The man accused in Phoenix’s “Canal Killer” case may have his ancestors to blame for his 2015 arrest.

Megan Cassidy, THE REPUBLIC - Nov. 30, 2016

A new tool stops a predator

Angela Brosso went for a bike ride along the Arizona Canal in Phoenix, Arizona one evening in 1991. She never returned home. Her body was found a few days later in a nearby vacant lot.

A few months later, in early 1992, Melanie Bernas went for a bike ride along the same canal. She never returned either. Her body was found floating in the canal a few days later.

Over the years, numerous suspects were investigated for these murders without success. Although DNA evidence was collected from the crime scenes, there was no CODIS hit.

The Phoenix Canal Murders remained unsolved until 2014, when Identifinders International was asked by the Phoenix Police Department to compare the Y-DNA collected from the crimes to genetic genealogy Y-STR profiles posted on public websites. Using proprietary software to interrogate numerous online databases, a Y-DNA match was found to a small group of men with the surname Miller, narrowing down the list of suspects from about 2000 to just five, one of whom was a close match to the psychological profile developed by the FBI.

PPD moved quickly to obtain this man’s DNA profile. Using a ruse, undercover officers invited him to lunch and secured a DNA sample from his water glass. On January 13, 2015 his DNA was found to be a CODIS match to the DNA profile developed from the victims in the original investigation and he was soon arrested.

The police continue to investigate possible links between Brian Patrick Miller and other victims of as yet unsolved homicides. Two surviving victims have stepped forward and also identified him as their assailant. After many delays, including the defendant’s rejected attempt to throw out the DNA evidence, his trial is expected to begin in September, 2021.

Phoenix police investigating the crime scene in November 1992. (The Republic photo)

These murders brought the city to its knees. By the time we heard about Colleen, we had exhausted all investigative leads. And what she did – it was miraculous. Without Colleen getting involved this case would never have been solved.

DOMINIC ROESTENBURG, Detective, Phoenix Police Department

Analyzing paternal lineage through the Y chromosome

The Phoenix Canal Murders are only one example of how cold case investigations benefit from the use of forensic genealogical resources. There are many others that have drawn on the ~300k Y-STR profiles that have been posted online by the genealogy community in thousands of public databases on thousands of public websites. A match between a forensic Y-profile and a profile found in a genetic genealogy database can advance a cold case investigation in many ways. Although not foolproof, a surname is sometimes indicated. As is ethnicity and ethnogeographic background.

The use of public genetic genealogy Y-STR databases to solve the 1992-1993 Phoenix Canal Murders was a breakthrough in forensic identification. The resolution of the nearly 25-year-old murders and sexual assaults is the first known instance where direct-to-consumer (DTC) DNA testing data was used to solve a cold case. Although DTC testing companies such as Ancestry and FamilyTreeDNA refused access to their proprietary databases for forensic cases at the time (Ancestry still does), a hit was found among Y-STR test results voluntarily posted by DTC customers on a public genetic genealogy website.

Moreover, while familial matching can now be used to identify close relatives of a suspect – parents, siblings, children, cousins – the Phoenix Canal case is the first example of a cold case hit based on a family connection perhaps generations in the past. Because Y-DNA is co-inherited along the male line along with the family’s surname in most cases, a match between Y-DNA left at the scene of a crime and a Y-STR profile found in a genetic genealogy database can produce leads even if the family link is a few hundred years old.

The Phoenix Canal Murders appeared to have been random. Without the use of public DTC genetic genealogy databases to link his Y-STR profile to Miller’s last name, the case could have remained unsolved forever.