2,795 people are thought to have died in the twin towers attacks.© GettyImagesNineteen months after the terrorist attacks on the World Trade Center, the effort to identify victims' remains goes on. Forensic scientists working on the project are being forced to break new ground. They have come up with innovative tests and ways to interpret the resulting data.
"The techniques developed for the World Trade Center will benefit the whole field," says forensic DNA expert John Butler of the US National Institute of Standards and Technology in Gaithersburg, Maryland. The new methods could be applied in war zones such as Iraq, Bosnia and Rwanda, and may help the identification of decomposed murder victims.
Researchers are working to match DNA samples from 14,249 body parts recovered from the wreckage of the twin towers to the 2,795 people who are thought to have died in the attacks. As of two weeks ago, 1,484 victims had been identified.
It's unlikely that everyone will be identified, says New York's chief forensic biologist, Robert Shaler. "At one time I hoped we were going to get 2,000. I'm not sure we're going to get that high now," he says.
The scale of the task has taken many by surprise. "I think at the beginning a lot of people expected that everyone would be identified in a few weeks," says forensic scientist Will Goodwin of the University of Central Lancashire, UK. "This has led to a realization of the limitations of DNA."
DNA samples
Researchers can compare DNA samples from bodies to those taken directly from the victim - from a hair- or toothbrush, for example - or from a family member.
Both comparisons have problems, says Charles Brenner, a consultant forensic mathematician based in Oakland, California. Many of the DNA samples thought to be from victims have turned out to be from someone else - it seems people are fairly relaxed about letting others use their toothbrush.
And standard DNA fingerprinting of relatives is not sensitive enough to cope with the number of victims. Coincidental sharing of genes between unrelated people can lead to bogus match-ups1.
“The scale of the task has led to a realization of the limitations of DNA”
Will Goodwin
University of Central Lancashire
"For every parent that has lost a child, there are likely to be quite a number of victims that are a plausible child of that parent," Brenner explains. There will be about one false relative among every 1,000 victims.
To solve this problem, Brenner recommended collecting DNA from several relatives of each victim. It's unlikely that one victim's DNA profile would erroneously match both the father and mother of another victim, for example.
But there are tens of millions of ways to match the more than 10,000 DNA profiles taken from victims' possessions and families to the thousands of unidentified samples. Brenner has written new software to search rapidly for the strongest matches from all of the possible combinations. The same approach to testing and analysis might help to identify the occupants of mass graves, Brenner believes.
Single letters
Fingerprinting is also limited by its requirement for intact stretches of DNA that are several hundred chemical letters long. Many twin towers body parts were too decomposed to yield such samples.
So Butler devised a test that works with shorter stretches about 100 letters long; it homes in on different regions of DNA.
“The techniques developed for the World Trade Center will benefit the whole field”
John Butler
National Institute of Standards and Technology
In addition, the team recently began looking for individuals' distinctive single-letter genetic changes, known as single-nucleotide polymorphisms (SNPs). These can turn up in runs of just 50 DNA letters.
SNPs are the focus of intense medical research, but are relatively untested in forensic applications. Results have been good so far, says Shaler. "I think it's going to work just fine," he predicts.
About ten new victims of the attacks are being identified each month. There is no set timescale for the project, but Shaler suspects that the effort may be wound down in around a year's time, when all of the latest tests have been run.
© 2018 Nature is part of Springer Nature. All Rights Reserved.
