The True Story of Two Women Using DNA to Solve America’s Most Puzzling Cold Cases

The True Story of Two Women Using DNA to Solve America’s Most Puzzling Cold Cases
Design by Steve Miluch

They called the dead man Joseph Newton Chandler, but that wasn't his name. He lived alone in an efficiency apartment in a nondescript complex in Eastlake. He rarely ventured far from home and the closest thing he had to a friend was a former coworker at a chemical factory he once listed as an emergency contact on his employment forms.

One day in July 2002, Chandler purchased a handgun — a .38-caliber Charter Arms revolver. The old man returned home. He locked his door and windows, turned off the air conditioner. Then he stepped into to his bathroom, faced the mirror, and put the barrel of the gun in his mouth. The last thing to enter his mind, other than that bullet, may have been the dark secret he was leaving behind.

By the time anyone noticed he was missing, Chandler's body had decomposed in the summer heat and become a gorging ground for maggots. It was impossible to lift fingerprints. Police reviewed paperwork from the rental office, searching for next of kin. He'd listed a sister, Mary Wilson, who lived at 1823 Center St., in Columbus. But the address led to an empty lot. It was important they find some relative: He'd left $82,000 in the bank.

The court appointed his coworker as the executor of his estate and enlisted a private investigator named Mike Lewis to track down Chandler's family. Lewis used Chandler's birth certificate to identify his parents and then cross-checked birth records in Buffalo, where Chandler was born, and in that way located a sister. She was rather surprised to hear the news because her brother, Joseph Newton Chandler, had died in a tragic car accident at age 8 along with their parents, four days before Christmas in 1945. The dead guy in Eastlake was not the real Joseph Newton Chandler.

Whoever he was, he'd carefully assumed the dead boy's identity over several years — first, by obtaining a copy of Chandler's birth certificate in Buffalo and then by using that document in 1978 to get a Social Security number in Rapid City, Iowa, before moving to Ohio. He got a job as a draftsman for the Lubrizol chemical company and earned a reputation there as a quiet eccentric — he liked to listen to the sound of static while he worked.

The Eastlake police ran down every lead they had but the case went cold. U.S. Marshal Pete Elliott heard about the mystery in 2014 while looking for some men who had escaped Alcatraz. Perhaps Chandler was actually one of these fugitives, he thought. Certainly, the man was running from something — why else would he take his true name to the grave?

Elliott located hospital records for the old man. One detailed a trip to the E.R. in 1989, when the man had been treated for lacerations on his penis, which he claimed were caused by having sex with a vacuum cleaner. Not long before he committed suicide, the erstwhile Joseph Newton Chandler was diagnosed with cancer and the hospital collected a tissue sample. They still had it. Elliott sent it to the lab. Unfortunately, the DNA was too degraded to be of much use to law enforcement. It seemed like the old man's true identity would never be found.

Then, on June 21, 2018, Pete Elliott called a press conference in the jury assembly room at federal courthouse in Cleveland.

In the news footage, there's an amused smile on Pete Elliott's lips as he steps to the podium to address reporters. He has a whopper of a story for them: The true identity of the man known as Joseph Newton Chandler has been discovered. But not by any cop.

"Now it is my honor to introduce the ladies from California," he said. "Dr. Colleen Fitzpatrick and Dr. Margaret Press not only put us in the ballpark with Joseph Newton Chandler, they told us the exact seat he was sitting in and then told us who paid for the ticket and the person who paid for that ticket was Mr. Robert Nichols. It's my honor to introduce the two doctors."

Elliott stepped away and Dr. Fitzpatrick came forward. She can hardly be seen behind the podium. "I want to thank the Marshals for trusting us to work on this case," she said, her face all smile, her glasses perched on top her short, wavy hair. She didn't open with an anecdote or charming story. Instead, she dove headlong into a particle-level explanation of Y-chromosomes and the proprietary software used to make sense of it all. Fitzpatrick and her partner had come up with a new way to catch bad guys.

