance that your DNA will match that of someone else? It depends
chromosome is inherited from father to son only, national DNA database has been found. on how many locations in the DNA (loci) you look at. If a forensic scientist looked
which means that all male relatives on the paternal at just one locus , the probability of this matching the same marker in another
side of the family will normally share the same Y Forensic DNA phenotyping individual would be relatively high (between 1 in 20 and 1 in 100).
chromosome. Uses DNA to make predictions about someone’s
But as more loci are examined, the probability of two individuals having an exact
appearance (eg hair colour, eye colour). It is
Mitochondrial DNA analysis match decreases rapidly. It’s similar to entering a lottery. Many people who buy
another way of narrowing down a pool of suspects
a ticket will match one number, but the chances of matching all of them and actually
Uses DNA from tiny energy factories found inside when no match in a national DNA database has
winning the lottery is very, very small.
all human cells called mitochondria. Mitochondrial been found. This technique uses DNA markers
DNA (mtDNA) is more abundant than other found in the genes that determine aspects Since European police forces today typically analyse STRs at 16 or more loci,
types of DNA, and can be useful in cases where of human appearance. As it is a new technique, the probability that two full DNA profiles match by chance is miniscule — in the
biological material is limited (eg if cells have been it has so far only been used in a very small region of 1 in 10 with 16 zeros after it (or 1 in 100 million billion)7. This means DNA
damaged by environmental exposure, such as heat, number of cases. profiling can be a very powerful tool in forensic investigations. Although in the UK
light or water, which break up the DNA strand). court, the statistics are always capped at 1 in a billion8.
MtDNA is inherited by a child from its mother, so Next-generation sequencing
all relatives on the maternal line of the family will Describes a suite of emerging DNA sequencing
share the same mtDNA. technologies, where sensitive tests can be done
simultaneously — i.e. you can do STR profiling,
SNP analysis biogeographic ancestry testing and phenotyping
Involves the detection of tiny variations in DNA tests at the same time. Their use in forensic
called single nucleotide polymorphisms. Because science is still in its infancy, but in future they 7
W
 elch LA, Gill P, Phillips C, Ansell R, Morling N, Parson W, Palo JU, Bastisch I (2012) European Network
SNPs are smaller and more abundant in each cell should enable more information to be obtained of Forensic Science Institutes (ENFSI): Evaluation of new commercial STR multiplexes that include
than STRs, SNP analysis can be useful when DNA from forensic DNA samples, eg making it easier the European Standard Set (ESS) of markers. Forensic Science International: Genetics 6(6): 819-26
8
is highly degraded. to distinguish between individual contributors E
 vett I, Pope S, Puch-Solis R (2016) Providing scientific guidance on DNA to the judiciary. Science
and Justice 56: 278-81
in a mixed sample of DNA.

12 WHAT CAN WE DETECT? 13 WHAT CAN WE DETECT?

WE’VE GOT DNA! NOW WHAT...
The Green River Killings
Swab used to take DNA sample
from crime scene to laboratory 1
(sometimes the DNA is degraded).
The development of forensic genetics has been a game The more people that
changer for certain criminal investigations, including the hunt appear to be in the
for one of the most prolific serial killers of all time. Throughout
the Eighties and Nineties, the bodies of numerous girls
mixture, the less sure you
and women were found dumped in overgrown and forested can be about the actual
areas near the Green River in Washington State, USA. All number of contributors.
had been raped and strangled, but despite the presence of For example, about 40%
chewing gum and cigarette butts at many of the dump sites, of mixtures from five
and even traces of semen on some of the victims, it wasn’t
until 2003 that their killer, Gary Ridgway, was finally caught.
people actually look like
a three person mixture,
Mr Ridgway first came under suspicion for the killings twenty
and virtually none would
Forensic scientist analyses years earlier, but police had no physical evidence against
2 sample in the lab to read him. In 1987, with more bodies appearing, police took hair show a definite indication
the pattern of DNA. and saliva samples from Mr Ridgway, but still couldn’t directly of 5 contributors because
connect him to the victims. It wasn’t until 2003 that more people share many of the
sensitive forensic DNA tests finally matched Mr Ridgway same markers.
to semen found on his earliest victims — women he killed
in 1982 and 19839. No-one knows precisely how many Denise Syndercombe Court
women he killed, but he was convicted of 48 murders, Reader in Forensic Genetics,
The DNA pattern is visualised as a graph and is now in prison for life. King’s College London
known as a DNA profile. If the sample 3 EUROFORGEN Consortium
was degraded it will give a partial profile. 9
 ttp://forensicoutreach.com/library/how-dna-changed-
h member
the-course-of-the-green-river-killer-investigation/
Full DNA profile Partial DNA profile

4
This DNA profile from the crime scene is then compared with another Partial profiles In an ideal situation you’d have enough DNA to generate a full DNA
DNA profile. Eg a mouth swab taken from the suspect or victim or it may profile (eg using at least 16 loci). However, this isn’t always possible.
be compared with DNA profiles held on the national DNA database. If DNA is only recovered in small amounts, or has been degraded
by temperature, moisture, or something else, some markers may
be missing. This leaves only partial DNA profiles. The reduced number
of markers makes it more difficult to distinguish between individuals,
and the chance of a partial profile matching another DNA profile
If the two DNA profiles have exactly the If parts of the two DNA profiles have the same
same pattern of DNA, this is a full match. A B pattern of DNA this is a partial match. 15 17 6 7 14 18 19 22 15 15 is much higher.
15 17 6 7 14 18 19 22 15 15
Full profile In addition, the world is a messy place, and DNA is rarely deposited
in neat packages from a single person. If a crime scene sample contains
the DNA from two or more individuals, then it is referred to as mixed
DNA profile. Because DNA gets everywhere, all crime scene DNA
samples are potential mixtures. This isn’t a problem, unless the DNA
15 17 6 7 14 18 19 22 15 15 15 17 6 7 14 18 19 22 15 15 15 17 6 7 14 18 19 22 15 15 15 7 14 18 19 15 15 15 7 14 18 19 15 15 you’re trying to analyse is present at such low levels that it becomes
15 7 14 18 19 15 15 confused with this background DNA, or with DNA from another
Full match Partial match contributor (eg a victim). In such situations, modern computerised
Partial profile
methods enable the strength of the evidence to be calculated —
something which should be communicated to investigators and juries.
N.B. This representation of a DNA profile only show 5 pairs of numbers/peaks (genetic markers),
but full DNA profiles would actually show at least 16.

14 WHAT CAN WE DETECT? 15 SECTION

WHAT CANHEADING
WE DETECT?
02 WHERE CAN WE DETECT DNA?

Each colour represents

a different person’s DNA
DNA gets everywhere. Besides the more obvious methods of DNA
transfer, including drops of blood or deposits of semen, small
amounts of DNA can also find their way onto people, places and
objects via droplets of saliva from talking, sneezing, skin cells shed
into house dust or by being left on the surfaces that people touch.
Given how easily DNA can be transferred, this means your DNA could be in a room
even if you weren’t. If your DNA is found at a location, it could be present because:

it’s so important for forensic experts to investigate a crime scene as soon as possible so that important DNA
(a) You have been there;
(b) You touched an object that was later carried to the location by someone See diagram p17

The DNA that is present in a room changes rapidly as people and objects interact within it, which is why
So DNA from individuals who have nothing to do with a crime might be present at a crime scene. DNA
else (eg an item of your clothing);

will last almost indefinitely untouched in a dry, cool environment, although activities such as cleaning
(c) You encountered a person, who soon after touches something at the
location, inadvertently leaving your DNA there (eg you shook hands
with them or you both previously touched the same surface).

