The Netherlands Forensic Institute (NFI) can do more and more with minimal and complex traces, and the technological possibilities in the field of DNA continue to increase. The NFI has now developed a new software program that makes comparisons with the DNA database even faster and more detailed. "This is a new milestone," says scientific researcher Jerry Hoogenboom. "The program is going to provide results in even more cases. The new search method also offers opportunities for cold cases in which DNA has been preserved."
The new program is called ProbRank. The program uses a new way of computing that is even more accurate than previous programs because it takes in more data. A DNA profile obtained from a forensic trace is represented by a series of peaks, with each peak representing a feature on the DNA. ProbRank, in addition to previous methods, incorporates not only the location of the peaks but also the peak heights into the calculations. "That's going to help when analyzing and comparing complicated traces whose features are missing, or traces that contain DNA from multiple individuals," Hoogenboom said.
"If a person has DNA characteristics that you see reflected in the trace DNA profile, the program also looks to see if the heights are easily explainable," Hoogenboom explains. When, in a mixed DNA profile, one person has contributed more DNA than the other, their peaks are different heights. "Basically, the more DNA, the higher the peaks. ProbRank takes into account in the calculation whether the peaks belonging to the same person also match in height. This allows the new method to detect matches even more accurately."
Now there are cases where when comparing a complex trace with the DNA criminal database, no persons are found because too many combinations are possible with the peaks. That's because many people contributed DNA to a trace. "In those cases, there is now more chance of finding a person who actually contributed DNA, because ProbRank also includes the highs and whether they match. That's very informative."
The new method also saves time and operations, says software designer Martin Slagter. He and his colleagues helped Hoogenboom program the new calculation program. ProbRank replaces two programs the NFI now uses to perform comparisons and calculations, namely SmartRank and DNAStatistX.
With SmartRank, the NFI has been searching the DNA database since 2017. The software compares all profiles in the DNA database with the characteristics of the DNA profile of a trace found at a crime scene or, for example, a piece of clothing. DNAStatistX then calculated the probability of obtaining the DNA profile of the trace if the person from the DNA database actually contributed to a trace, compared to if an unknown random person instead of that person from the DNA database had done so. "We call that probability 'the probative value,' and only when it's high enough does the expert mention the person's name in the report," he said. ProbRank merges what the programs SmartRank and DNAStatistX do," Slagter says.
ProbRank will be introduced this summer to the Fast ID line, in which the NFI has automated the entire DNA process from A to Z. Police and prosecutors in Rotterdam use the Rapid ID Line and by default receive automatic answers after only three days as to whether DNA found at a crime scene provides a match with the DNA database or not. "The Rapid ID Line now only reports the preliminary results of the person who contributed relatively much to a trace, the main donor. Thanks to ProbRank, the Rapid ID line will soon also be able to compare the DNA of persons whose DNA is less in the trace with the DNA database and indicate whether a person has been found, with the corresponding probative value," Slagter said. The NFI is now in the validation phase.
How did the new calculation method come about? Hoogenboom laughs, "In 2021, we made adjustments to DNAStatistX that made the program much faster. For example - at first jokingly - the question arose as to whether it would now be fast enough to be able to calculate the probative value with a mixed DNA profile of a trace in a case for all 400,000 people in the DNA database.... I did some programming one Friday afternoon. It turned out to work. So I had created a first proof of concept, but it needed some neater programming. It was a project to move forward with."
Slagter and his team hooked up and the rest is now history. Slagter beams, "Sweden has already indicated they would like to use it soon. That means ProbRank will also be used outside the Netherlands. That makes me incredibly proud."
