Artefacts and Noise Flashcards
Define artefacts
- things that happen during testing that can’t really be reproduced
- aren’t reflective of the DNA that is associated with a sample
What is a stutter (state the type of artefact)
- biological artefact
- small peaks found in position immediately before the peak that gives rise to them
- quite common
How does a stutter occur?
- occurs during PCR amplification
- when DNA polymerase has a hiccup and slips a bit forwards or backwards
- makes a PCR product which is a little bit shorter than true fragment
Why are stutters not a problem in DNA analysis
- software makes use of a stutter filter
- any peak, in the position immediately before the (actual) peak with a value lower than the cut-off – (usually 12%, sometimes 15%) is not labelled and is written off as an artefact (we know is a stutter)
What does helicase enzyme do?
- enzyme used in DNA replication to unwind DNA molecule from its tightly woven form
What is a spike (state the type of artefact)
- instrumental artefact (technical)
- a peak that you see in EPG that is too tall and narrow relative to what we would typically expect to see in the range of all the other peaks
What is a blob (state the type of artefact)
- instrumental artefact (technical)
- short and squat peak in comparison to what we would expect
How can we tell spikes and blobs from actual peaks?
- typical peaks have an expected range of height to area or height to weight ratios
- if they do not fit in this expected range
What causes spikes and blobs?
- originated from testing process
- not rooted from any DNA that is associated with a sample
- not actually known
- possibly associated with voltage spikes when sample is running through genetic analyser
- maybe particles that are passing through capillaries or dust moving in front of the camera
What is important for the DNA analyst to do when it comes and spikes and blobs?
- important that the DNA analyst can recognise them and say that this is not something that is rooted in the DNA associated with the sample
How are spikes and blobs recognised?
- get peaks that come from reference samples from positive controls
- spikes/blobs not part of reference positive control samples fall outside this range
- they appear in top-left hand quadrant of X and Y’ chart
What are RFUs?
- relative fluorescent units
- unit of measurement used in analysis which employs fluorescence-detection
How is fluorescence detected?
What is detection used to determine?
What does a high RFU value correspond to?
- using a charged couple device (CCD) array when labelled fragments, which are separated within a capillary by using electrophoresis, are energised by laser light and travel across the detection window
- a computer program measures the results, determining the quantity or size of the fragments from the level of fluorescence intensity
- high RFU value = contain higher quantities of amplified DNA
How do we measure peak height ratio?
- (height / height) x 100
What does it mean if the peak height ratio is low (peaks not the same height)
- not looking at a single source, but a mixture
- one person contributed more DNA than the other initially
- artefact associated with the sample and these peaks were not tall to begin with
What is a stochastic effect?
- when we have small amounts of starting material
- probabilistic effects that occur by chance
What is primer binding site mutations?
- most common form of mutation
- maybe one allele has a mutation that prevented it from generating a profile as well
What is a peak height balance ratio used for?
- when labs use a testing kit they use a cut-off of a 70 % peak height balance (some will use 60 %)
- if the peak height balance (smaller/larger x 100) is higher than that they are balanced
What does it mean if two peaks are balanced?
- likely from a single source
- no problems with amplification process
What does it mean if a peak height balance is below the threshold set by the lab?
- could be a mixed sample
- might be dealing with a small amount of material
- primer binding site mutation has taken place
What is a synonym for technical/instrumental artefacts?
- degradation
- inhibition
Define degradation
- degradation is deterioration of DNA
- as DNA is a chemical, it can interact with other chemicals which can cause DNA to get damaged/broken
- when this happens DNA molecules cannot be amplified through PCR process
- therefore they cannot contribute to EPG’s or at least not as much as before degradation occurred
Define inhibition
- inhibition is poor PCR amplification
- sometimes other chemicals that make it harder for DNA polymerases to do their jobs during PCR
- therefore DNA doesn’t get amplified as efficiently as it would have in absence of chemicals
- smaller peaks
What is the relationship between size of DNA fragment and target for degradation/inhibitors
Explain why
- bigger fragment = better target for degradation/more likely to be affected by inhibitors in amplification process
- a bit piece of DNA takes more time to go through capillaries
- it is harder to get amplified during PCR amplification process than small one
- if got something in mix that is making it hard for DNA polymerase to amplify the DNA during PCR that effect is going to be observed more for bigger pieces of DNA that smaller ones
Give four examples of stochastic effects?
- imbalance
- drop-in
- stutter
- drop-out
What is difference between homozygote and heterozygote peak heights?
- homozygote peaks - singular peaks of same allele
- taller than hetero peaks because there are two of the same alleles
- heterozygote peaks - two peaks that are different alleles found at the same locus
- smaller peak height compared to homozygous peaks
Where do different size fragments appear on the EPG?
- smaller = LHS
- smaller fragments of DNA are first ones that come off the capillary during capillary electrophoresis
- RHS because they take longer to travel through the capillary during electrophoresis
How does degradation and inhibition effect an EPG?
- RHS is getting shorter and that height decreases progressively as move from left to right
- this is because the larger peaks are more likely to be degraded
- this creates a ski slope (slant on EPG left to right - hallmark of a sample which has suffered from inhibition/degradation)
- this ski slope means we can get allele drop-out take place at far RHS
What is allelic dropout?
- the test has failed - got incomplete picture of actual DNA profile
- with degradation and inhibition, dropout can easily occur particularly for loci that give rise to peaks on RHS of EPG
- alleles present in less degraded sample but not showing up on more degraded sample
What is locus drop out?
- when both alleles in a particular locus have dropped out so there is no alleles present on EPG at that locus
What type of issues are more of a type of problem in evidence samples rather than reference samples?
- peak height imbalance
- degradation
- inhibition
What can be said about recognition of artefacts?
- often they can be recognised and at the very least they raise some questions about how certain we can be about the reliability of the test and the conclusions that can be drawn from it
When undertaking interpretations of EPG, what do we need to understand?
- need to understand clearly what is signal and what is noise
What is baseline noise?
- unbalanced third peak at locus
What probabilistic effects we may encounter with low-level DNA material?
- stochastic effects
In which way do labs distinguish between signal and noise?
- have a minimum peak height threshold
- if it is not above this, it is written off
- most labs settle around 150RFUs
- some 200, some 100
How are the minimum peak height thresholds determined?
- during validation studies
What can be said about thresholds by labs?
- they eliminate noise (even at the cost of eliminating signal)
- some will say they can arbitrarily remove legitimate signal
How are baseline fluctuations/noise created?
- any kind of instrument that is recording voltages or light impulses there will be fluctuation of the baseline, just a random noise (a static), sort of effect
- sometimes the static will be for instance louder than at another time, we can measure the heights of those random bits of noise and get an average value and a st. dev. associated with that the average value for background noise
What is LOD?
- limit of detection
- mean background signal + 3 st. dev.
- if something above this, very unusual it is noise or something else
What is LOQ?
- limit of quantification
- mean background signal + 10 st. dev.
When are there more likely to be problems with allelic and locus drop out?
- when small amounts of material are used to generate DNA profiles
