Please post your questions for this week in the comments section below. Penny will select the most liked/interesting questions and publish her response to these on this step by Wednesday of Week 4.
Please ‘like’ questions posted by other learners if you are also interested in having these answered.
Thank you very much for all of your great comments and questions this week, it’s fantastic to see that you are all still enjoying our case, and that you’re thinking so carefully about the details. You put forward some really great questions about the DNA material, it’s great to see that you’re understanding this topic so well. I hope you find the answers below useful, and remember there will be another ‘Ask Penny’ next week. For those of you who commented that you couldn’t find the answers to the ‘Ask Penny’ questions in Week 1, these were posted on the ‘Ask Penny’ step in Week 1.
Peter C asked how the different chromosomes are separated before extracting the relevant STRs for profiling, and the answer is that they aren’t. When we do DNA profiling we target multiple STR regions around the genome, and these are usually all on different chromosomes. However, when the DNA is extracted from a sample, all of the genetic material is extracted together and so the chromosomes aren’t separated. We therefore use downstream processes to separate out the different STRs. Firstly, we use a technique called the polymerase chain reaction (PCR) to make many copies of all of the targeted STR regions from different chromosomes in the genome. In order to identify which region is which, we target different length regions around each of the STRs, and label the copies with different coloured fluorescent dyes. We then determine what is in the sample by determining the length of the copies, and the colour of dye that they are labelled with. The combination of length and colour separates the different STR regions, allowing us to build a profile of what genetic types an individual has at each of these regions in their genome.
Fiona B asked an excellent question related to this, about how we know that the STRs used for comparison are from the same place on the same chromosome, and why it is not possible for the primers to attach to the same base sequence elsewhere in the DNA. The answer is that the primers are long enough that the sequence they contain should be unique in the genome, and thus they should only bind in the specific region you are targeting—this is also controlled by the annealing temperature used in the PCR, and this will be set at a level high enough that only sequence-specific binding can take place. If there were any binding anywhere else in the genome then in order to produce any amplicons, both of the primers in the pair would need to bind close enough together that the DNA polymerase enzyme could copy across the gap, which would be unlikely anywhere except the specific target region. In addition, all of the commercial STR kits have been optimised, tested and validated, and shown to reliably amplify only the specific region they are targeting. The only situation where a problem might arise is if an individual had a duplication of a region that is targeted by one of the STR primer pairs, as then this could amplify this STR from two regions. However, this would be very rare, and would likely be detected as an anomaly, because the profile produced at this STR region for that individual would probably have more than two alleles, and this would be flagged up and investigated further.
Anne Weatherly asked whether DNA results are subject to the same independent review as fingerprint results. This will differ in different forensic labs, but most labs will have some form of verification process whereby the results determined by one DNA analyst will be checked by another. In addition to this, there are various processes that are used to continually check that DNA profiling procedures are working correctly. This includes proficiency testing, and most labs will carry out regular testing of their analysts by checking the results they get for samples with known DNA profiles. Often the analyst will not know they are being tested so that they don’t unconsciously change the way they work. All of the techniques used in the DNA profiling procedure are also validated, i.e. tested with known samples to determine that they are working correctly. In addition, labs who do DNA profiling should be accredited, which means that they are regularly checked and inspected by accrediting bodies to make sure they are carrying out procedures correctly. In the UK, for example, there are fewer than 20 labs who are accredited to generate DNA profiles for loading onto the National DNA Database, and this is to ensure that the data is as reliable as possible. In addition, labs will always use positive and negative control samples to ensure that all of the regents and techniques are working properly, and that there has not been any contamination of samples. They will also usually have DNA elimination databases, which contains the DNA profiles of all of their staff and any individuals who may come into contact with evidence samples. This means that all DNA profiling results can be checked against this database to ensure that there has not been any contamination in the lab.