2 Reality
degrade it. DNA deposited before the crime so unrelated to it is called background DNA.
DNA is transferred by body fluids, shedding skin cells, talking, coughing and sneezing.
As the sensitivity of forensic DNA tests
increases, so do the problems…
We’ve noted that when forensic DNA analysis was When there’s a serious crime,
first invented in the Eighties, a fairly large sample with very little evidence
of biological material was needed to generate a DNA
profile. But as forensic techniques have improved, their available to indicate the
ability to detect smaller and smaller amounts of DNA perpetrator, investigators may
has increased. This means that tiny, invisible traces resort to sampling areas that
of DNA can now be recovered and analysed. could have been touched
Undoubtedly, this has been beneficial in forensic by a potential suspect, such

does not get moved and new DNA is not introduced.

investigations, enabling difficult cases to be resolved, as a door handle or table
and the perpetrators brought to justice. However, it is surface. DNA profiles will
also creating problems. Such is the power of DNA
that some crime scene investigators have taken certainly be found, but it won’t

There is even DNA present in house dust.

DNA IS EVERYWHERE
to speculatively swabbing crime scenes, targeting be known if any of these
areas they suspect a perpetrator might have had direct belong to the perpetrator, so
contact with, and looking for DNA even when there’s this could result in false leads.
no visible stain. This may result in useful investigative

WHERE CAN WE DETECT DNA?

leads but it can also result in the detection of DNA Peter Gill
that’s irrelevant to an investigation, such as a DNA Professor of Forensic
profile from someone who has never visited the crime Genetics, University of Oslo
scene and/or has no connection to the crime. Like EUROFORGEN Consortium
any other analytic tool, if used without discrimination, member

1 Expectation
DNA profiling can be costly and distracting.

17
16 WHERE CAN WE DETECT DNA?
WHERE CAN WE DETECT DNA?
03 CONTEXT IS KEY

As forensic DNA techniques become more and more sensitive, there’s an increased The presence of DNA doesn’t necessarily tell us when or how
chance that DNA recovered from a crime scene is actually; it got there. But it can still be a phenomenally useful tool to police
(a)  ackground DNA: deposited before the crime took place and unrelated
B investigations. What matters is context.
to it (see page 17);
(b)  econdary transfer DNA: DNA from someone who was never there, picked
S Some types of DNA evidence are less likely to have been deposited through
up from them by contact and then left at the crime scene by another individual. innocent means than others. For example, a visible blood stain is not so easily
transferred unnoticed as an invisible smear of saliva, or a smattering of skin cells.
(c) The result of contamination by an investigator after the crime took place.
Latex gloves can carry DNA and accidentally transfer it between items and Questions such as: ‘When and how was the DNA deposited on to the surfaces
locations if an investigator forgets to change them. Other examination tools, tested?’ and: ‘How was the DNA collected by crime scene investigators?’
such as fingerprinting brushes, can also inadvertently transfer DNA between are crucial to understanding whether DNA is relevant to an investigation;
surfaces. Accidental contamination can also occur within a forensic laboratory or if it’s background DNA; the result of secondary transfer; or contamination.
(as with the Adam Scott case, see page 7). Additional context may also be provided by other, non-DNA evidence, such
as fibres, footwear marks, or fingerprints.
Shedder Status

One way in which we release DNA into our environment is through the constant shedding of skin Activity and context
cells onto our clothes and the surfaces we touch. But not everyone does this at the same rate. If a large, visible, sample of body fluid, such as a blood stain, is found at a crime
People who shed lots of skin cells — possibly because of a skin condition such as eczema, scene, then it is easy to obtain DNA from it. And if that blood is found on smashed
dermatitis, dandruff, or even sunburn — are known as ‘high status shedders’ and are more likely window glass, say, then there is a good chance it is relevant to the investigation. The
to deposit DNA. For instance, a recent study found that people with atopic dermatitis shed four suspect may have cut himself while breaking into the property, for example. In other
times as much DNA as healthy individuals10. Conversely, a ‘low status shedder’ is less likely words, the evidence is considered relevant because it is directly associated with See diagram p20
to deposit DNA. But not everyone does this at the same rate and it will also vary within the same the activity of the crime — in this case breaking a window. Contrast this with a DNA
person at different times. profile taken from a surface, such as a kitchen table, where there’s no identifiable
body fluid. It is more difficult to propose an activity that could explain its presence
because of the lack of context.
Another important change that has come with advances in forensic DNA techniques
is that when invisible samples are used to generate a DNA profile, there’s usually So… DNA alone doesn’t solve crimes. It’s an important detection tool, but it’s
no information about which body tissue it came from, or when the DNA was deposited. certainly not a detective.
When larger amounts of DNA were needed to generate a DNA profile, it was usually
possible to run other tests to determine whether a fluid was saliva or semen, say. And
in the case of blood, its colour could provide clues about how fresh it was. These tests
can’t be done on really small biological samples, but DNA can often still be detected.
Other types of forensic evidence can provide important corroboration
WHEN COULD DNA BE DEPOSITED? of DNA results, and vice versa. As an example: fibre analysis depends on having
a reference sample from the textile source. If fibres have been recovered from
Before the crime: During the crime: After the crime: a victim not matching the victim’s clothes, there is not much that can be done.
Background DNA Perpetrator’s DNA Potential contamination However, these fibres may suddenly become informative if a suspect has been
Secondary Transfer
identified by DNA profiling, and clothes are found in the suspect’s apartment
Time matching the fibres from the victim. Then you have corroborative evidence.

Investigators Peter Schneider Professor of Forensic Molecular Genetics, Institute of Legal Medicine,
arrive and Lab Analysis University of Cologne
Crime Discovery detect DNA analysis complete EUROFORGEN Consortium member

10
Kamphausen T et al (2012) International Journal of Legal Medicine 126(1): 179-83

18 WHERE CAN WE DETECT DNA? 19 CONTEXT IS KEY

 When local taxi driver David Butler’s DNA was recovered from the fingernails of murdered
DNA is found on the victim’s hands. You were also at the party, and had no direct contact

was tiny, but enough to generate a hit against the UK’s DNA database (see next chapter),
DNA on the knife blade. Methods used to collect and store the evidence were also found

of secondary DNA transfer — or they had been transferred to her through other innocent
However, you knew the victim, and often visited him, so your DNA could have been there

However, Mr Butler’s defence team queried precisely how that DNA got onto the victim’s
In the final judgement, the court accepted the defence version of events and exonerated

of the dry skin condition he suffered from, and suggested that perhaps some of his skin
to be sub-standard, so the DNA could also have got there through cross-contamination.