Fitzpatrick and Press at a press conference
Fitzpatrick and Press at a press conference

* * *

Colleen Fitzpatrick has a vest with 42 pockets, and before she explains how she changed forensic science forever, she wants to explain what each pocket is used for. We're at her favorite Persian restaurant — Darya, in Costa Mesa, California. She's just come off a plane from Cleveland, a day after Elliott's press conference, and the vest made traveling easier.

"It's a SCOTTeVEST," she says. "It was on Shark Tank, the TV show? It will change your life. Look, you can fit an iPad in this pocket and here's a pocket you can store your credit cards because it blocks RFID signals so nobody can scan for your information, people do that now, and this pocket is for your glasses and here, the piece de resistance ..." She opens a pocket to reveal a card-sized piece of cloth with diagrams sketched on it. It's a map to all her pockets so she knows where everything goes and where everything is.

Colleen grew up in Lakeview, an upper-middle-class section of New Orleans that drowned in Katrina. She was the oldest of four and her memories of the Big Easy from the '50s and '60s conjure a world that doesn't exist anymore. A safe, cozy place in a time before Mardi Gras became a bacchanal. Her family had the only pool in the neighborhood and her mother let the Red Cross train kids to swim there. Her father sold silk flowers, wholesale and retail, in a large shop and designed and hung Christmas decorations for businesses in the winter.

From an early age, Colleen exhibited savant-like behavior. She was proficient in French by the time she was 4. She skipped eighth grade. And she was always working on some fantastical project.

"Colleen would hold little fairs at our house," recalls her sister, Bebe. "Like county fairs, you know? With booths and games for kids. One year, she made a spook-house in the garage. You sat in a wagon and she pulled you through."

The entire family enjoyed thought-provoking games. There were long matches of Clue with the extended family, and her grandmother was the reigning champion for many years. Her father taught them to play Hearts.

The Fitzpatrick kids liked to collect Mardi Gras beads in the French Quarter and along Bourbon Street, and it was always a challenge to see who could procure the most without being indecent. One year Colleen returned with great piles of beads.

"She came up with such a great idea," says Bebe. "She made this giant clown face, with a big open mouth for a target. She stood on the sidewalk and as the parade came by, she challenged everyone to throw their beads in its mouth."

Though nobody gave a name to it, it was clear Colleen was different. She thought differently, saw the world differently. She assessed the world in a series of win/loss challenges, looking for any advantage.

"People expected her to do something fabulous when she left," says Bebe.

When it came time for college, though, funds were tight. Much of her father's income was going to care for her dying grandmother. Colleen had her mind set on the physics program at Rice University, but she knew she'd probably end up at Loyola. As the enrollment deadline approached, Colleen focused on her science fair project: She was conducting experiments with a Benham disk. An illusion first discovered by an English toymaker, a Benham disk is a circle with a specific black-and-white pattern of curved lines that, when spun, appears to create bands of color. Different people see different colors, and to this day nobody really understands how it works. Young Colleen set out to discover what was going on. She ran a number of tests and constructed variations of the Benham disk to see if it had any impact on the colors produced.

The day her enrollment paperwork for Loyola was due, a telegram arrived at Colleen's house. Her science fair project was one of 10 in the country selected for an award by Tomorrow's Scientists and Engineers, a program sponsored by Humble Oil. The prize was $6,000 toward the college of her choice.

"My life is a series of events like this," says Colleen. "I'm a scientist, but science can't describe everything. We cannot describe ourselves scientifically. How do I describe why we are here? I believe in intuition, in being directed down a certain path because of events like this. Is there a master puppeteer directing our path through life? Or do we engineer it all ourselves? I don't know."

Colleen studied physics at Rice and earned her Ph.D. from Duke University. Along the way, she became fascinated with an illusion of a different sort: holograms. She taught herself holography, the method of creating a three-dimensional image on a two-dimension surface of glass. It was art; it was science. And it required great patience and focus: The glass had to remain completely still during the process. If it wavered even a billionth of a meter the image would not develop. She built a holography lab in her garage and it took her five years to get it right.

