Predicting phenotype from genotype
Published November 2016
It is increasingly possible to predict an organism’s phenotype (observable physical properties including behaviour), or aspects of an organism’s phenotype, by studying their genotype (complete set of genes). The focus here is on using this technique in humans.
Are there recent scientific, legal or social developments?
Genetic ancestry testing has been around for a number of years and involves looking at either the Y chromosome, mitochondrial DNA or single nucleotide polymorphisms (SNPs) genome to estimate a person’s ethnic background in a way that can be broken down into percentages e.g. 50% African, 50% European. A newer technology, DNA phenotyping uses a DNA sample to predict physical traits of the individual and build up a physical likeness of them. Researchers are building databases using information from volunteers to look for connections between facial shape and patterns of SNPs to enable them to then ‘reverse engineer’ a face from a DNA sample. Genome-wide association studies (GWAS) look for associations between SNPs and certain phenotypic traits across the whole genome. As well as looking at physical traits such as freckles or hair morphology these studies can look at particular diseases and offer a promising way to study complex, common diseases (such as diabetes or Parkinson’s) in which many genetic variations contribute to a person’s risk. Furthermore, forensic DNA phenotyping may eventually be able to make probabilistic predications about mental or behavioural traits.
There have been many recent developments in genetics, genomics and molecular biology which are likely to make human forensic case work more reliable in the near future. The UK Border Agency (UKBA) ran the Human Provenance Pilot (HPP) Project between Feb 2009 and March 2010 which aimed to investigate whether ethnicity testing might be applicable to the work of the UKBA, in particular in cases where it was suspected that asylum seekers were lying about their country of origin. The project was discontinued following a review.
Are there complex ethical issues?
Such technologies offer huge potential to help solve crimes and protect individuals, to better understand health and disease or to help identify missing persons or disaster victims. However there are concerns that we might come to falsely rely upon their conclusions. Much of the accuracy of this technology depends on the quantity and quality of the databases being used for comparison. This raises the question of whether and how more samples should be attained, and how to ensure representative data across populations. Ensuring that people who volunteer their genetic and phenotypic information to such databases are able to give truly informed consent may be difficult, especially as the research possibilities continue to expand exponentially. The need for large datasets also needs to be balanced with concerns about the security of the data being stored, and questions of privacy. There are risks associated with having an unbalanced number of samples from one sub-group of the population. The council’s report on bioinformation cautions that inferring ethnic identity from biological samples risks reinforcing racist views. Data generated by DNA phenotyping tests are merely predictions and indicators of probability for phenotypic traits. This raises the concern that the results will be misconstrued, or taken/presented as more definite that they are, with the specific worry of them being used to justify the targeting of specific racial groups.
There is a key ethical distinction between testing known and unknown samples. If law enforcement agencies perform intrusive tests on unknown samples, and these ultimately become known individuals, then there are issues of consent, privacy and autonomy to consider. These individuals are unable to consent to the use of their DNA and previously unknown or concealed traits may be discovered, such as whether a person has had plastic surgery or has changed genders, or they may reveal unexpected information relevant to the individual’s health.
Is there a potential policy impact?
Further regulation will be required to determine how these new technologies should be applied within human forensics or immigration services.
Is it a subject of public concern?
There was widespread criticism of the HPP project from the Human Genetics Commission and the wider public. This included concern about the accuracy and validity of the scientific techniques involved (which fail to determine which political borders a person has come from) as well as issues of consent and the use of a vulnerable test group. Issues concerning the control of immigration are particularly prominent within the public psyche at this time, especially in the wake of Brexit.
Is the consideration timely?
In April 2010 The Crime and Security Act introduced several changes to the law, namely that DNA profiles of non-convicted individuals can only be kept for a maximum of six years (three years for under 18s) and DNA profiles of volunteers should no longer be added to the database.
Can the Council offer a distinctive contribution?
There is likely to be a fair amount of overlap with the bioinformation report as well as other Council reports on genome editing, emerging biotechnologies, genetics and human behaviour, medical profiling and online medicine, and genetic screening. The Council may, however, be able to offer a freshly updated contribution on the issues in forensics, in the light of the more recent technological developments.
Possible future work topics
This is one of the topics that have been suggested as possible project areas for further investigation by the Council. These topic summaries do not aim for comprehensiveness; rather, they are intended to sign-post some of the key considerations and to provide a starting point for discussion. Each summary includes links to relevant publications on the topic.