sex worker Anne Marie Foy, it seemed like an open and shut case. It was presumed that

nails. They established that he was sometimes known by the nickname “flaky”, because
park near Liverpool city centre in September 2005. The amount of DNA found by police
A man is attacked by someone wearing a mask, while walking home from a party. Your

with the victim, but you both poured a glass of wine from the same bottle. This caused

though other evidence was lacking, the DNA evidence was enough to see him charged
she co-habited with Ms Kercher, there was a ready explanation for the presence of her
contained small traces of Ms Kercher’s DNA on the blade, and a DNA profile from Ms

of circumstances, while ignoring other possibilities. For instance, Ms Knox could have
used the knife to cut bread (starch grains were also observed on the blade), and since
transferred to the knife when Ms Kercher was stabbed with it, although no blood was

Ms Foy had torn at his skin as he hit and strangled her, before dumping her body in a
detected. Possibly, this was because the knife was cleaned with bleach, they argued.

cells had transferred to bank notes that were later used to pay Ms Foy — an example
Knox was recovered from the handle. Prosecutors suggested that that the DNA was
Meredith Kercher was stabbed to death in Perugia11, Italy, in 2007, and her flat-mate,
 man is found dead at his home, and your DNA is recovered from the crime scene.

a psychological effect where an individual(s) fits the evidence to a presupposed set

Amanda Knox, was a key suspect. A knife retrieved from Ms Knox’s boyfriend’s flat

This is a classic example of confirmation bias, (see Adam Scott case, see page 7),

identifying Mr Butler as the source. He denied ever having met the victim, but even
your DNA to be transferred to their hands. This is known as secondary transfer.
for weeks before he was killed. This is known as background DNA.

 ill P (2016) Analysis and implications of the miscarriages of justice of Amanda Knox and Raffaele Sollecito.

http://www.telegraph.co.uk/news/science/9115916/The-case-against-DNA.html (accessed January 2017)

means12 . Mr Butler was acquitted.

Forensic Science International: Genetics 23: 9-18

the defendants.

with murder.
Secondary transfer?
Background DNA?

SECTION HEADING
IS KEY
A

CONTEXT
Scenario

Scenario
example

example
Real-life

Real-life

21
12
11
WHO ARE WE DISCOVERING AT THE SCENE? HOW DID MY DNA GET THERE?
DNA is everywhere and context is everything. There have been major developments in the techniques for identifying individuals
from samples of DNA. Since the 1990s tests to detect invisible DNA have been developed. Yet, this ability to measure smaller microgram = one millionth of a gram
and smaller samples has come with problems. With invisible traces it is not always possible to know whether the DNA came nanogram = one thousand millionth
from a body fluid, skin or hair. There is also no time stamp on DNA whereas a visible stain eg blood might give some indication of a gram
of when it appeared. So, as techniques have improved, the importance of evaluating DNA in context of other information picogram = one thousand billionth
and evidence has also increased. of a gram
one cell = 6 picograms
Sample size
A large visible trace; BLOOD STAIN A small visible trace; a tiny STAIN A small invisible trace (could be from A very small invisible trace; a ‘touch’
(2cm) was found at the scene. (millimetre size) was found at the scene. body fluids or skin flakes), was found trace was found at the scene.
at the scene.
A large amount of DNA was obtained A small amount of DNA was obtained A very small amount of DNA was
from the stain (some micrograms). (some nanograms) from the stain. A small amount of DNA was obtained obtained (less than a hundred picograms)
(several hundred picograms) from from the trace.
A DNA profile was generated and a match A DNA profile was generated and a match the trace.
with your DNA was found! with your DNA was found! A partial DNA profile was generated
A DNA profile was generated and a match and a match with your DNA was found!
The type of body tissue Using a screening test it is
with your DNA was found!
(blood) is known and a very general possible to identify the type of body The type of body tissue cannot
idea of how old the blood stain can tissue from the stain eg blood, semen The type of body tissue cannot be determined without consuming
be estimated (range: hours vs days). or saliva without ‘consuming’ the trace. be determined without consuming the the trace (using the entire sample)
This means some of the DNA remains trace (using the entire sample) in the test in the test because there are currently
It’s highly likely you were
intact for future tests. because there are currently no screening no screening tests for invisible traces.
there at the time, or recently.
tests for invisible traces.
Highly likely you had been Were you there or not?...
You were at the premises
to the premises where the crime was Were you there or not?... Unsure. An invisible trace could have been
— you cut yourself on broken glass.
committed (but not necessarily recently). Unsure. An invisible trace could have been left during the crime, before or after or you
left during the crime, before or after, or you may have never been at the crime scene.
You’d had a nose bleed
may have never been at the crime scene.
in the morning, then went to the premises The victim touched the same
in the afternoon and sneezed. The crime You had drunk from a glass glass as you at a pub before he/she was
was committed in the evening. at the premises, but it could have attacked (secondary transfer).
been days or even months before
(background DNA).
Importance of context
20 CONTEXT IS KEY
 04 WHAT ARE DNA DATABASES FOR?
Contamination?
Scenario As a security guard, you had legitimate access to a building, before a stabbing took place
there. The police investigator wore plastic gloves to recover a knife from the crime scene, Once a DNA profile has been generated from crime scene material,
but first touched a door handle that you had also previously touched. As a result, your the next step is to compare it to DNA profiles of known individuals
DNA was accidentally transferred to the knife. This is known as contamination.
in order to find a match — eg to a known suspect. Or in many cases,
where no suspect has been identified, this will involve searching a
Real-life Farah Jama, was wrongfully convicted of raping a woman in Melbourne, Australia, DNA database.
example in 2008 and spent 15 months in prison after a sample of his DNA had contaminated
a sample taken from the alleged rape victim13. The mistake is believed to have occurred Most European countries have their own national DNA database. The
because 28 hours earlier, the same forensic medical officer had taken a DNA swab from circumstances under which an individual’s DNA can be taken, and whose
a woman with whom Mr Jama had had sex (no charges were made)14. The precise DNA profiles can be retained on such databases vary widely between different
mechanism of contamination is unknown, but as the two samples were not ‘mixed up’, countries. In many places, samples can be taken on arrest, but DNA profiles can
it is most likely that the examination room or the equipment used were not cleaned. only be retained for a short period of time - unless the person is convicted of a
serious crime. In other countries, DNA profiles can be retained from individuals
convicted of any offence. National DNA databases also hold DNA profiles retrieved
from crime scenes, in case someone whose DNA isn’t currently on the database
is arrested in the future. This also enables investigators to link crimes which may
have been committed by the same person.
Although many national DNA databases are large, they don’t contain DNA from
everyone living in a country. This means that even if DNA is retrieved from a crime
scene, unless the perpetrator’s DNA is already in the database, it won’t necessarily
generate a ‘hit’ — or match.

The UK National DNA Database by Numbers 15

The UK national DNA database (NDNAD) was launched in 1995 and holds 5 million DNA profiles
from individuals, (plus about half a million crime scene profiles) — equivalent to nine percent
of the UK population — (or 14% of the male population since 80% of profiles are from men)15.

About 40 vetted home office officials have access to the database. Police forces do not have
access, although they own the records on the database and receive notifications of any matches.
The database can only be used to prevent and investigate crime, to prosecute those accused
and to identify deceased persons16.

The UK NDNAD generates matches or hits for more than 32,000 crimes per year.

In 2014/15, the chance of a crime scene profile matching a subject on the database was 63.2%.

