The Registers/Entry 4.05/One technique, and what it can and cannot carry
Familial searching
Searching for a partial match to find a relative rather than the person turns a register into something wider than it was authorised to be.
When the database doesn't match the suspect directly, investigators may search it for a near-miss — and that changes what a forensic register actually is.
A partial match is not an identification
A conventional database search looks for a full profile match: the crime-scene profile at every locus aligns with a stored profile, and the probability of a coincidental match is vanishingly small. Familial searching works differently. It looks for candidates who share enough alleles at enough loci to be plausible close relatives of the unknown contributor — and then uses that biological relationship as a lead, not a conclusion.
The logic rests on inheritance. A parent and child share one allele at every locus. Full siblings share both with roughly a quarter of loci, one at about half, and none at the remaining quarter. A partial match between a crime-scene profile and a database entry can therefore point toward a family, even when the person responsible was never sampled. Investigators then pursue conventional intelligence work — checking ages, locations, sex — to narrow the field before any DNA from the identified relative is sought.
From the register
How the search works — step by step
- Crime-scene profile generated and enteredthe full profile of the unknown contributor
- Partial-match search runcandidates ranked by number of shared alleles, filtered by sex and age
- Intelligence developmentinvestigators check whether a candidate's known relatives could fit the profile of the unknown contributor
- Voluntary or arrest sample soughtfrom the identified relative, not the database subject
- Conventional profile comparisonthe new sample matched against the crime-scene profile by standard methods; familial logic plays no further evidential role
There is no federal familial searching policy in the United States; the technique has been used in several states that have adopted their own authorising frameworks, Colorado being the first to formalise one in 2008. In England and Wales, the Forensic Science Regulator and the Home Office set out conditions under which the National DNA Database may be searched for partial matches, and every such search requires senior police authorisation.
The register becomes something else
The problem is scope. When a national forensic register was built, its legal and ethical justification rested on retaining profiles of people who had been arrested or convicted — a specific population, for a specific purpose, under a specific legal framework. Familial searching extends the investigative reach of that register to people who were never sampled: the relatives who happen to share alleles with someone already on the file.
This is the core tension. S. and Marper v. United Kingdom ↗, the 2008 European Court of Human Rights judgment, found that indefinite retention of profiles engaged the right to private life under Article 8 of the European Convention. Familial searching sharpens that tension further: a person on a database has had their rights considered, however inadequately. Their brother, who has never been arrested, has not. Yet if their brother's profile is inferrable — and it is, statistically — their presence on the register carries consequences for him.
From the register
Where it is authorised
- England and Wales
- Home Office and Forensic Science Regulator conditions; senior authorisation required per search
- United States
- no federal standard; state-by-state; Colorado first to formalise a policy (2008); California, Virginia and others followed
- Much of the EU
- either not authorised or not publicly documented as operational practice
The technique also introduces a structural bias. Populations that are historically over-represented in forensic databases are disproportionately affected: more people from those communities are indexed, so more of their relatives are reachable by a familial search. Research published through the National Academy of Sciences noted that forensic databases reflect patterns of policing, not patterns of offending, and familial searching multiplies that disparity across entire families who have committed no offence.
What it actually establishes
A familial search result is an investigative pointer, not evidence. It identifies candidate relatives; it does not place anyone at a scene, it does not establish contact, and it cannot carry a case on its own. DNA from the identified relative must be obtained — through elimination sampling, through intelligence development, or through a new crime-scene exhibit — and that profile must be interpreted by conventional means before anything evidential is produced.
The interpretation step carries its own demands. Mixture analysis, secondary transfer, and the distinction between depositing and being present all still apply, exactly as they would in any other case. A familial hit that leads to a voluntary sample that matches a crime-scene profile is still a profile-to-profile comparison, and every limit of that comparison still holds.
What changes with familial searching is not the science — the allele-counting, the statistical framework, the instrument — but the population the science is operating on. A register authorised to index one group is being used to investigate a much larger one. That is a policy question, not a laboratory question, and the two should not be confused.