Comment
Greg Hampikian professor of biology and criminal
justice director, Idaho Innocence Project
Too much information G
alileo taught us that science proceeds more from what it has learned to ignore, than by what it takes into account, and that is the basis of all validation.
Nowhere is this more important than in forensic science, which is fast approaching single molecule resolution. To put this level of sensitivity in perspective, when I was in graduate school in the 1980s, we typically used 3μg of DNA for a ‘DNA fingerprint’. In 30 years, we have gone from needing a million cells to less than 10; and already some labs are reporting single molecule resolution. While this leap is impressive, the fact is that methods of evidence collection, handling, and storage have not improved nearly as much. Also consider that all modern forensic testing relies on the polymerase chain reaction, which doubles the number of molecules in a cell about 30 times, enough theoretically to produce about a billion copies of each DNA molecule in a sample. If a few molecules of this amplification, or perhaps even one, should contaminate a glove, lab solution, instrument or tube, it could result in time wasted on bad leads, a lost case against a true offender, or a wrongful prosecution against an innocent citizen. Does this happen? Yes. But no one has any idea how often, because there is no central repository for unexplained forensic results, and no national organisation tracks these issues. If
you think this is unnecessarily alarmist, consider the Phantom of Heilbronn, described by police as Germany’s most lethal woman. She left DNA at crime scenes all over Europe, and plagued international police for a decade. Her profile turned up on a swabbed teacup at a murder scene, on guns, and in syringes. After years of man-hours, it was finally discovered that the Phantom was actually a (still unidentified) worker at the Bavarian company that made swabs.
What is particularly telling about this case is that over the years no one
reported her profile in a control sample, which would have indicated that the phantom was a laboratory contaminant.
The truth is that someone likely did notice such a problem, but the profile was probably dismissed as one of the many unexplained
results that occur in crime labs all over the world. If we took the challenge of sensitivity seriously,
there would be a central repository of these unexplained results, where patterns would quickly emerge, and we would have some idea about the accuracy of ultra-sensitive DNA testing. The Amanda Knox case demonstrated another aspect of increased sensitivity. I worked with the defence team on that case, and we were able to show that a partial profile of the victim found on a knife blade was likely just contaminant or ‘unintended transfer’. Since Knox’s DNA was on the knife’s handle, this piece of evidence had played a crucial role in her original conviction. Fortunately, our work on the case was verified by two independent Italian DNA experts who were appointed by the appeal court judge in Perugia. The knife had no trace of blood, and was
Chemistry&Industry • November 2012
‘In the 1980s, we typically used 3μg of DNA for a “DNA fingerprint”. In 30 years, we have gone from needing a million cells to less than 10’
in fact covered in starch. More importantly, we pointed out that the crime lab looked far below the normal cut off for DNA evidence, straying into an area where they could not provide a validated protocol. They were not able to reproduce the result (reportedly because of a lack of material), had unsatisfactory quantification data, and never disclosed their standard operating procedures for this type of work. Was their DNA from the victim? Yes; was it transferred during a lethal wound to the victim? No. There were dozens of victim samples handled by the police and the lab, and we will never know how the victim’s DNA came to be associated with the knife, but the Italian DNA experts and the appeal court determined that it was likely a contamination event.
A final example demonstrates the problem of
sensitivity. I recently dealt with a case where a single sperm cell on a victim’s underwear was seen as evidence of sexual activity. Since an average ejaculate is 300m sperm cells, how significant is one?
Imagine if a swimmer enters a public pool carrying a semen sample on their person or suit. Since an Olympic pool holds 300,000 gallons of water, in order to get enough material for a DNA fingerprint in the 1980s, we would have needed 1000 gallons of pool water. Today, you would need a couple of ounces, and for those labs that can do single cell PCR, less than a teaspoon would do. Swimmers, is this too much information?
Find C&I online at
www.soci.org/chemistryandindustry
Victor De Schwanberg/Science Photo Library
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64