Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

7BBFM128: Forensic Biology > DNA Mixtures > Flashcards

DNA Mixtures Flashcards

You may prefer our related Brainscape-certified flashcards:
AP® Biology flashcards Biology 101 flashcards
1
Q

Mixture recognition

A
  • Safe recognition of a mixture of DNA from more than one person is an important part of forensic analysis
  • Ability to detect a mixture will depend on the relative contributions to any mixture present
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the types of things we need to find out from DNA mixtures?

A
  • How many contributors?
  • To ensure that allelic artefacts (stutters) and non- allelic artefacts are not confused with alleles
    o Stutters to the left are under 15% of large peak
    o Stutters to right are under 5% of large peak
  • To identify loci where potential drop-out has occurred (limited or degraded samples)
  • Many guidelines have been developed:
    o Clayton guidelines (1998)
     Use these to decide whether further statistical analysis needs to be done
    o ISFG (2006)
    o SWGDAM (2017 and 2021)
    o UK Forensic Science Regulator (2018)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are the three types of alleles in a profile?

A

Unambiguous allele

Allele masked by a stutter

Alleles that has dropped out and therefore have not been detected

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Whats an unambiguous allele?

A

o If more than two peaks at any one locus a mixture is present
o If peaks are unbalanced, then a mixture might be suspected
o Peak imbalance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What can lead to peak imbalance with unambiguous alleles?

A

 Natural variation – generally up to 60% peak height variation
 Primer binding mutation- to find this out you could change where your primer binds by designing your own or using different manufacturers kit

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

A Mixed STR profile

A
  • This is just an example of a mixed profile – with DNA from more than one person present. Instead of only two peaks at each area tested, there are three or four. It’s obviously more complicated because we don’t know which ‘bits’ of the profile come from one person and which bits come from the other, but the heights of the peaks can help us work this out
  • This mixture includes at least one male
  • This imbalance of X and Y could indicate female present but also may not, lack of confidence with this
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Tell me about the situation when an allele may me masked by a stutter

A

o A high level of stutter may suggest the presence of an allele co-incident with the stutter
o Typically stutter will have peak < 15% of height of parent peak but this is variable
o Over-stutter typically has peaks <5% of height of parent peak

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Typically stutter will have peak < 15% of height of parent peak but this is variable and depends on what?

A

Locus – some have low levels of stutter – e.g., Th01, Pentas (why we go to 4 or 5 bp)
Low level DNA – stochastic effects associated with low level DNA may lead to extra amplification of the stutter allele (stochastic effect is variability to do with amplification)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How does stutter product form?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Alleles in stutter positions in a mixture could be…?

A

o Stutters
o Alleles
o Indistinguishable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

If the height is greater than expected stutter ratio for that locus, then = ?

A

allele

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

If height below expected stutter ratio, then…?

A

o = possible allele if that height fits with other peaks in non-stutter position in the mixture
o = likely stutter no evidence against that

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Tell me about an allele that has dropped out and therefore not detected

A

o If evidence that the DNA is at a low level then, even if no evidence of more than one contributor, cannot assume only one contributor
o Stochastic effects could have meant that the peak wasn’t amplified

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Poor quality profile- drop-out

A
  • RFU= Relative fluorescence unit which gives an indication of the amount of DNA you have (RFU represented under the allele number)
  • Low level partial profile from at least one person and at least one male present
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Tell me about situations of allele drop-out

A
  • If allele ‘a’ at very low level, then ‘b’ may have dropped-out
    o If suspect is ‘ab’ then drop-out must be considered
  • Probability (drop-out) decreases as the height of ‘a’ increases
    o Leaving the locus out will be prejudicial against the suspect
  • If drop-out is required to explain the prosecution hypothesis (that the suspect’s DNA is there) then the height of the observed allele must support drop-out
  • Interpretation should not be attempted where background noise dominates a profile
  • Other reasons for peak imbalance in isolated loci should also be considered
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What can lead to peak imbalance/ lack of concordance in isolated loci?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Whats low template DNA?

A
  • Heterozygote balance and stutter rules do not apply where stochastic effects are likely to be enhanced (100 – 200pg analysed)
  • Must consider drop-out and drop-in (contamination) in the analysis
  • Helpful to enhance the analysis
  • If analysing <100pg then could run into more stochastic effects than expected
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Improved detection can be done by increasing the amount of DNA, what can a peak imbalance then suggest?

A

1. This is a single homozygote 15,15 with a large stutter because of the low amount of DNA being analyses,
OR
2. This is a mixture of more than one person
Not possible to say what the combinations are
3. Could the 14 be a DROP IN? Need to repeat the analysis

Say at least X number of people when writing witness statement

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are some examples of low template artefacts?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

According to Clayton guidelines (1998), when identifying a mixture and there are extra bands present, what could this be due to?