Even so, the likelihood of crime scene DNA being retrieved and sent for testing is low,
and forensic analysis tends to be reserved for serious cases eg homicides. For example,
in 2014/15, crime scene investigators were sent to look for forensic evidence in 96% of homicide
crime scenes, and found DNA at 65% of those examined, whereas they were only sent to 27%
of vehicle thefts, and just 30% of these yielded DNA evidence.

13
I nquiry Into the Circumstances that led to the Conviction of Mr Farah Abdulkadir Jama 15
http://www.parliament.vic.gov.au/papers/govpub/VPARL2006-10No301.pdf  ational DNA Database Strategy Board Annual Report 2014/15 https://www.gov.uk/government/uploads/system/uploads/
N
14 attachment_data/file/484937/52921_NPCC_National_DNA_Database_web_pdf.pdf
http://www.smh.com.au/national/dna-fiasco-rape-conviction-quashed-20091207-kfc3.html 16
(accessed January 2017)  ffice for National Statistics: United Kingdom population mid-year estimate https://www.ons.gov.uk/
O
peoplepopulationandcommunity/populationandmigration/populationestimates/timeseries/ukpop/pop (Accessed January 2017)

22 CONTEXT IS KEY 23 WHAT ARE DNA DATABASES FOR?

WHAT ARE DNA DATABASES FOR? WHAT ARE DNA DATABASES FOR?

Social and ethical aspects of DNA databases Match vs identity

There are clear benefits to maintaining a forensic DNA database and some people Suppose a DNA profile from a crime scene is compared against a national
think they should hold information from everyone in a country. But others have DNA database. If a match occurs, it’s either because that person’s DNA was
concerns about privacy, data security, and fairness. Questions about whose DNA found at the crime scene, there was contamination, or it’s a false positive match
profiles should be retained on the DNA database; for how long; for what purposes; (a chance match with an individual who was not involved in the crime). The next
and who should have access to, and oversight of national DNA databases, continue step is to calculate the strength of the DNA evidence to support the proposition
to be debated in many countries. These questions grow more pertinent as interest that a specified named individual has contributed to the sample.
in the sharing of information across national borders — and even across different
Although the chances of two full DNA profiles from two different unrelated
types of databases (eg medical and commercial information) — increases, because
individuals matching are extremely small, DNA profiles from crime scenes
each database has different security controls, access policies and retention periods.
are rarely perfect, and they may not contain information about every genetic
There are also concerns that certain minority groups are disproportionately marker analysed (i.e. they are a partial DNA profile). The smaller the number
represented on national DNA databases. Some argue that such inequalities of genetic markers the DNA profile is composed of, the greater the risk of a false
could stoke feelings that certain groups are being unfairly criminalised match occurring.
or discriminated against17,18.
In 1999, Raymond Easton, a 49-year-old man from Swindon, UK, was arrested
and charged with a burglary in Bolton, UK (~175 miles away), after a DNA sample
The right to privacy from the crime scene matched his DNA profile in the UK national DNA database.
Mr Easton was in the advanced stages of Parkinson’s disease, and was unable
to walk more than ten metres without help. His DNA profile had been loaded
Should governments have the right to hold the DNA of people without a conviction indefinitely?
onto the database four years earlier following a domestic dispute. The crime
Some people have argued that if everyone in a country were held on a national DNA database,
scene sample matched Mr Easton’s DNA profile at six loci, which was considered
far more crimes would be solved, but many are opposed to this idea on the grounds of personal
enough to secure an identification at that time — although the total number
privacy and human dignity. Instead, many countries only keep DNA profiles on the national DNA
of STRs required has since been extended to 16 plus a sex marker. The chances
database from people who have been convicted, or who choose to give their DNA voluntarily.
of a match was reported as 37-million-to-one 21. Mr Easton spent several months
So what if you’re arrested, but not guilty of the crime? In 2001, an 11-year old boy, known as S., in custody before his solicitor persuaded police to run further DNA tests, which
and another British man called Mr Marper, were arrested for separate offences, but neither was eliminated him22 .
charged. Both had DNA swabs taken from the inside of their cheeks and their samples were
Importantly, a match does not imply identity, something we discuss further
added to the UK national DNA database (NDNAD). After their release, they applied to have their
in chapter 5. Even with a probability as low as one-in-37-million23 , you could find
DNA profiles (and fingerprints) removed from police databases including the NDNAD, but these
one or more innocent British men in the UK population who would match the crime
applications were rejected by British Appeal Courts.
scene profile described above. The larger the database, the higher the risks of false
In 2008, the European Court of Human Rights asserted that the retention of fingerprints, cellular positive matches, which is why it is essential to consider the non-DNA evidence
samples and DNA in such circumstances was in breach of Article 8 of the European Convention as well. If this is not done there is a danger that the DNA evidence will be over-
on Human Rights — the right to respect for privacy and family life19. As a result, there has been weighted by a jury.
a change to UK law, meaning that people who have been arrested for, but not charged with,
The Raymond Easton case emphasises the need to regard DNA evidence
imprisonable offences (and certain non-prisonable ones), must normally have their DNA profiles
as investigative in the first instance, and not a panacea for a prosecution. This
removed from the DNA database within 3 years20.
is not to say it can’t be valuable or form an important part of a conviction, but the
context of the DNA evidence and whether it can be corroborated with other non-
DNA evidence, such as fibres, fingerprints, or eyewitness statements, matters
17
D
 uster T (2004) Selective Arrests, an Ever-Expanding DNA Forensic Database and the Specter of an Early-Twenty-First Century Equivalent and needs to be considered if these are available in a case.
of Phrenology in DNA and the Criminal Justice System: The Technology of Justice, ed. David Lazer (Cambridge M.A: MIT Press), p315-334
18
 olicy Equality Statement for the National DNA Database www.gov.uk/government/uploads/system/uploads/attachment_data/
P
file/245240/Policy_Equality_Statement_for_the_National_DNA_Database_Sep13.pdf 21
 ttp://www.heraldscotland.com/news/12440889.Guilty_by_a_handshake__Crime_scene_DNA_
h
19
h
 ttps://justice.org.uk/s-marper-v-uk-2008/ (accessed January 2017) tests_may_not_be_as_accurate_as_we_are_led_to_believe/
20
P
 rotection of Freedoms Act 2012: how DNA and fingerprint evidence is protected in law https://www.gov.uk/government/publications/ 22
Huff C, Killias M (2008) Wrongful conviction. 1st ed. Philadelphia: Temple University Press
protection-of-freedoms-act-2012-dna-and-fingerprint-provisions/protection-of-freedoms-act-2012-how-dna-and-fingerprint-evidence-is-
23
protected-in-law (accessed January 2017) Gill P (2014) Misleading DNA evidence: Reasons for Miscarriages of Justice. 1st ed. Academic Press

24 WHAT ARE DNA DATABASES FOR? 25 WHAT ARE DNA DATABASES FOR?
WHAT ARE DNA DATABASES FOR?
05 THE MEANING OF A MATCH

Familial searching DNA doesn’t give a simple ‘yes’ or ‘no’ answer.

