DNA – analysis part 2

In its simplest form, interpretation of the results consists of comparing the assigned alleles for each STR in the sample from the accused and victim, in the case of crimes against the person, with those in samples collected in the investigation. For example, the blood stains in the car in our case were compared with the reference samples from Mr. and Mrs. Ward, resulting in identification of the source of each of the two sets of blood spatters. Having gone through all this and produced a DNA profile, the next question to be dealt with is, what does it mean?

For example, if we have a DNA profile obtained from a swab recovered from a victim and the profile is no different to that of the suspect, now what? The “now what” is based on the calculation of the frequency of the profile in the population. Here are some of the factors to be considered. How often would we expect to encounter this profile in the population? Can we combine this data for the different STRs? What do we mean by population? Could the DNA profile be from someone else? Let’s start with the third question. The frequency of occurrence of the alleles in any of our STRs is not the same everywhere in the world.

Alleles are the result of mutations, and the distribution of any allele in the global population will reflect the physical migration and inter-breeding of the progenitors’ offspring. Right away, we are faced with a problem. How wide do we wish to cast the net? And what arbitrary distinctions do we wish to make? In practice, most DNA databases keep population data for the main racial and ethnic groups. We can make whatever refinement we wish to this classification eventually getting down to a small community, perhaps geographically isolated to a degree and where many of the inhabitants are related to some degree. There is a major non-genetic factor to consider too, namely opportunity.

It can be argued that the relevant population to consider in an investigation is that made up of people who could have been involved, and this group may cross accepted racial and ethnic divisions. With that caveat, if we accept a conventional definition of a racial and ethnic population, the allele frequencies can be counted in samples from the target population, usually at least 200 unrelated people, and statistically tested for validity. Frequencies for STR alleles in generally accepted racial and ethnic groups have been widely published, and we have provided some material in the resources for you to look at. The second question was, can we combine this data for the different STRs? And the answer is, yes.

The rules of population statistics and genetics provide a tool called the product rule, which says that in the case of genes which are inherited in a manner such that there is no association of one with the other, than the frequency in the population of combinations of them is equal to the product of the frequency of each. Note that to make the mathematical calculations simpler, the individual frequencies are expressed in decimal form. An allele that is observed in 5 in 200 people has a frequency of 0.025. We saw earlier that the loci of STRs used in forensic profiling are on different chromosomes, and therefore it is reasonable to assume that the genes are inherited independently of one another.

When applied to the STRs in a typical multiplex, the combined population frequency soon becomes extremely small. Full statistical evaluations are complex and controversial and beyond the scope of this MOOC. It did not take a long from the introduction of DNA testing as a routine part of an investigation of crimes against a person for it to be realised that a DNA database would be a valuable tool to link crimes and identify suspects. Fingerprint databases had been doing this for almost a century by then, but DNA had the additional advantage of a close association between the evidence and the crime.

If a woman says she was attacked by a workman during a follow up call, and his fingerprints are found in the house, this is of little probative value. However, if his DNA is found on the victim, that is entirely different. At least 120 countries have some form of operational DNA database. The first was the National DNA Database, established in the UK in 1995. And the largest in terms of samples is the CODIS database in the US. There is still controversy about the population of databases in regard to what samples are entered as reference samples and what policies are implemented regarding retention times. Most agree that profiles from persons convicted of serious offences are kept permanently.

One of the surprising discoveries on implementation of DNA databases is the association between samples found in relatively minor crime scenes and hits with those in major offences, so-called cold hits. There is some information about this in the resources.

So far, whenever we have referred to DNA, we have meant human nuclear DNA found in homologous chromosomes. The two main other DNAs of forensic interest are mitochondrial DNA and Y-chromosome STRs. Other applications include using animal DNA, and there is some information about this in the resources. Mitochondria are organelles present in the cytoplasm of cells, where their main function is in energy production. They have many interesting features from an evolutionary biology perspective, but we will deal only with those of a forensic science application. Mitochondria contain DNA which display structural variations that are inherited, which means that it can be used as a genetic marker.

The main rule of mitochondrial DNA is coding for genes involved in energy metabolism, but there are two non-coding regions that display considerable variation in the nucleotide sequences, and the variations that are inherited along with the rest of the mitochondrial DNA. These regions, each consisting of around 700 nucleotides, are named hypervariable region 1 and hypervariable region 2. The analysis of the HVRs is similar to that for nuclear STRs in that it begins with PCR amplification, but the subsequent step is different and more complex. In comparing the mitochondrial DNA with nuclear DNA, the following observations can be made. The DNA molecule is a double-helix just as nuclear DNA is.

But it is in the form of a closed loop, and there are many fewer nucleotides in the mitochondria genome. Because there are hundreds to thousands of mitochondria in the cell, there is much more mitochondrial DNA than nuclear DNA. The inheritance line is from mother to daughter, which means that only one female relative needs to be traced to identify a body, valuable in mass disaster and other unidentified remains cases. The maternal line inheritance also means that it is not possible to come to a conclusion of absolute identity, since every female in the maternal line will have the same mitochondrial DNA.

The conventional use of mitochondrial DNA is in circumstances where there is no usable nuclear DNA and where the much higher amounts of mitochondrial DNA allow for profiling. Mass disaster victim identification is one application. Also where contact evidence is critical in a criminal investigation, mitochondrial DNA analysis is preferred to, for example, traditional microscopy. The main non-forensic application is in evolutionary biology and genealogy, including investigation of family lines and remains that are centuries old. The controversy over whether or not Anastasia, the youngest daughter of Czar Nicholas II, escaped death when the Bolshevik secret police executed the Romanovs is one of the more well known cases of this type.

DNA testing, including mitochondrial DNA analysis, demonstrated that there are no living relatives of the Czar’s family. The Y chromosome is also used for forensic purposes and is only found in males passed on from father to son. Like the other chromosomes, its DNA contains STRs, and these can be used in an analogous way to mitochondrial DNA.