A

 Trisomy
 Somatic mutations / chimera
 Aberrant locus specific sequences
 Non-specific artefacts
 N bands
 Software (matrix) problems

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Isolated incidences of 3 peaks when there is one contributor, can lead to what types of patterning?

A

* Type 1: two peaks similar peak and 1 peak greater
* Type 2: three peaks of similar heights

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Tell me about type 1 patterns

A
  • Type 1 patterns are due to somatic mutations during an individual’s development
  • Produces a chimera – some cells have the original allele and others have the mutant allele
  • Most common in the D18S51 locus
  • Peak heights are unbalanced
23
Q

Tell me about a type 2 pattern

A
  • A type II pattern may cause by a chromosome duplication – a trisomy
  • Peaks are balanced
  • Commonly seen in association with Chromosome 21 – Down syndrome
  • Or may be the result of a locus specific sequence aberration
24
Q

What are the two types of chimeras and tell me about each type

A

* Artificial chimera – bone marrow transplant
o Tends to micro chimerism with mixtures seen in all loci
o Affects blood progenitor cells

* Natural chimera
o Merging of non-identical twins
o Different tissues have different profiles

25
Q

Non-specific artefacts

A
26
Q

Tell me about** N and N+1 bands **

A
  • To do with adenylation with PCR
  • At end of PCR is elongation step at 4˚c for infinity for the adding of adenine band at end of product
  • Get split peaks when adenylation is incomplete
  • Can also get split peaks if have too much DNA
  • 1000-2000 RFU is optimal for DNA levels
  • This at >6000 RFU has too much DNA to add the extra adenine on the end
  • Flat peak which is near off-scale, is where the height of the peak is probably much larger than that it is just off the scale
  • Need to ensure don’t have microvariant which is 1bp smaller
27
Q

Tell me about polymerase adenylation

A
  • Taq polymerase often adds an extra nucleotide to the end of a PCR product – usually an A (adenine)
    o Called ‘adenylation’
    o Some manufacturers will add a sequence to their primers to promote adenylation
  • An extension of time at the end of the PCR cycle often introduced to give the polymerase more time to add the adenine
  • If you have too much DNA template, then you may have incomplete adenylation because you don’t have enough polymerase
28
Q

Artefact, or DNA from another person?

A
  • Dark purple line would be the software highlighting something to look at, a potential problem
    o Due to poor calibration of system, known as pull-up (dye in the same position as an allele) i.e., the allele 10 in green does not exist but has been pulled-up
  • Difference in peak height could be imbalance or the presence or another person
29
Q

Matrix effects

A
30
Q

Tell me the type of things to look at for a Mixture assessment

A
  • Alleles of a person of interest present or absent
    o Simple but dangerous
  • Need to evaluate:
    o What is a safe ‘major’ contributor?
     Needs to be assessed across all loci
     If someone is a major contributor, are saying that those alleles come from that person
    o Stochastic level contributors
    o Degradation / inhibition
    o Whether DNA from a known individual could have contributed
  • Deconvolution is ideal
    o Often very difficult and time consuming
31
Q

Whats A ‘major’ contributor?

A
32
Q

What are two thresholds that need to be determined when analysing a mixture?
What do each of these tresholds represent?

A
33
Q

According to the Clayton guidelines (1998) what needs to be identified?

A
  • Identify potential number of contributors
    o Number of bands per locus
    o Peak imbalance
  • Useful to consider what type of mixture you have
    o No defined major
    o Clear major/minor
    o Low level minor – possibility of drop-out
    o All low level – possibility of drop-out
34
Q

How many contributors?
What are the likely combinations?

A
  • A mixture
  • A mixture of at least 3 people (as have 5 alleles in first group so must have at least 3 people)
  • The alleles of those three people cannot be defined from this directly
35
Q

Clayton guidelines (1998)

A
  • Consider ratio of components in the mixture
    o If two DNA templates are mixed 3:1, then this approximate ratio will be maintained when the peak areas of the different component alleles within a locus are compared
  • Estimate the mixture proportion
    o Given the mixture proportion, and the observed peak areas, it is possible to calculate the expected peak areas for all possible combinations of genotypes that can be conditioned on.
  • Determine the possible pairwise contributions to the mixture
    o ESSENTIAL – this must be done independent of knowledge of any reference profiles
    o Objective assessment
  • PROBLEM
    o Increased sensitivity of new multiplexes has led to very complex mixtures that these guidelines cannot reasonably be used
    o Can we assess the strength of the opinion – suspect A is/is not in the mixture?
36
Q

Deconvolution

A
37
Q

DNA commission (ISFG): guideline highlighting the importance of propositions

A
  • This is a lengthy document
  • The likelihood ratio is recommended and confirmed for use
  • There are two roles for the forensic scientist
    o Investigative (there is no suspect – a database search is needed)
    o Evaluative (there is a suspect and the strength of evidence is to be reported to the court
38
Q

What needs to be looked at when assessing the weighting of the evidence?