Colleen lectured at Sam Houston State University for a couple of years but found she didn't care for teaching. She left for a job with Rockwell International, a defense contractor out of Seal Beach. There, Colleen worked on the laser radar system used in Ronald Reagan's Star Wars initiative. Later, at Spectron Laser Systems, she was the principle investigator on the LITE Laser for the space shuttle, an experiment that used lasers to identify chemicals and pollution in the atmosphere. In 1989, she founded her own company, Rice Systems, Inc. She continued her work on the use of lasers for high-resolution measurements and bid on contracts with NASA, the National Science Foundation and DARPA. She was working on the next spacecraft to Jupiter when funding was redirected to a return voyage to the moon by the Bush administration in 2005. As a kind of release, Colleen then began writing a book about her other passion: genealogy. Colleen knew there was a way to apply science to genealogy in a fashion that had not really been done before. Hers was the very first book on the subject, according to Forensic Magazine.

Part of genealogy involves the examination of old photographs in order to identify subjects and time periods. At the time, most investigators were focused on the details of the clothing worn by the subjects in the pictures, trying to identify the manufacturer and time period when a blouse or pair of pants had been produced. But that method was very imprecise: People could keep the same pair of pants for years. So instead, Colleen studied the thickness of the cardboard the photographs were printed on. She researched the history of cardboard and learned how the thickness increased over time. She also researched the aspect ratios of the photos, which reveals the size of the negatives the photographer used, which in turn identifies the particular brand of camera that took the photo. She built a website —— and blogged about her findings.

In 2006, Colleen was contracted by Hebron Investments, an international business that located people whose homes had fallen into escrow due to unpaid taxes. She worked 75 cases, locating people in 30 countries, sometimes using Google Translate to communicate. She found them all save two.

After that, Colleen spent some years identifying Holocaust survivors who had been taken from their families during the war. She helped expose fraudulent authors who claimed to be survivors of the Holocaust. And she got really good at identifying people using archival records and DNA. The techniques she was using and honing were about to be deployed on a new question: the identity of a serial killer.

* * *

On the morning of Nov. 9, 1992, a young woman's nude, headless body was discovered near a bike path along the Arizona Canal in Phoenix. Police were quick to connect this body to the report of a missing woman that had come in the night before: Angela Brosso, 22, had gone for a bike ride along the canal path and never returned home. It was Angela's birthday. Her bike was never located. Her head was found 11 days later in a canal grate.

Ten months later, 17-year-old Melanie Bernas went for a bike ride along the Arizona Canal and never returned. Her body was found in the water the day after she disappeared. She'd been stabbed to death. Her bike was also gone. A serial killer had come to Phoenix.

And for 22 years, the man remained unidentified.

In 2014, the 25th International Symposium on Human Identification was held in Phoenix and several local detectives attended the event to learn about new techniques that could be used in forensic identification. Colleen Fitzpatrick was there, speaking about her attempts to identify the extended family of Abraham Lincoln's mother.

Colleen offered to meet with the police to explain a new way they might use DNA to find Angela and Melanie's killer. An officer picked her up and brought her to the station. If there was any chance this woman had a new trick up her sleeve, the detectives were interested.

"These murders brought the city to its knees," says Detective Dominic Roestenberg. "We hadn't seen anything that violent. By the time we heard about Colleen, we had exhausted all investigative leads. And what she did ... it was miraculous."

What Colleen did was find a needle in a factory of haystacks. The state lab had tested the suspect's DNA and isolated his Y-DNA profile. Y-DNA is very important to genealogists. Every person has 23 pairs of chromosomes, a mix of genetic information from their mother and father. The 23rd pair, however, is either two X-shaped chromosomes for a girl or an X and a Y-shaped chromosome for a boy. The Y-shaped chromosome is passed from father to son — like a surname. If a person's Y-DNA profile can be sequenced and compared to other profiles in a large enough database, it's possible to figure out what that person's last name likely is.

