Amerithrax case closed thanks to forensic genomics (and lots of other leg work)

I have just come across this report from the US Department of Justice, which documents all the evidence in the Amerithrax case and explains why they are closing the case. Anyone interested in bacterial pathogenomics should read the report, not just because success in solving the crime depended on forensic use of genome sequencing, but because of all the bizarre twists in the evidence trail and in the behaviour of the perpetrator of this horrific criminal act.

The story begins in September 2001, only a few days after the 9/11 attacks. Letters containing anthrax spores were sent by post to two US Senators and some news media offices. As a result, at least twenty-two people developed anthrax, with eleven cases of inhalational infection and five deaths. Cross-contamination of letters in the postal system meant that several people who had nothing to do with the intended targets were among those infected.

At first, the finger was pointed at Saddam Hussein, but the focus quickly shifted to a home-grown perpetrator, when molecular typing made clear that the strain used was a laboratory strain, the Ames strain, which was originally isolated in the US. I remember that forensic applications of whole-genome sequencing were discussed at a meeting I attended in Florida in November 2003. At that meeting I met Paul Keim who did the anthrax typing and Jacques Ravel who was doing the genome sequencing. The talk in the bar was that the epidemiological typing results delayed the start of the Iraq war by at least a year, because unless Saddam had been ruled out, the Bush Administration would have used the Amerithrax incident as a pretext for invasion (of course, they found other reasons anyhow).

Nonetheless, it took nearly five years more before the FBI closed the net on the perpetrator, a US anthrax researcher called Bruce Ivins. Genomics helped pinpoint single nucleotide polymorphisms associated with a batch of Bacillus anthracis managed by Ivins. Quite what genome sequencing was done is not revealed in detail in the report, but a key factor appears to be the presence of a unique combination of colony (and  SNP) variants in that batch.

This case sets an interesting precedent for use of bacterial genome sequencing for forensic and epidemiological purposes—something which is likely to become increasing common with the arrival of high-throughput technologies. But I would urge you to read the report for all the other stuff in there too. You could not make this up! The perpetrator’s obsession with codes and pictures of blindfolded women, working late in the lab and taking six hour drives just for fun. Paranoid delusions documented in prose and poetry. A fascination with the book Gödel, Escher and Bach (note the other writer on this blog, Nick Loman, has had my copy on long-term loan for far too long!). A hidden message in the Amerithrax letters written in the genetic code. This is John Grisham meets Michael Crichton, but all for real! A bacteriological Unabomber... but one caught by the application of bacterial genomics!