A
  • The need to assess two propositions
  • The need to assign the probability of the observations, given the two propositions
  • The ratio of the two probabilities is the LIKELIHOOD RATIO (LR)
  • The LR is the measure of the extent of support to one proposition over the other
39
Q

Whats the hierarchy of propositions?

A
40
Q

Whats the importance of the framework to the propositions?

A
  • Semen from Mr X is present in the sample v
  • Semen from Mr X is not present in the sample
    o Source level (not addressed by DNA)
    o No knowledge of number of contributors
    o No information about other individual’s potential relationship to Mr X
  • The DNA in the sample is that of Mr X v
  • The DNA in the sample is from an unknown person, unrelated to Mr X
    o One contributor – alternative unrelated to Mr X
    o Sub-source level
  • Propositions compared must be mutually exclusive, but not necessarily exhaustive
41
Q

UK forensic science regulators guidelines

A

1. Attempt to assign a value to the (minimum) number of contributors – record reasoning

2. Consider if any of the reference profiles can be used conditionally (undisputed) – record reasoning

3. Can the person of interest (POI) be excluded – consider 1 and 2, quality of profile
o DO NOT increase the number of possible contributors in order not to bias the LR
 Observation AA – appears single source
 POI is AB – probable exclusion
 If two contributors are considered (without supporting evidence) then the LR will be higher and lead to a misinterpretation

**4. **Formulate one or more propositions that represent the prosecution point of view
o Ms A alleges rape. The mixture observed is of at least three persons
o The sample is a mixture of DNA from Ms A, the POI, and an unknown unrelated individual

5. Formulate an alternative (proxy) proposition to counter each prosecution proposition
o The sample is a mixture of DNA from Ms A and at least two unknown persons unrelated to the POI, to Ms A and each other
o No requirement for the total number of persons to be the same as in 4 but it is not normally of benefit to the defendant

**6. **In circumstances where there are multiple POIs a simple pair of propositions is not adequate
o The DNA is a mixture of Mr A and Mr B v
o The DNA is a mixture of two unknown persons, unrelated to A and B
o BUT
 this assumes equal weight to genetic evidence – one of the apparent contributions may be at a very low level
 one of the contributors may claim innocence and implicate the other
o The number of comparisons increases significantly when more people are implicated, and possible kinship complicates this more and it may be necessary to frame this as an INVESTIGATIVE rather than an EVALUATIVE report

7. If the association with the questioned sample and the crime is uncertain consider a staged approach to forming the propositions and offer an initial investigative report making sure the limitations are clearly stated
o For example, a female claims she has been held by the POI at his property; he denies all knowledge of the crime
o DNA from a flannel in the bathroom is a two-person mixture and can be fully explained by a mixture of the POI and the female
o Stage 1 – consider if female + unknown v 2 unknowns strongly support the former
o Stage 2 – conditioning on the female, consider whether POI is present in the mixture

**8. **Always consider a relative of the POI as an alternative source under the defence proposition if the circumstances suggest this might be relevant

9. Sometimes number of contributors may be very uncertain – more than 4 or 5, due to ambiguous peaks, increased stutter, forward stutter. These need special approaches often needing sophisticated software and ranges of calculation outcomes should be reported
o Increase number of potential contributors – report both
o Make use of specialist software that will consider the large range of possible combinations based on different probabilities

**10. **Only use a major/minor approach only if you can
o Attribute an unambiguous set of components to one profile
o Can justify the major without considering the POIs reference profile
o Differentially changing the number of contributors in either prosecution or defence propositions
 Increasing number of contributors in the defence proposition only tends to favour the prosecution hypothesis
 Increasing number of contributors in the prosecution hypothesis but not the defence hypothesis must be scrutinised carefully

11. A quantitative (LR) approach is possible and should be used if:
o Pairs of propositions can be formulated
o Component (artefacts) excluded with certainty
o Calculations are made within validated capabilities of the software
o Population databases used are relevant within the context of the case circumstances

**UK forensic science regulators guidelines- qualitative approaches **

12. Counting matching alleles is potentially prejudicial and should not be done
‘I have seen 28/30 Mr A’s components in this complex profile and cannot exclude him from contributing’

13. Qualitative opinion where a quantitative approach is impossible and should only be given as an interim provisional report

**14. **Any scientist giving a qualitative approach must be shown to be providing reliable opinion based on a continuous assessment against a quantitative method