Gwynne Harper asked about whether there is an international consensus on which STRs to sequence and amplify, and whether this makes it possible for different countries to search each other’s DNA databases. Different countries have historically used different set of STR loci, or sets that only partially overlap, but over the last decade countries have reviewed these and there is now a much greater consensus over which STRs are analysed. Until recently, the standard DNA profile in the UK consisted of ten STR regions plus a marker to determine the sex chromosome complement of the donor of the sample, i.e. whether they have two X chromosomes or and X and Y chromosome. This was known as the SGM Plus system, and was developed by the UK Forensic Science Service in 1999. However, in the last few years labs in the UK have moved to new systems. In England and Wales, since 2014 a DNA profile now consists of 16 STR loci plus the sex marker, and in Scotland we moved to a system with 21 STR loci and three sex markers in 2015. Part of the reason for implementing these new kits, as well as increasing discriminatory power by increasing the number of loci, was to bring the UK in line with the recommendations of the European Network of Forensic Science Institutes and the European DNA Profiling Group, who specified a European Standard Set of loci. In terms of sharing DNA profile information more widely among countries, this is a subject of much debate, and may people have concerns about the sharing of DNA profile data between countries, for example due to the potentially sensitive information stored within an individual’s DNA profile, and the wide variation in scientific standards among different countries. Sharing of DNA data is not done automatically, but depends on legal agreements between different countries. For example, several EU member countries have signed bilateral agreements with the United States to share DNA profiling information. Countries who are members of INTERPOL are able to submit DNA profiles to the DNA Gateway, which automatically searches against DNA profiles contributed from 73 member states and returns a result within 15 minutes. Within the European Union, the sharing of DNA profiling information is also carried out through the Prüm Convention, a treaty signed in 2005 allowing EU member states to make direct searches of DNA profiles in national DNA databases of other EU member states. Currently, not all member states are signed up to this agreement, and the UK initially opted out of the treaty due to concerns over the increased chance of adventitious or false matches due to the unusually large size of the UK database. This is exacerbated by the fact that until recently there was only limited overlap between the UK and countries in Europe in terms of the regions of the genome that were tested. This meant that a search could be made under the Prüm Convention with data for only six regions (whereas at the time typically ten regions were tested in the UK), which makes the likelihood of a coincidental match much higher. The newly adopted DNA-17 system in England/Wales and DNA-24 system in Scotland means that there is now much greater overlap with DNA profiles produced in Europe, and so the chance of adventitious matches should be lower. As a result of this and advice from law enforcement agencies that Prüm offers significant benefits for the investigation and prevention of crime in the UK, in December 2015 the UK parliament voted to join the Prüm Convention. This is considered to be much more efficient than current methods of cross-border cooperation through INTERPOL, and will include safeguards against the issue of potential adventitious matches, with the National Crime Agency acting as a gatekeeper for the process. In late 2016, the UK government confirmed that this would go ahead despite the UK EU referendum result in June 2016, although it remains to be seen whether this will be the case.
Sandra Harper asked whether in rape cases, the DNA analysis provides information on whether the suspect is a secretor or not, that is whether they are one of the ~80% of the population who secrete their blood type antigens into their body fluids, meaning that blood group typing can be determined from saliva and semen samples. This cannot be determined using DNA analysis, as STR profiling just looks at the length of different genetic regions, and does not include any antigen testing. Blood group typing is no longer carried out for forensic purposes and has been superseded by the greater discriminatory power of DNA profiling. Sandra also asked whether DNA can be extracted from a suspect who has had a vasectomy, as there are no sperm cells present, and this is where the majority of DNA in semen comes from. As Nadia Haouaoui and Anna Dion point out, although there are no sperm cells in the semen, other cellular components are present, e.g. epithelial cells, from which DNA can be extracted. This can cause difficulties in that the amount of DNA extracted is much lower than it would be if sperm cells were present, and if the semen is mixed with biological material from another individual, e.g. the other individual involved in an alleged sexual offence, then this can swamp the DNA from the semen, making it difficult to analyse. There are some ways around this, for example if there are two individuals in the DNA mixture and only one of them has a Y chromosome, then the analysis can focus on STR regions only on the Y, meaning that the larger DNA component from the other individual in the mixture is not analysed. There has also been some progress in the use of laser microdissection to isolate single cells from one individual so that the DNA can be extracted only from these cells - you can read more about this here.
Ludovica Gorza asked two questions, the first about whether PCR is destructive and how this is dealt with in court if no other samples can be tested. PCR is destructive in the sense that part of a DNA sample is used up during the analysis, and when samples have very low levels of DNA in them (as if often the case for forensic samples) this can mean that only a limited number of tests can be carried out. If this is the case then the forensic DNA analyst has to be very careful when deciding what tests should be carried out on the sample, and try to preserve this as much as possible. If there is not enough material to carry out the appropriate tests then this can prove difficult when cases get to court, and the DNA evidence may not be very powerful in these cases. However, it is important to note that the DNA profiling kits being used now are incredibly sensitive. Most commercial kits currently on the market are optimised for use with around 0.2-1 nanograms of DNA, or 200-1000 picograms. 200 picograms represents the DNA from approximately 30 cells, and this quantity of DNA can reliably and robustly produce a DNA profile, as long as there are no inhibitory chemicals in the sample, which can be the case in some specific types of samples—for example the haem present in blood, indigo dye in denim, calcium in bone or melanin in hair can inhibit the reactions used in the DNA profiling process. Many kits will work with smaller quantities of DNA than this, and various adjustments can be made to DNA profiling techniques to increase the chances of generating a useable profile from samples containing small quantities of DNA, although this can lead to increased chances of artefacts in the profile. Useable, full DNA profiles have been generated from only 10-20 pigograms of DNA, representing only around 2-4 cells.
Ludovica’s second question related to monozygotic twins, and whether there is any way of genetically differentiating them, even though they will have the same DNA profile and in theory have identical genome sequences. Until recently, one of the limitations of forensic DNA analysis was that it could not be used to tell identical twins apart. However, in 2014 a German company called Eurofins published a paper in which they sequenced the whole genome of a set of identical twins, as well as the child of one of the twins. Out of the three billion letters in the human genome, they were able to identify five genetic mutations that were present in the child and the twin who was his father. This has the potential to allow investigators to distinguish between identical twins in paternity cases, or criminal cases where one of a set of twins is the alleged source of DNA found at a crime scene. The study used a technique called next-generation sequencing, which is a method for very quickly and accurately sequencing a large amount of DNA. Although such methods are currently too expensive to be routinely implemented in forensic casework, the use of these new technologies looks set to revolutionise the way forensic DNA analysis is carried out in future. You can read more about the use of this technique for discriminating twins here.
Dorothy R asked about mitochondrial DNA and why it isn’t used more often, given that it is present in higher quantities than nuclear DNA. The main reason is that mitochondrial DNA has much less power to discriminate among different individuals, because it is a single chromosome that is inherited in one block, and so it essentially has the power of a single genetic region, whereas using STRs we are able to look at many different independent regions in the genome. This means that there is much more chance of different individuals having different profiles for STRs than for mitochondrial DNA—all individuals in the same maternal lineage will have the same mitochondrial DNA, for example. In addition, analysis of mitochondrial DNA is rarely carried out as standard in forensic DNA laboratories, and usually requires a specialist laboratory with different types of expertise. As mentioned above, next-generation sequencing is being increasingly used in forensic DNA analysis, and so there are some labs moving to sequencing the whole mitochondrial genome, which is likely to provide greater discriminatory power.
Doug Karo asked three questions about DNA evidence, the first relating to the issue of body fluid samples deposited at different times, and whether there is any way to determine the time of deposition of different samples. With the increasing sensitivity and efficiency of DNA profiling techniques, Doug is correct that DNA can often be isolated from old samples, and this can raise the issue of whether samples relate to the incident in question, or whether they could have been deposited prior to the event. Conversely, if a body fluid is known to be involved in an incident then working out the time at which it was deposited can give information about when the incident occurred. There are currently no tests that can reliably determine the time of deposition of body fluid stains, but there is a great deal of research being done in this area. One promising line of enquiry involves extracting RNA from samples—RNA is a molecule similar to DNA but whereas your DNA is the same in almost every cell of your body, RNA differs between different cell types depending on their function. A number of researchers have been testing whether the rate of degradation of RNA molecules in samples is related to the length of time since they were deposited, and there are some promising results coming out.
Doug also asked whether there is ever any contention over the probabilities put forward in court in relation to the matching of DNA profiles between evidence samples and suspects, and the answer is not really—the techniques of DNA profiling and the associated statistical evaluation of the results are well established and accepted, and usually go uncontested in court. Doug raises the point that judges and juries may not be able to evaluate this evidence fully, and this is an important point in relation to the presentation of forensic evidence in court. Scientists should communicate the principles behind their science in a manner that can be understood by those who do not have scientific background, but this can be challenging, particularly with quite technical evidence types such as DNA.
Finally, Doug asked if different labs analyse the same sample, how similar are the results likely to be. The answer is that they should be absolutely identical, if they are analysed using the same kits, or kits that examine the same STR loci. The only time differences might be expected to arise is when samples have very low levels of DNA in them, as this can cause stochastic variation in the profiling of these samples, which can give slightly different results. However, in most laboratories, very carefully controlled procedures are in place to ensure a consistent approach to analysing these types of samples, to maximise the reliability of the results.
This leads us on to the question from Norman Banfield, about what the pitfalls of DNA analysis are, even though it has been proven to be reliable. Although the techniques used in DNA profiling are well established and accepted, there are a number of pitfalls associated with difficult sample types, and with the interpretation of DNA evidence. One issue is around the analysis of samples with very low levels of DNA in them, as described above, and this can still be very challenging. Another issue relates to the fact that we can’t tell from a DNA profile what types of tissue or body fluid the DNA came from. There are some chemical tests that can be done that indicate a particular body fluid, e.g. blood, semen, saliva is present, but these are not conclusive and there is no test for skin cells or sweat, or for vaginal samples. Even if a positive test for e.g. blood is found and then a DNA profile is recovered, there is no information in the DNA profile that definitively links the DNA profile to the blood. Often this link is assumed, and sometimes this is a reasonable inference to make, but as DNA profiling techniques become increasingly sensitive, it is going to become harder for scientists to make that assumption, as we will increasingly be picking up ‘background’ DNA profiles in samples, which have no connection to the alleged incident. This also relates to the issue discussed above about our inability to determine how and when a sample was deposited, and DNA analysts have to be very careful in their interpretation of DNA evidence to consider both how the DNA was transferred to wherever it was recovered from, and when. Given that DNA can be deposited on a surface by secondary transfer via an intermediary object or person, and under the right conditions can be retained intact on a surface for significant periods of time, we have to consider the possibility that the DNA may not have been deposited by the person it matches, or in the time frame in which the incident under investigation occurred. Another area that can raise difficulties in the analysis of DNA from forensic samples is in the interpretation of DNA mixtures, i.e. DNA profiles that are produced when more than one individual has contributed to the sample—this is particularly common in sexual offences, because samples will often contain biological material from an individual who has been attacked, and from their attacker. Mixed DNA profiles are extremely common, and as DNA profiling techniques become more and more sensitive, this issue will become even more widespread. This type of result can be very difficult to interpret and it can be very challenging to determine which components in the profile come from which individual. Many forensic laboratories will only interpret mixed DNA profiles with relatively small numbers of contributions (e.g. from two or perhaps three individuals), and this is a major challenge in the field. Some very complex statistical analyses are being introduced into DNA profile interpretation, which can assist in separating out the most likely contributions from different individuals in the mixture, but this is not at all straightforward.