If investigators can’t match a crime scene DNA profile to someone on a national Forensic geneticists typically use 16 or more genetic markers, selected from
DNA database, or to a suspect who has been identified through other means, the human genome, plus an indicator of sex, to generate an individual’s DNA
they may occasionally search for relatives of the person who deposited the DNA profile. These markers have been chosen because they are extremely variable,
by looking for similar DNA matches. This is based on the principle that unrelated so if you have a full DNA profile (with information available for all 16 markers)
people have relatively few DNA markers in common, whereas related people will the chances of finding another unrelated person with exactly the same DNA
have more — a parent and child, for instance, will match at least half of the genetic sequence at each location is very small. This means that the risk of DNA retrieved
markers analysed, as they share 50% of their autosomal DNA. As we’ve noted, from a crime scene matching someone unrelated to the true source is extremely
we have two copies of DNA at each locus, because we inherit one copy from each low (less than 1 in a billion, and often many orders of magnitude lower than this).
parent. So a parent and child always share one of the two digits (representing
an STR repeat of the same length) at each locus. However, many of the DNA profiles retrieved from crime scenes aren’t full DNA
profiles because they’re missing some genetic markers or there is a mixture
The more distantly related people are, the less similar their DNA is. This means of DNA from two or more people. So was it the suspect who left their DNA
with standard (autosomal) DNA profiling, familial searching is only useful for at the crime scene? The DNA evidence won’t give a ‘yes’ or ‘no’ answer: it can
finding close relatives (such as parent/child/sibling). It typically generates many only ever be expressed in terms of probability.
false leads, so investigators will often limit the size of the pool they search within,
eg by confining it to a specific geographic area, in order to produce a useful
shortlist to follow up24 . Investigators can then look at other evidence that might How certain?
imply a relative of an individual was involved. Familial searching is only used for
A DNA profile is said to ‘match’ if all of the markers in the crime stain profile are the
very serious crimes, and is allowed in some countries including the UK and the
same as the markers in the sample it’s being compared to (the reference sample).
Netherlands, but not, for example, Germany. In the UK, there were only 16 familial
Remember that this is a matching of samples, not a matching of a sample to a
searches conducted during 2014-15 25.
specific person. It is possible that several individuals in a database may ‘match’
Familial searching raises ethical questions, because it risks casting suspicion a crime scene DNA profile, when none of them actually deposited it.
on family members who were not involved in the offence. However, it can
Once a DNA sample has been analysed, two kinds of statistics may be reported.
sometimes provide an investigative lead where none existed before, and has led
The simplest is the match probability, which addresses the question of how
to successful prosecutions.
rare a DNA profile is in a population of random unrelated individuals. This must
not be confused (but often is) with how likely the person is to be innocent
Familial searching in action: Finding justice for Lynette White of the crime. For example, if a DNA profile from the crime scene matches
the suspect’s DNA and the probability of such a match is 1 in 100 million if the
Lynette White was a 21-year-old prostitute who was stabbed to death on Valentine’s Day, 1988, DNA came from someone else, this does not mean that the chance of the suspect
in Cardiff, UK. Three local men, Stephen Miller, Tony Paris and Yusef Abdullahi, were wrongfully being innocent is 1 in 100 million. This serious misinterpretation is known as the
convicted of her murder and spent two years in prison, before their conviction was quashed. prosecutor’s fallacy.
The case was reopened in 2000, and thanks to newer, more sensitive methods of DNA analysis,
a DNA profile could be obtained from blood on a skirting board that had been found near These match probabilities are fine if you are dealing with full DNA profiles from
to her body. This was run against the national DNA database, but no full matches were found. a single individual and they can often be used for a partial DNA profile too. But this
However, there was a partial DNA match with a local 14-year-old boy, who was already known figure cannot be used for mixtures of two or more individuals or if information
to police. He was only 2 years old when Ms White was murdered, but was possibly related about some of the genetic markers is missing. Under these circumstances
to the killer. When police took DNA samples from other members of his family, the boy’s uncle, a second kind of analysis is applied: a likelihood ratio.
Jeffrey Gafoor, provided a full match. Soon afterwards Mr Gafoor confessed, and was sentenced
to life imprisonment 26 .

24
F
 amilial Searching, inferring ethnicity and research uses http://nuffieldbioethics.org/wp-content/uploads/
Bioinformation-Chapter-6-Familial-searching-inferring-ethnicity-and-research-uses.pdf
25
N
 ational DNA Database Strategy Board Annual Report 2014/15 https://www.gov.uk/government/uploads/
system/uploads/attachment_data/file/484937/52921_NPCC_National_DNA_Database_web_pdf.pdf
26
http://thejusticegap.com/2011/12/what-price-justice-the-cardiff-3/ (Accessed January 2017)

26 WHAT ARE DNA DATABASES FOR? 27 THE MEANING OF A MATCH

THE MEANING OF A MATCH THE MEANING OF A MATCH

A likelihood ratio weighs the evidence in favour of competing ‘stories’ Calculating complexity
(or hypotheses), one from the prosecution perspective and one from the defence.
It compares how probable the observed evidence is under each story. Because of the complexity of the calculations – particularly once several people’s
DNA gets mixed together — DNA analysts now use specially designed computer
So in the case of a DNA profile from a sample of blood, you would be looking at: programs. Just like the physical methods of analysing DNA, these programs need
a) Assuming the blood comes from the suspect, what is the probability to be validated to ensure they have been properly formulated and use theory that
of seeing the match? is widely accepted by the scientific community.
b) Assuming the blood was someone else’s, what is the probability of seeing A number of such programs are in use around the world, but because different
the match? programs are prepared using different mathematical approaches and assumptions
about the data, they can give different values for the likelihood ratio. There
The likelihood ratio is obtained by dividing (a) by (b). If the answer is larger than one
have been cases where the prosecution has used one program and the defence
the prosecution’s version of events is better supported; if it’s lower than one then
another, meaning the court has been presented with two different answers about
the defendant’s is.
the strength of the DNA evidence. Usually the difference is only small, but cases
The guidance given alongside is usually: where one program favours the prosecution and another favours the defence are
obviously important and need further investigation.
Ratio Expert guidance: the forensic findings… It is important to note that several computer programs have been developed by
commercial companies, and may not be as available for use by defence experts
1 ... do not support one proposition over the other
because they have less money to pay for them. This could make it difficult
2-10 ... provide weak support to scrutinise how the prosecution generated its statistics. Since open source
software is freely available and is more transparent, there is currently a debate
10-100 ... provide moderate support within the forensic community on the relative benefits of commercial vs open
100-1,000 ... provide moderately strong support source software.

1,000-10,000 ... provide strong support

How reliable is forensic DNA profiling?
10,000-1 million ... provide very strong support
In recent years, confidence in other (non-DNA) forensic techniques, such as
Over 1 million ... provide extremely strong support 27 bite mark analysis or hair microscopy, has been damaged by the discovery
that different analysts can reach different conclusions about the same piece
Although statistical experts agree that likelihood ratios are the best approach of evidence. Some of these older forensic techniques were never properly
for complex DNA profiles, their adoption by many labs has proved slow. In part this validated, or subjected to rigorous scientific examination, before entering
is because of fears that courts may misunderstand the ratio. the criminal justice system. They involved a lot of subjectivity.