15. Qualitative opinions should only be presented as INVESTIGATIVE for INTELLIGENCE

16. Qualitative EVALUATIONS should not be given in circumstances where it has not been possible to use a quantitative approach

17. Expressions of probability should not be presented in a manner that favours the prosecution
o ‘It is possible that Mr A has contributed to the mixture’
o ‘It is possible that Mr A has contributed to the mixture, but it is also possible that Mr A has not contributed to the mixture’
o Be wary of expressions ‘could have’, ‘is consisted with’ unless the converse it also quoted

42
Q

Complex mixture of at least 3

A
43
Q

R v Dlugosz [2013] EWCA Crim 2

A
  • ‘19/20 of the (matching) components of the appellant’s DNA had been present in the mixture …’
  • Is statistical evidence required if an evaluative opinion is to be given?
    o Doheny & Adams suggests not
    o Atkins & Atkins, and T, suggest you can
    o Experts in Dlugosz unable to give statistical support to their observation
  • Is a hierarchy of support required if evaluative evidence is to be given?
    o Used in Atkins & Atkins
    o Not a measurable scale and does not have a scientific basis
    o Experts in Dlugosz not prepared to assign a hierarchical statement of support to this evidence
  • Scientist in Dlugosz had evaluated thousands of cases
  • He considered it ‘rare’ to observe all 20, by chance
  • Opinion based on conclusions of unpublished research that he was aware of, but was not available for review
  • ‘An expert who spends years studying this kind of comparison can properly form a judgment as to the significance of what he has found in any case. It is a judgment based on his experience. A jury is entitled to be informed of his assessment.’
44
Q

What are the types of
mixture software?

A

Commercial or free
LRmix(Studio); EuroForMix; Genoproof Mixture; LikeLTD; DNAmixtures; FST; TrueAllele; STRmix ; ArmedXpert; LiRa; LabRetriever; DNA View; Mixture Solution; maSTR; MixCal; MatchIt

45
Q

What are the main differences between the mixture softwares?

A
  • Main differences
    o Open source / commercial
    o Non-continuous / continuous
    o Relatedness considered / not considered
    o Maximum number of unprofiled contributors under Hd
  • Differences
    o Getting the number of potential contributors wrong has more effect in a semi-continuous model
    o No true LR – suggested that a difference of one unit in a log10 scale is negligible (think of LR in terms of Log i.e., LR of 10 is log1)
46
Q

Euroforgen promotes open-source software

A

o LRMix – Hinda Haned and Peter Gill (R software)
o LRMix (Studio) – NFI developed (Java)
 https://github.com/smartrank/lrmixstudio/releases
o EuroForMix - Øyvind Bleka
o DNAXs – NFI developed expert system based on EuroForMix

47
Q

Why EuroForMix?

A
  • 2-person mixture
  • Non-continuous models look at all combinations of 15, 16, 17, 18
  • Obvious that 15/18 and 16/17 combination are more likely from the peak heights
  • EuroForMix is a continuous model as it uses more information (peak heights) than a non-continuous model such as LRMix
  • Includes a non-continuous model
  • Includes an assessment for number of contributors
  • Problems
    o Peak heights are stochastic
48
Q

Consider the following

A
49
Q

Suggested model explained by stutters and masking

A
50
Q

Compromised samples

A
51
Q

Whats are the Elements of EuroForMix?

A
52
Q

What are the Model features of EuroForMix?

A
  • Multiple contributors
    o Can condition on any number of references
    o Can specify any number of unknowns (practical limit is 4)
  • Additions
    o Drop-out and drop-in
    o Forward and backward stutters
    o Degradation
    o Subpopulation structure
  • Replicates
    o Consensus preparation not needed
    o Assumes replicates have same contributors and peak height properties
53
Q

What are the advantages of EuroForMix?

A
  • Flexible input data
    o Any kit (any number of markers)
    o Any number of replicates
    o Any population frequencies
  • Flexible model
    o Back-stutter models can be included
    o Global degradation model trend can be included
  • Model properties selected as best explaining observations
    o Peak height distribution fitted through observed peak heights
    o Drop-out properties considered as part of the fitted peak height distribution
  • Uncertainty of model parameters can be considered
  • Deconvolution and database search modules included
54
Q

What are the limitations of EuroForMix?

A
  • Effective only up to four unknown contributors because of slow speed – potentially days if greater
    o Filtering for stutter can speed the process as you remove alleles
  • Model limitations
    o Peak heights assumed to be gamma-distributed
    o Possibility that markers have different amplification efficiencies is not considered
    o Stutters have same distribution for all alleles (for the given amount of DNA)
  • Deconvolution module cannot consider uncertainties in the model parameter estimates

7BBFM128: Forensic Biology (19 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions