This week’s post is dedicated to exploring the differences between clinical and forensic samples, specifically relating to obtaining data for investigative genetic genealogy purposes. In a clinical setting it’s usually relatively straightforward to obtain high quality samples for DNA sequencing, as you generally have a living, breathing person in front of you. One might take some blood, get a fresh saliva sample, swab their cheek, or possibly even use a tissue sample from a biopsy. From there, the sample can be sent straight to the sequencing lab.
However, in a forensic setting, things are much different. In the case of violent crime such as a homicide or sexual assault, samples are usually less than ideal. Sources of DNA often include things such as blood stains or semen – either left at the scene, or swabs from a sexual assault kit. In the case of identifying human remains, the samples often contain even poorer quality DNA – often teeth, bones, or hair. In all of these instances, there are four main factors that make these samples more difficult to work with:
- Quantity – This is fairly straightforward. As opposed to a blood or saliva sample where you may get hundreds or even thousands of nanograms of DNA, forensic samples often contain much smaller quantities of DNA. This is especially true in unidentified remains cases where a bone extraction often results in less than a nanogram of DNA – sometimes even as low as only dozens of picograms. The larger the quantity of DNA available, the greater the chance of getting the sequencing coverage desired/needed. However, I have seen success such as in the Hudson OH John Doe case, with only a few cells worth of DNA.
- Degradation – In a fresh DNA sample, fragments may be thousands of base pairs (bp) long. However, over time DNA degrades, leaving much smaller fragments to work with. Forensic samples may often have fragments of <150bp, often even quite a bit smaller. In the case of DNA extracted from rootless hair, fragments are often only 40-50bp or so. To better utilize these smaller fragments, specific sequencing library prep kits such as Claret Bio’s SRSLY product are often used to obtain the best results.
- Contamination – Another hurdle I encounter very regularly is exogenous contamination, especially bacterial contamination. Quite often after doing a shallow round of DNA sequencing, we will find that only 5-10% of the sequencing reads map to the human genome, sometimes even less. The vast majority of the DNA will be bacterial DNA, which is useless for investigative genetic genealogy purposes. In these instances enrichment may be attempted, or it might even require going back to the drawing board and finding a new sample. This is especially prevalent in identifying human remains when using DNA extracted from bone/teeth, which may have been exposed to the environment for months, years, or even decades.
- Mixtures – This is really another form of contamination, specifically when DNA from another person is mixed in with the person of interest. This can be an especially difficult issue to overcome, depending on the specifics of the mixture. So far there has not been much published on resolving dense SNP mixtures, but the science is ever-evolving.
As you can see, there are many additional factors involved when sequencing DNA from forensic samples as opposed to standard clinical sequencing. Thankfully, many of these issues can be overcome by using specific lab protocols, methods and products, in addition to advanced bioinformatics techniques. However, it’s extremely important to plan for these issues, so that the best possible result can be achieved.