There are a number of websites with large, public databases of Y-DNA profiles, which are used by genealogists to trace lineage. There are huge websites like, but also smaller, individual sites that focus on specific surnames, like the Smith Project, the Jones Project and the Fitzpatrick Project. In fact, Colleen knew there were over 200,000 publicly available Y-DNA profiles and she had the software to mine them.

click to enlarge Joseph Newton Chandler
Joseph Newton Chandler

When she compared the canal murderer's Y-DNA to these publicly available profiles, she got eight matches, all with the same last name.

"She gave us a last name, Miller," says Roestenberg. "We did a follow-up and found his name in our files. We found a suspect with that name who we had contacted years earlier." Detectives had talked to Bryan Patrick Miller 20 years earlier, when he was 19. At the time, he denied involvement and was dismissed. Now he became their No. 1 suspect.

Of course, they needed to confirm that they had the right man. So the detectives surreptitiously collected Miller's DNA (since the criminal case is still pending, they won't say how exactly they did this) and sent the sample to the lab so the genetic material could be compared to the DNA left at the crime scene.

Days later, as the detectives gathered for a weekly meeting, an analyst from the lab called. Could they come over? Like right now?

"They literally walked over," says Roestenberg. "That never happens. The lab employees never just walk over to the police department to personally deliver information. And they came in and said, 'We got him.'"

Roestenberg was so shocked, all he could say was, "Are you sure?" To which the lab technician replied, the chances they were wrong was one in several quintillion. "Yeah, it's definitely him. I do believe I shed a tear."

It was 8 p.m. and Fitzpatrick was at dinner with a friend when the detective called. "He said, 'You were right, you were right! His name's Miller!'"

The next day a SWAT team arrested Miller, now 43-years-old, for the murders of Angela Brosso and Melanie Bernas. He goes to trial soon and police say he is a suspect in several additional cold case murders.

"Without Colleen getting involved, I feel strongly that this case would never have been solved," says Roestenberg. "I don't know if you ever get closure, but, yes, it was good for the family."

Thanks to Colleen, police had been given a new tool to crack cold cases, and with her track record growing, law enforcement had a new secret weapon. Without meaning to, Colleen Fitzpatrick had become a post-modern Sherlock Holmes. And every Holmes needs a Watson.

* * *

Before she found genealogy, Margaret Press fell in love with linguistics. This was at Berkeley, in the late 1960s — where the culture was evolving day by day. Margaret's mother was an activist in Pasadena and she'd taken Margaret to political protests as a girl, where she had witnessed the power of words in the signs and chants used to fight racism and McCarthyism. By the time she got to Berkeley, revolution was the word of the day. Not just in the streets but within the study of linguistics as well. A young rockstar of a linguist, Noam Chomsky, was challenging long-held beliefs about the origins of language. Chomsky proposed that the capacity for language was innate in humans, that we are born with an understanding of the structure of language that is not learned — as unlikely a thing as a child who never played chess knowing where each piece goes at the beginning of his first game.

"Suddenly the planets were going around the sun, not the earth," says Margaret, recalling the excitement she felt reading Chomsky's theories for the first time. "This universal structure of language has to be hard-wired inside us. And it is. Like our DNA. If you work backward from any language, you will see where it diverges from a common, shared language hundreds of years ago." Margaret explains how in most IndoEuropean languages, the word for "father" begins with either a "p" or an "f." Like a mutation on the Y-chromosome, which letter they use lets you figure out which languages are related to each other today.

For her dissertation, Margaret compiled a descriptive grammar of Chemehuevi, a nearly-extinct Native American language. After grad school, she found work writing in the language of computers and worked in software development until she retired two years ago. The software work was always meant to supplement her other passion — writing murder mysteries. Margaret is the author of several fiction and nonfiction books based in Salem, Massachusetts. In 1996, she published Counterpoint, which detailed her research into the real-life murder of Martha Brailsford.

It was another mystery writer, though, who inspired Margaret to test the boundaries of forensic science.

In 2000, the novelist Sue Grafton learned about a cold case from Santa Barbara as she was casting about for a mystery she could use for the next book in her Alphabet series. In 1969, a young woman's body was found by hunters in a quarry, her murder never solved, her identity never recovered. There were scant clues. She'd had buck teeth and new fillings. She wore bell-bottoms and horseshoe earrings. Grafton became enamored of the case and used it for the basic plot of Q is for Quarry.

