Important forensic DNA advance published in the journal Law, Probability and Risk

Dec 10 2009

Cybergenetics is pleased to announce the publication of an important forensic DNA advance in the journal Law, Probability and Risk. The paper "Match likelihood ratio for uncertain genotypes" enables scientists to extract far more identification information from the same DNA evidence. The prominent forensic statistician Dr. John Buckleton, Principal Scientist at ESR in New Zealand, considers this paper to be "a particularly elegant piece of work."

Every genetically distinct person has a unique genotype. However, evidence that is damaged, mixed or low-level may produce ambiguous data. Such zombie^™ DNA is currently analyzed by crime labs in ways that may discard considerable identification information. This information loss occurs because current DNA likelihood ratio (LR) match statistics focus on special cases where the genotype is assumed to have a definite value.

In the match likelihood ratio (MLR) paper, the authors embrace genotype uncertainty, and show how to tame it using probability. Lead author Dr. Mark Perlin, CEO of Cybergenetics, says that "Just as quantum mechanics extended the explanatory power of physics by treating particles as probability distributions, so too does MLR conserve DNA identification power by representing genotypes using probability." The MLR permits a simple match of these uncertain genotypes in intuitive ways that can be explained visually.

The paper provides a theoretical foundation, and gives examples of how the MLR typically preserves one million times more match strength than current analyses of zombie DNA. In Commonwealth v. Foley, the FBI had used a popular "inclusion" interpretation of fingernail homicide evidence that had a small amount of assailant DNA mixed in with the victim, reporting a 13 thousand LR. Using the TrueAllele^® Genetic Calculator to infer an uncertain genotype, and MLR to match it with the suspect, Dr. Perlin established a more accurate reading of 189 billion.

The National Research Council has criticized forensic science for lacking a sound statistical match basis. The MLR paper notes that "the continuing debate over DNA mixture interpretation and low-level DNA" has raised similar questions. The MLR advance enables forensic scientists to preserve the identification power of DNA evidence through genotype probability, and confidently present objective, rigorous and highly informative DNA match results in court.