Remember: DNA doesn’t give ‘yes’ or ’no’ answers, but rather enables Ideally, analytical techniques used in forensic science should be validated before
us to assess probabilities. So DNA evidence can be very strong, very weak they’re allowed into court. For a method to be validated it must have a scientific
and everything in between. backing that is underpinned by peer reviewed papers in the scientific literature. An
important part of testing is to show that when a completely unconnected individual
is compared with the ‘crime stain’ DNA profile, the likelihood ratio obtained is well
below 1 and never very high. Any limitations of the test must also be made clear.
The most important feature distinguishing DNA profiling from other non-DNA
forensic identification techniques is that for DNA profiling we can base calculations
on the exceptionally well understood established theory of genetics. No such
underpinning theory is available for most non-DNA forensic techniques.

27
E
 NFSI Guideline for Evaluative Reporting in Forensic Science http://www.forensic-isotopes.org/assets/
ENFSI%20Guideline%20Evaluative%20Reporting%20March%202015.pdf

28 THE MEANING OF A MATCH 29 THE MEANING OF A MATCH

06 PREDICTING APPEARANCE AND BIOGEOGRAPHIC WHO TO PROFILE? FORENSIC DNA PHENOTYPIC
ANCESTRY FROM DNA TESTS TO NARROW DOWN SUSPECT POOL
Our DNA underpins many aspects of the way we look, but scientists If a DNA profile is obtained from a crime scene and there is no match on a national DNA database,
have only recently begun to get to grips with how individual genes forensic DNA phenotyping may be used to predict a suspect’s appearance, to narrow down from a
population to a pool of possible suspects for investigation. Tests can currently predict eg hair colour,
produce physical traits such as hair or eye colour.
sex, eye colour. But these tests aren’t 100% accurate, so they cannot (with the exception of sex) be used
A practical application of this work is the development of forensic DNA tests that to eliminate suspects. They are still in development, so are rarely used in forensic investigations currently.
can predict aspects of someone’s physical appearance. This approach is called
forensic DNA phenotyping28 .
But not all externally visible characteristics are equally predictable from DNA
Population of people
information — at least not today. The simplest is biological sex, because females
carry two X chromosomes, and males carry one X and one Y, which is easily
detectable by DNA testing. Eye colour is already harder, because it’s influenced
by many genes, of which six are currently used in forensic DNA phenotyping tests.
Predicting other externally visible traits such as height is even harder, and currently
not yet possible because they are determined by large numbers of genes, many
of which remain unknown. They are also influenced by environmental factors such Predicting sex
as nutrition, which cannot be predicted from DNA. Accuracy = close to 100%
Because human geneticists have only understood the genetic basis of a few = male
externally visible characteristics so far, forensic DNA phenotyping is still in its
infancy. In general, it’s only used in cases where a DNA profile from crime scene
material does not match a profile on the DNA database or any other known
suspect. Even then it is only used as an investigative tool, to reduce the number Predicting broad geographic ancestry:
of potential suspects when the suspect pool is very large and to help prioritise who Accuracy = 97-100% for African,
to focus on first or next. It is not used as final evidence in court. + Eurasian (including Europe, Middle East
and South Asian) and East Asian.
For instance, forensic DNA phenotyping might be used to narrow the pool = male with Eurasian ancestry
of suspects where investigators already have some idea about where a perpetrator
lives (in a cluster of villages, say). One way of doing this is through familial
searching (described on page 26) to identify potential relatives. If this doesn’t
Predicting hair colour
generate any leads, then forensic DNA phenotyping might be able to provide clues
(red/blond/brown/black)
about the physical appearance of the person they’re looking for. The ultimate goal
+ Accuracy = 80-90%
is to identify a group of individuals whose DNA could be sampled (by voluntary
screening) and analysed to generate a standard DNA profile, in order to find = male with Eurasian ancestry
a match with the DNA profile from the crime scene. and black hair

Predicting eye colour

(blue/intermediate/brown)
+ + + Accuracy = 75-90%
= male with Eurasian ancestry, 
black hair and blue eyes

Once a suspect pool has been narrowed down to this final group, first DNA samples are taken from each
of these individuals. Their DNA profiles are generated and then compared to the DNA profile obtained from the
28 crime scene to see if there is a match. If no match is found the next group up is tested and so on. These groups
 ayser M (2015) Forensic DNA Phenotyping: Predicting human appearance from crime scene material
K
for investigative purposes. Forensic Science International: Genetics 18: 33-48 enable investigators to prioritise which individuals in a population to screen. This is a much more practical
and cost effective than simply taking samples from the whole population (which is the alternative approach).

30 PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA 31 PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA
PREDICTING APPEARANCE AND BIOGEOGRAPHIC
ANCESTRY FROM DNA

Can DNA tell what the face of a suspected criminal looks like? The face of litter?
It is currently possible to predict eye and hair colour from a DNA sample — although
none of these tests are 100% accurate. Some of these tests have been forensically Behind every piece of litter is a person who dropped Some physical traits, such
validated (see page 31), and the results of these studies published in scientific See diagram p31 it. So when Hong Kong launched an initiative to tidy as skin colour, are closely
journals. Skin colour is likely to be the next appearance trait that forensic scientists up its streets, it tried to shame litterbugs by plastering
reconstructions of their faces on billboards across the associated with commonplace
will be able to predict from DNA — tests are currently being developed and validated.
city. The information for these facial reconstructions understandings of biogeographic
However, knowledge about the genetic basis of any other physical traits is not yet came from DNA, which had been retrieved from ancestry, ‘race’, or ‘ethnic identity’.
advanced enough for them to be predicted from a DNA sequence. In particular, discarded items such as chewing gum or cigarette
the genetics of human facial structure is highly complex, and the scientific studies Because of this, care needs
butts, and sent to a USA-based company for forensic
that have been published in this area have identified only a few out of the hundreds to be exercised in the application
DNA phenotyping.
or possibly thousands of genes that scientists expect to be involved — each with of forensic DNA phenotyping
a very small effect. It made predictions about their sex, hair, skin and eye so that police inquiries are
colour, freckling, biogeographic ancestry, and facial
This hasn’t stopped a USA-based company from marketing a service to reconstruct shape. Some of these things can be predicted from not seen as reinforcing racial
a face from DNA. The company has not yet published information about its DNA, but eg black hair is almost universal within stereotyping and perceptions
methods (in a peer reviewed journal), nor a validation study which is particularly the population of Hong Kong. However, facial shape of unequal treatment amongst
important in forensic science (see page 29). Yet some police forces have started cannot currently be predicted from DNA because
using these facial reconstruction tests, which has been covered uncritically minority communities.
it involves the complex interplay of very many genes
in the news. (see page 32). Robin Williams
Forensic DNA phenotyping raises some ethical issues too. The campaign conveyed the idea that DNA from
Whereas standard forensic DNA profiling involves genetic Professor of Forensic Science
rubbish could be used to reconstruct what the person Studies, Northumbria University,
markers found in parts of the human genome that are not who dropped it looks like; that they would be found Newcastle-upon-Tyne
within genes (non-coding regions), the markers used in forensic and prosecuted. The reality is quite different. EUROFORGEN Consortium
DNA phenotyping are located within or close to genes involved member
in the externally visible trait being predicted (coding regions).
If forensic DNA phenotyping techniques were extended Currently, eye colour, hair
to also include non-visible characteristics, such as genetic colour and skin colour
risk of disease, they could reveal personally sensitive, private can be predicted reliably
information, to whoever is doing the testing — which could
be of interest to medical insurance companies or certain and with practically useful
employers. This can be avoided through regulation, eg in the accuracy from crime
Netherlands, where externally visible characteristics are legally scene DNA, but not yet The idea that a person’s face is reconstructed from DNA traces alone,
allowed to be used in forensic DNA phenotyping, but non-visible any other externally visible and the result publicly displayed as a ‘photo-fit’ to aid police investigations
disease traits aren’t. is disconcerting. DNA analysis may be able to predict but cannot determine
characteristic.
When DNA is used to make predictions about what a suspect the actual likeness of a person. However, some may take such images at face
looks like, this will result in estimates of probability and error value. This could lead to endangering or stigmatising groups of people who may
for each predicted trait, meaning that the weight of this type Manfred Kayser
be considered to look similar to such DNA-generated images, even though they
of evidence can be assessed quite well. It’s important that Professor of Forensic Molecular
police understand the differences between standard forensic are not remotely connected to a crime, or may be innocent.
Biology, Erasmus MC University
DNA profiling — which can identify individual DNA profiles that Medical Centre, Rotterdam Matthias Wienroth Research Fellow in Social Science, King’s College London
match -- and forensic DNA phenotyping — which can’t identify EUROFORGEN Consortium
individuals yet, but provides information that allows individuals member EUROFORGEN Consortium member
to be placed into groups defined by specific visible traits
and biogeographic ancestry.