Éva Rompos asked a series of questions about section 3.12 but I’m not sure which part you mean as this is the section on extracting your own DNA from a saliva sample. It looks from the questions like the issues are around swabbing for DNA on evidence items, so in that context, you don’t know where the DNA is on an item as you can’t see it if there isn’t an obvious body fluid stain present. In that case, you would either swab the whole surface of an item, or think carefully about how an item might have been handled to suggest where it might be most appropriate to sample. It would be unlikely you would recover all of the DNA from the surface so some would likely remain, but swabbing with a wet swab and then a dry swab can maximise the chances of recovering as much DNA as possible. It is certainly possible that more than one person’s DNA is present on an item that is swabbed, and this can cause difficulties with DNA interpretation, as discussed above—separating these profiles after amplification can be very challenging indeed.
Michael Boyle asked whether forensic experts work in isolation for the rest of the case in order to avoid bias, or whether they are kept up to date with developments in the case in order to allow them to prioritise their examinations. This is a really important question, and the answer is that it really depends on the circumstance of the case, and different labs/organisations will work differently. There are good arguments that scientists should not be given any case information, to minimise any unconscious bias in their assessment of the evidence, but sometimes it is necessary for them to be given some information in order to be able to best focus their efforts. For example, it can be important for the second fingerprint examiner in a case not to know what the first examiner’s conclusion was, to avoid them being unavoidably and unconsciously biased by this information, but in the case of sexual offences, it can be important for an examiner to know the details of what has been alleged in the incident, in order for them to know what types of evidence might be found and where.
Leslie Grey McCawley asked two questions, the first about whether if a surface has multiple overlapping fingermarks can they somehow be effectively differentiated for closer examination. This can be extremely difficult, and often nothing can be done to separate the fingermarks. Sometimes, if the marks have been deposited in different substances, different enhancement techniques can be used on the different marks and then photographed between each treatment to give an enhanced version of the marks one at a time. In addition, if a fresh mark is overlaid on older marks then the fresher mark will take up fingerprint powder better and so will be enhanced over the other marks so can sometimes be recovered. There is also research work being done by chemists using spectroscopic methods to distinguish between overlapped fingermarks, and by computer scientists scanning overlapping fingermarks and using digital algorithms to try and separate them out. I don’t know a great deal about these methods but I found a couple of research papers relating to the spectroscopic and scanning methods and another one here. Leslie also asked whether it is illegal to give blood in someone else’s name, and I’m afraid I have no idea! In the UK when you give blood you have to confirm your identity, so I think it would be difficult to give blood under a false identity, but I’m not aware of any legislation relating to this.
Philip Parish asked a question about a scenario of a woman who has sex with someone at a party, and then whilst walking home is raped by another person who uses a condom, but she is unaware of this. When she reports the incident and intimate samples are taken, semen is detected in the samples and a DNA profile from the person she had sex with at the party is recovered. This highlights an area where forensic science is often not able to assist police investigations, as it can only be used to examine what physical evidence is present, it cannot tell you anything about consent. In these circumstances, the police would likely investigate all individuals involved and consider all of the different accounts of the evening. They would always consider the possibility that a condom had been used in an incident, and some labs will test intimate swabs for the presence of lubricants from condoms. The police may also search for a discarded used condom if the area where the incident occurred could be identified. If a condom were recovered then as well as detecting the semen and DNA profile of the attacker inside the condom, scientists would also swab the exterior of the condom to try and recover epithelial cells from the vagina of the victim—if a DNA profile could be recovered from these cells then this would provide a link between the victim and her attacker.
Finally, Larry Groseclose asked a question about why there might not be any blood stains on Mr Ward’s clothing, and if this related to the fact that Mrs Ward was shot at close range. Both close range and long range gunshots would produce very large forces impacting into blood and so would produce very small blood spatter, less than 1-2mm in diameter, and often in the form of a mist. With longer range shots the force may be somewhat reduced and so the spatter may be slightly larger than with closer range shots, and with closer range shots the gases from the muzzle of the firearm may influence the travel of the blood spots, so the directionality of the spots could be less reliable in determining the origin of the force. However, in general, the blood spatter would be fairly similar, and is likely to be extremely small in size. In addition, a gunshot wound to the head with no exit wound will result in instant death and so does not produce a wound that would bleed a great deal, so there may not actually be a large amount of blood present.
© University of Strathclyde