"As of April 2005, when the process was suspended, we had identified only 1,592 out of the presumed 2,749 who died," said Shaler, the former director of the department of forensic biology, OCME, New York City and the man responsible for overseeing the DNA identification of World Trade Center victims.
From the destruction of the Twin Towers in 2001 until his retirement in July 2005, Shaler oversaw New York City's efforts in identifying victims through DNA analysis. In the process of identifications, he and his colleagues pushed the edges of science and the technologies applicable to forensic investigations.
"We lacked the tools needed to do the job properly or efficiently," Shaler said in his recent book, "Who They Were: Inside the World Trade Center DNA story: The Unprecedented Effort to Identify the Missing," (Free Press 2005). He adds the staff had "No bar-coding capability, no laboratory information management system . . . to keep track of samples."
Worse than a laboratory not ready for the monumental effort of identifying the victims of this disaster were the problems inherent with the remains. While about 2,749 people died in the disaster, there were 19,915 individual samples to analyze, identify and return to victims' families. Of these, forensic scientists identified 54 percent. DNA analysis of one kind or another played a part in 89 percent of all identifications and was the sole method of identification in 86 percent of the cases.
Within the first weeks, it became obvious that many remains were commingled, that fire and water at the site had badly degraded many specimens, and that, as time progressed, the quality of DNA retrievable decreased dramatically.
Until Septenber 2001, crime labs across the country generally used STRs – short tandem repeats – in forensics work. STRs are biological markers in nuclear DNA that are inherited from both parents and vary widely among people. The Combined DNA Index System – CODIS – used by the FBI, uses 13 of these markers for identification purposes. Shaler's group used up to 16 markers. If sufficient markers matched known DNA samples, then there was an identification.
However, to do any matching, analysts needed DNA samples from the missing individuals including toothbrush, hairbrush and razor samples and DNA samples from close relatives. The New York City Police Department obtained these reference samples.
Collecting and recording large numbers of samples perfectly is difficult. Errors and omissions occur. Throughout the investigation, the OCME maintained contact with families of the victims and worked with them. The contact brought perspective to the project, friendships and, at times, emotional turmoil.
"Working with the families is critical to getting the job done properly, and the truth is that the medical examiner and the police need to establish a synergistic relationship from the outset," said Shaler.
Certainly in this emotional, heart-wrenching environment, patience and understanding were aided by familiarity and trust.
While samples of the living are replaceable, it quickly became apparent that degraded DNA made the use of STR by itself impossible in many victim identifications.
Shaler and colleagues decided to look at two other DNA methods -- mitotyping and SNPs. Mitotyping involves looking at mitochondrial DNA, DNA inherited only from the mother, and comparing it to a standard reference. While mitochondrial DNA is not unique -- entire maternal ancestry lines share the same mtDNA -- it, along with STR, could help in identification.
SNPs, or single nucleotide polymorphisms, are genetic variations that occur at a single DNA location. While standard practice requires only 13 STRs to identify an individual, 40 SNPs are necessary for an identification with the same statistical accuracy.
Although mtDNA analysis is uncommon, it has been used in forensic settings in the U.S. "SNPs represent a new adventure in testing for forensic scientists," says Shaler. No laboratory had used them day in and day out nor had they been used on compromised samples. Eventually, forensic scientists used all three methods of DNA analysis on the 9/11 remains. A separate version of CODIS, WTC CODIS, unconnected to the FBI, was set up; a version of BEAST – Bar Coded Evidence Analysis Statistics and Tracking – installed; and a variety of government agencies and corporations helped in the analysis and identification of the remains.
"In the instance of the World Trade Center attacks, first responders worked tirelessly for nine months," said Shaler. "In contrast, my staff would work doggedly for more than three years." Shaler suffered a heart attack during the identification period, personnel came and went, and companies and agencies assisted on projects and then dropped out of the program.
In the end, hundreds of people cooperated in the identifications. Not all individuals who died on Sept. 11 were identified, but the best possible effort given the circumstances and the level of technology at the time produced all the identifications that it could. In time, new techniques may identify more individuals.