32 PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA 33 SECTION HEADING

PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA
PREDICTING APPEARANCE AND BIOGEOGRAPHIC PREDICTING APPEARANCE AND BIOGEOGRAPHIC
ANCESTRY FROM DNA ANCESTRY FROM DNA

Predicting biogeographic ancestry However, forensic DNA phenotyping has also been
A number of genetic markers (known as ancestry informative markers) have a game changer
been discovered that are much more common among people from some parts Eva Blanco case, 1997
of the world (eg Africans) compared to others (eg Europeans). These can be used
to predict an individual’s biogeographic ancestry, i.e. the broad geographic region Eva Blanco Puig was a 16-year-old Spanish high school student, who was raped
their biological ancestors originated from. For now, the available tests that are and murdered in Algete, near Madrid in 1997. Police made an application to take
suitable for crime sample analysis can only reliably predict to which of the major DNA samples from men in Algete, including relatives and acquaintances, in the
continental groups a person belongs i.e. African, Western Eurasian, East and South hope of identifying the killer, but the application was turned down. In 2015, semen
Asian, or Native American. They cannot say which country someone comes from. recovered from Ms Blanco’s body was subjected to forensic DNA phenotyping,
revealing that the perpetrator was likely of North African origin. So investigators
The terms ethnicity and race are sometimes used interchangeably with narrowed their search, focusing on men from this group who were living in the
biogeographic ancestry, but strictly, ethnicity reflects a person’s social and cultural area in 1997. They took DNA samples from 300 willing volunteers, including two
background, which cannot be detected from DNA. However a person’s ethnicity brothers who gave partial matches with conventional DNA profiling. This led them
may be strongly associated with their biogeographic ancestry which to some to a third brother, Ahmed Chelh, who was arrested and charged with the murder
extent can be informed by DNA. The term ‘race’ refers to a largely outmoded in October 2015 32 . The case was never heard in court; Mr Chelh was found dead
concept of human classification. in his cell in January 2016.

DNA tests for refugees by the UK Border Agency

In 2009, the UK Border Agency piloted a scheme to determine the biogeographical
ancestry of asylum seekers through DNA testing. It came in response to concerns
that people were lying about their country of origin in order to boost their chances The use of new DNA tests to predict ancestry, skin, hair and eye colour
of a successful asylum application. Major scientific29 and ethical criticisms led reopened this cold case and provided a key shift in focus for the investigators.
to it being scaled back and then abandoned in 2011. Although DNA tests can We were fortunate to have completed reference databases, containing genetic
predict whether someone’s ancestors came from large geographic regions such
as continents, such tests cannot predict someone’s nationality30,31.
data from many people of North African origin, earlier in 2015 — as well as
benefiting from a close and proactive relationship with the investigating team.
Premature use of forensic DNA phenotyping
Christopher Phillips Researcher in forensic genetics,
Operation Minstead, rapes 1992-2009 University of Santiago de Compostela, Spain
EUROFORGEN Consortium member
When London’s Metropolitan Police were struggling to catch a serial burglar and
rapist, they turned to a USA-based, DNA testing company to help them establish
his ancestral origins via DNA analysis. The company used unspecified ancestry
and pigmentation markers to predict that the assailant came from Southern
Caribbean regions, so investigators flew to Trinidad. When the perpetrator was All in all, predicting physical appearance from DNA is still in its infancy.
finally caught, it turned out he was from Jamaica. Biogeographic ancestry tests But advances in the science are happening so quickly that the investigative
can only narrow down to broad geographic regions not specific countries. and ethical implications need to be considered sooner rather than later.

29
B
 alding D, Weale M, Richards M, Thomas M (2010) Genetic and isotopic analysis and the UK Border Agency.
Significance 7: 58-61
30
T
 utton R, Hauskeller C, Sturdy S (2014) Suspect technologies: Forensic testing of asylum seekers at the UK
border. Ethnic and Racial Studies 37(5): 738-52
31
P
 hillips C (2015) Forensic genetic analysis of bio-geographical ancestry. Forensic Science International:
32
Genetics 18: 49-65 h
 ttp://elpais.com/elpais/2015/10/04/opinion/1443970167_046935.html (accessed January 2017)

34 PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA 35 PREDICTING APPEARANCE AND BIOGEOGRAPHIC ANCESTRY FROM DNA
FINAL THOUGHTS FROM EUROFORGEN USEFUL TERMS TO KNOW

DNA profiling is a powerful tool, both for securing criminal Autosomal DNA Likelihood Ratio
convictions, and excluding or exonerating the innocent. DNA from the 22 pairs of non-sex chromosomes, A statistical calculation that summarises the
found in the cell nucleus. relative support for two hypotheses provided
Yet its story is far from complete. Forensic genetics continues to be an innovative,
by some evidence. When the likelihood ratio is 1,
dynamic and evolving field of research, and the amount of information that can
Confirmation bias the evidence available provides equal support
be gleaned from the tiniest traces of DNA continues to grow. It is time to take
for both hypotheses. (Formally, it is the ratio
stock of these increased possibilities, and address the challenges that enhanced The tendency to interpret new evidence as of the probability of the evidence under the
DNA analysis could bring. confirmation of one’s existing beliefs or theories. two hypotheses).
The contributors to this guide have been at the forefront of the development
of new types of forensic genetic tests. One strength of the EUROFORGEN Chromosome Locus (plural loci)
Consortium is its multi-national composition and its awareness that different
countries have adopted different strategies for processing, interpreting The human genome is composed of 23 pairs A specific, identifiable place in human DNA
and presenting complex DNA evidence. of chromosomes (46 in total), each of which where there is variability between individuals
contain thousands of genes and non-coding DNA. (genetic markers such as STRs, SNPs).
As DNA profiling continues to grow more sensitive, and it is used in more
investigations, the need for accurate communication between scientists and non-
scientists only grows - both to ensure that their expectations of the technology are DNA Mitochondrial DNA (mtDNA)
realistic, and its limits are properly understood. The collaboration between Sense DNA is the molecule that carries the genetic DNA from mitochondria, which are small
about Science and the EUROFORGEN Consortium aims to improve this process information of most organisms including humans. energy factories existing in the cell in numerous
and inform the public about new developments in this exciting field. copies outside of the nucleus. As there are lots
DNA profile of mitochondria in cells their DNA is present
in larger amounts and can be more easily
In the forensic context, this describes the detected when DNA from the nucleus is limited
visualisation of the genetic markers which have or degraded.
been analysed in an individual’s DNA. The most
commonly used is an STR DNA profile.
Mixed DNA Profile
Forensic analysis A DNA profile involving two or more contributors,
eg victim(s) and suspect(s).
Scientific tests or techniques relevant to legal
proceedings.
National DNA Database
Genetic Marker Most European countries have national
forensic DNA databases storing DNA profiles
Sections of the genome that can have different from unsolved criminal cases, as well as from
forms (alleles). These sections are highly variable convicted offenders. In a number of countries,
so are chosen to distinguish between individuals. such as the UK, DNA profiles from persons
These can be detected in a laboratory and used arrested but not convicted may also be stored
to generate a DNA profile. Every individual has for varying periods of time. In this guide,
two copies of each genetic marker (because we the term ‘DNA database’ refers always to these
inherit one version from our mother and one from forensic databases.
our father).

36 FINAL THOUGHTS FROM EUROFORGEN 37 USEFUL TERMS TO KNOW

USEFUL TERMS TO KNOW
07 DELVING DEEPER

Nuclear DNA Short Tandem Repeats (STRs) Royal Statistical Society: Practitioner guides Forensic science: A sociological introduction
DNA from the cell’s nucleus, which encodes Small sections of DNA (found throughout the Four guides intended to assist judges, lawyers, Christopher Lawless (2016) draws on a wealth
the vast majority of an organism’s genes. human genome), which are made up of short forensic scientists and other expert witnesses of international research and case studies
The most common form of DNA profiling sequences that are repeated. The number in coping with the demands of modern criminal to explore the intersection of science, technology,
involves nuclear DNA (including autosomal DNA, of times this sequence is repeated (and hence litigation. law and society and examine the production
Y-chromosomal DNA, but excludes mtDNA, the length of the section), tends to differ between of forensic knowledge.
http://bit.ly/2hxZLKa
that is found outside the cell nucleus). unrelated individuals and can be measured using
STR analysis. This principle forms the basis of the
most common types of forensic DNA profiling Gov.uk DNA guidance Inside the cell: The dark side of forensic DNA
Partial DNA Profile using autosomal STRs (ie those located on the Erin Murphy (2016) probes the scientific,
An incomplete DNA profile, where some non-sex chromosomes). Each STR marker carries Brings together guidance on DNA published by the
statistical, legal, and ethical challenges presented
of the genetic markers analysed are missing. two repeats, one inherited from the mother Forensic Science Regulator.
by forensic DNA testing.
This can be because the DNA has been and the other from the father. They are identified https://www.gov.uk/government/collections/
degraded by, for example, exposure to heat, by numbers to signify the length of repeat dna-guidance
water or microorganisms, or because DNA sequences: “11 / 15” or “11 / 11”. Misleading DNA evidence: reasons
is present at such low levels that accurate marker for miscarriages of justice
Y-STRs are STRs found on the Y chromosome Genetics, technology, security and justice:
information cannot be obtained.
(male only; see below). The other sex crossing, contesting and comparing boundaries Peter Gill (2014) Elsevier. Published under
chromosome is the X chromosome; females EUROFORGEN funding. The book provides
Phenotype normally carry two X chromosomes, males carry Six Economic and Social Research Council a deep analysis of the Adam Scott, Farah Jama
one X and one Y. seminars that critically examine aspects of and Meredith Kercher cases, and describes the
The physical characteristics of an individual the contributions of forensic genetics to the utility and pitfalls of National DNA databases.
which are a result of the expression of their production of security and justice in the UK and
genes, as well as environmental factors. Single Nucleotide Polymorphisms (SNPs) other contemporary European societies.
Forensic DNA phenotyping is the prediction Another form of variability in the DNA https://www.northumbria.ac.uk/research/
Probability and statistics in forensic science
of one or  more externally visible aspects of individuals, this time at a single position academic-departments/applied-sciences/ Isaac Newton Institute for Mathematical Sciences
of these physical characteristics from the DNA in the DNA sequence (rather than a repetitive wwwnorthumbriaacukforensicgenetics/ aiming to produce guidelines for reliability
eg eye and hair colour. section like with STRs).
Matthias Wienroth discusses the series in an estimates for specific forensic techniques
ESRC blog. https://blog.esrc.ac.uk/2016/11/29/ https://www.newton.ac.uk/event/fos/seminars
Y Chromosome DNA geneticstechnology-security-and-justice-the-
DNA from the Y chromosome, one of two sex social-lifeof-dna/#more-1773
chromosomes, inherited from father to son Read our guides at senseaboutscience.org
so only carried by males. It is found in the The forensic use of bioinformation:
cell nucleus.
ethical issues
Fingerprinting and DNA profiling are valuable
tools in the fight against crime, but there is a
debate about whether police powers to keep
people’s details on record are justified. Nuffield’s
report (2007) makes recommendations in areas
including the use of the National DNA Database. Find out about EUROFORGEN-NoE research
http://nuffieldbioethics.org/project/ and information resources at
bioinformation/ www.euroforgen.eu/training/online-resources/

38 USEFUL TERMS TO KNOW 39 DELVING DEEPER

 ABOUT US...
The European Forensic Genetics Network For more copies or further
of Excellence has been operating and information contact
developing during five years thanks to the Sense about Science:
funding of the European Union to bring
hello@senseaboutscience.org
together forensic scientists, social and legal
researchers from nine European countries, +44 20 7490 9590
who study novel forms of forensic DNA www.senseaboutscience.org
profiling and searching techniques. The
EUROFORGEN Community will continue to
exist in the framework of the International
Society for Forensic Genetics.
To achieve a broad public awareness about
the advances and key issues in forensic
genetics, EUROFORGEN has partnered with
Sense about Science to produce “Making
Sense of Forensic Genetics”. Sense about
Science, is an independent campaigning Published in 2017 by
charity that challenges the misrepresentation
of science and evidence in public life.
Sense about Science advocates openness
and honesty about research findings, and
works to ensure the public interest in sound
science and evidence is recognised in public
Sense about Science
discussion and policy making. Sense about 14a Clerkenwell Green
Science focuses on socially and scientifically London EC1R 0DP
difficult issues where evidence is neglected,
Registered Charity No. 1146170
politicised or misleading. Sense about Science Company No: 6771027
is a small team working with thousands of
supporters, from world-leading researchers to This document is licensed under Creative
community groups. Commons Attribution-Noncommercial-
No Derivative Works 2.0 UK:
England & Wales License.

This project was financially supported from the European Union

Seventh Framework Programme (FP7/2007-2013) under grant
agreement n° 285487 (EUROFORGEN-NoE).