In 2001, Grafton paid to have the woman's body exhumed. DNA was collected. A facial-reconstruction was done. But none of the leads panned out.

Margaret read about the case and wondered if she might be able to help. She knew that adoptees in the United States were having luck tracking down their parents by having their DNA sequenced by companies such as Ancestry and 23andMe. Why couldn't that same technology be used to identify the Santa Barbara Jane Doe? So she contacted the sheriff's office. "They wouldn't return my calls," she says. "They thought I was a crank, like the psychics who call and say they know who the murderer is. A genealogist calls and says I can solve your case? Okay, right."

What she needed was credibility. Colleen had street cred. She was already working with law enforcement on similar cases. So Margaret reached out to Colleen on Facebook and laid out her plan: She wanted to get the dead woman's genetic profile from the detectives in Santa Barbara and upload it into the 23andMe database. It would surely kick up a relative match, a second cousin in Boise or something.

But Colleen and others said such a thing would be impossible. Ancestry and 23andMe do not allow users to upload their own genetic profiles. They test the samples sent in by subscribers themselves and then use proprietary software to find specific markers known as single nucleotide polymorphisms, which you'll see written as "SNPs", commonly called "snips." These markers are then compared to other profiles in their database.

“They thought I was a crank, like the psychics who call and say they know who the murderer is. A genealogist calls and says I cansolve your case? Okay, right.”

tweet this

They brainstormed ideas: Perhaps they could attend a symposium where executives from the genetic companies would be present and convince them that they should allow personal uploads.

And then a hobbyist figured out a workaround. It was a member of their Facebook genealogy group: He'd submitted a DNA sample of his dead father (gathered from a tissue sample collected during a hospital stay) and had its entire genome sequenced. That sequencing included not just the Y-DNA that many genealogists relied upon to trace lineage, but also the DNA from the other 22 chromosome pairs inherited from both parents — what's called autosomal DNA. And while the private ancestry companies don't allow uploads, they do allow users to download a text file that contains coded information about specific markers that were identified in their DNA. That text file can then be uploaded to a public repository of autosomal DNA on the web, called GEDmatch. The GEDmatch software then looks for matches between all their users' data sets. Using a program to isolate these same snips from the whole genome data, a file could be created and uploaded. It worked. The man succeeded in creating a 23andMe file for his father on his own. GEDmatch located the man's relatives, and Colleen and Margaret found their solution. The same strategy could be used to identify any John or Jane Doe.

"It was the right moment, the perfect storm," says Margaret. "The public databases were getting bigger, and the genetic testing was getting cheaper. Suddenly, entire-genome testing was affordable." Margaret still holds out hope that one day the private companies will see the benefit of working with them too. "We'd have access to another 15 million profiles if Ancestry and 23andMe allowed clients to upload data. We could solve these cases lickety-split."

Grafton died last December. The Santa Barbara Jane Doe has yet to be identified. But Margaret hopes for an answer soon. "I would have loved to tell her we're going to do this case."

From that moment on, Colleen and Margaret have spoken nearly every day. Colleen contacts law enforcement and Margaret manages a volunteer army of genealogists who gather to discuss leads in their private Facebook group.

"She's the nuts and bolts," says Colleen. "I'm more of the marketing end. I'm more the one with the contacts, who knows the agencies, knows how to speak to them. Our personalities are very compatible."

Together, Colleen and Margaret founded the DNA Doe Project, a nonprofit organization that uses autosomal DNA to identify John and Jane Does. The project also helps raise funds for agencies that want to use their services but who can't afford it.

Over the years they proved, through numerous and varied cases, that the DNA Doe Project worked. "This was unprecedented," says Colleen.

But there was a growing concern that others in the genealogical community would view what they were doing as an invasion of privacy. Each time they solved a new case, it brought more attention. Eventually, GEDMatch would have to decide what was permissible. And what would happen to their efforts to identify their John and Jane Does if the database they relied on suddenly changed its terms of service? What if it simply shut down? "If GEDmatch rolled up the sidewalk, we were toast."

Margaret sighs. "And then the Golden State Killer happened."

* * *

It seems callous to reduce the evil of this man to a paragraph, but there's simply too much to give every last horror its due.

His crimes began with burglaries (about a hundred), which advanced to rapes (at least 50), and then to murder (at least 13). His killing fields were the suburbs of Sacramento, Modesto, Visalia. The serial killer's reign lasted from 1974 to 1986, and then he went silent. Some suspected he had died. The police have given him many names over the years — The East Area Rapist, the Original Night Stalker, the Golden State Killer. DNA linked all of these crimes together but the perpetrator remained unknown: Police found no match in CODIS.

Paul Holes, a retired investigator for the Contra Costa District Attorney's office, consulted with his own forensic genealogist, a woman named Barbara Rae-Venter, a retired intellectual-property attorney. They compared the Golden State Killer's DNA with the profiles on GEDmatch, and identified possible relatives. From there they narrowed down the list to a couple possible candidates, which pointed them to Joseph DeAngelo, a former police officer from Citrus Heights.

Eighty-year-old Curtis Rogers was in bed the morning of April 25, reading the news on his computer, as is his routine. That's when he saw the breaking news — the Golden State Killer was caught — and the police had used his website to do it.

"I could feel my hair turn gray," says Rogers. In another life, he worked in international business, servicing clients like Quaker Oats and Mennen. He's retired now, but not less busy. GEDmatch was his idea. But he'd never envisioned where it would lead him. "I think I sat there and just tried to absorb the meaning of what was happening. I looked outside. There were antennas. There were reporters at my door."

Rogers was faced with a dilemma. What was to be done about GEDmatch? What responsibility did he have? "I was very concerned," he says. "Were we violating the privacy of our users, allowing these forensic genealogists to compare a suspect's DNA to their profile? At the same time, here was this killer that made Jack the Ripper look like a choirboy. I did a lot of thinking. The ethics are very bothersome."

“I think I sat there and just tried to absorbthe meaning of what was happening.I looked outside. There were antennas.There were reporters at my door.”

tweet this

Rogers is not a crime fighter by choice. He was a member of the Rogers surname project, and claims to be the third cousin, three times removed, of Winston Churchill. Rogers had become frustrated by having to make so many calls to members of other surname projects in order to trace back his family tree. So, out of necessity, he came up with a simple idea: What if he were to build a website that could serve as a central clearinghouse for autosomal DNA data?

It began in 2010 and he named it after GEDCOM (Genealogical Data Communication), a data format used for comparing DNA profiles that was developed by the Church of Jesus Christ of Latter Day Saints. Genealogy is a big part of the Mormon faith because of their belief in baptism-by-proxy. If they can specifically identify a person who died, no matter how long ago it happened, they can be saved if a member of the church is baptized in their place. Their mission is to save us all.

GEDmatch was built so genealogists could analyze their DNA, not to catch criminals. But in the end, the ends justified the means. "The dam was broken with the Golden State Killer," says Rogers. "There's no going back. What we need to do is educate our users."

GEDmatch will remain up and open, he says, and can be used to track serial killers and rapists. It's a new era for DNA.

"The police were used to things that were easy," says Rogers. "Black and white. Fingerprints are either a match or not. DNA matched a suspect or it didn't. Now, DNA is no longer black and white, no longer yes or no. It's more complicated now. The police have to use people like Colleen, experts at finding people. Law enforcement can't do it alone right now because they simply don't know how, but maybe in the future they'll train their own experts."

Currently, there is nothing regulating this new field of investigation, and for now, anyway, the new frontier of forensic genealogy is wide open. Colleen has begun accepting criminal cases through her private company, Identifinders International, and the DNA Doe Project remains dedicated to giving names to John and Jane Does.

* * *

U.S. Marshal Pete Elliott had heard about Colleen's work over the years and asked if she could help him on one of his mysteries. The lab had already isolated the man we knew as Joseph Newton Chandler's Y-DNA profile. In that profile were segments of DNA where the nucleotides (the A-G-T-C bases you learned about in grade school) repeated. Everyone's DNA has these small, repeating bits of genetic code, in different places — what's called a "short tandem repeat," or STR. Colleen compared the STR profile generated on the case to hundreds of thousands of Y-DNA profiles stored online at places like Out of all those available profiles, only one matched all 17 markers. That person's last name was Nicholas.

This line of Nicholases descended from a colonial immigrant, George Nicholas. George's son was the first treasurer of the Colony of Virginia. It was a prominent family with old ties going back centuries. There were literally thousands of descendants.

click to enlarge Robert Ivan Nichols
Robert Ivan Nichols

So Elliott, at Colleen and Margaret's request, resubmitted the DNA to the lab for full genome sequencing. When the data came back, Colleen uploaded the file to GEDmatch, where the new information was compared to the other autosomal DNA files in its database. This identified hundreds of distant relatives.

"We were working with fumes here," says Colleen. So they requested that the lab test the DNA extract one more time — a risky gamble, since this would destroy the remaining DNA the lab had for this man. Once it was gone, there could be no more tests. The lab came back with a slightly different data set this time. That's when Margaret suggested they combined two files. "I explain it like having two photographs where the resolution is not great and then overlaying them to fill in the face in the picture," says Margaret. This time GEDmatch identified several relatives they'd never seen before. And one new match led them to a woman named Schreiber who had four sons with a man whose last name was Nichols.

Three of the boys had passed away. The fourth, Robert Ivan Nichols, was still alive. The DNA Doe volunteers scoured social media for more information about the family. They searched public records and phone numbers. And one very smart sleuth noticed an odd coincidence. She'd found that the Nichols family home was at 1823 Center St. in New Albany, Indiana. When Chandler had filled out his rental agreement, he'd listed 1823 Center St., Columbus, as an address.

"It wasn't a coincidence," says Elliott. "Liars always lie close to home."

Soon, Elliott stood on the doorstep of man named Phil Nichols, who he now believed was the mystery man's son. "I drove down with another Marshal and knocked on his door. When he opened the door, it was like Joseph Newton Chandler was standing there. They looked so much alike. I knew we had our guy."

Of course, that was only half the mystery. Elliott knows the man's real name, now, but he still doesn't know what Nichols was running from.

What he's been able to gather from family stories and scarce public records is that Nichols served on the USS Aaron Ward, which was bombed by the Japanese during World War II. He suffered wounds in battle and burned his uniform when he returned home. He left his wife and kids shortly after, telling his wife, "In due time, you'll know why." Phil says the last contact he had with his father was in 1965, when he received an envelope postmarked from Napa, California, with nothing inside but a single penny. Marshals found the elder Nichols had resided in Dearborn, Michigan, in 1964 before making his way out to the San Francisco area in 1965, the same year his parents reported him missing.

"Typically, when people are hiding, when people are scared, it's typically fugitives on the run," Elliot said at the press conference earlier this year. "There's a reason he went missing and assumed the name of a deceased 8-year-old boy in 1978 and hid for so many years. There's a reason he never again contacted his family. Robert Ivan Nichols never wanted to be found throughout his lifetime and into his death, and someone out there may hold the key as to why." 

Fitzpatrick thinks the answer will be more mundane and personal.

"I think it's more likely we're dealing with a man who suffered from PTSD from the war, which caused paranoia," she says.

Her role in the case is done, but there's plenty more lined up. Now that police know what a forensic genealogist can do, her phone rings incessantly. "I'm getting calls and emails every day. I'm working on a dozen cases."

Her house is not large, a couple of rooms and bedrooms upstairs. She sublets one of her rooms to a friend. For her dream job to survive, the DNA Doe Project will need funding.

No matter what happens, Colleen and Margaret have each other's back. "When you work these cases, you feel that insidious gloom and it can affect you," says Margaret. "We rely on each other and our team and we focus on the excitement of being the ones to solve a case. And we are. We're solving them Doe by Doe by Doe. It's a jungle of bad out there, but we're chopping through it."