The early success of Operation Advance shows that techniques for characterizing human DNA have improved significantly during the past several decades. Forensic DNA analysis has evolved from an examination of cleaved DNA fragments to the characterization of target nucleotide sequences amplified from an infinitesimal quantity of DNA. As the Forensic Science Serviceís Alan Matthews told the BBC, ìInstead of looking previously for amounts of blood, perhaps the size of a coin, to even begin to start looking for DNA, weíre now looking at microscopic amounts which amount to a few cells.î Further advances in DNA forensics may shift analysis from the lab to the crime scene.

The RFLP Riff
The story of DNA profiling begins in the early 1980s with Dr. (now Sir) Alec Jeffreysí investigation of genetic stutters. The Leicester University geneticist had been interested in regions of human DNA - known as minisatellites - that contain core nucleotide sequences repeated end-to-end, often up to hundreds of times. The number of reiterations contained within any particular minisatellite varies among individuals. And this means that distances between the nucleotide sequences at either end of one of these genetic stutters also vary, a phenomenon called length polymorphism.

Here's an analogy. If a core nucleotide sequence is a boxcar, then a segment of tandem repeats would be a line of identical boxcars. And the number of boxcars - residing between the engine and the caboose - determines the length of the train. The trick is to visualize this length.

Jeffreys visualized length polymorphisms by treating DNA samples with restriction enzymes, which cleave DNA to produce fragments that contain a repeating segment. He then employed the molecular sieving technique of gel electrophoresis to separate DNA fragments by size. After transferring the separated DNA fragments from the gel to a nylon membrane, Jeffreys treated the membranes with a radioactive probe that binds with a selected repetitive nucleotide sequence. When the nylon sheets were placed against X-ray sensitive film, the positions of DNA fragments carrying the radioactive markers appeared as a series of lines that resemble bar codes. These bar codes have two important qualities: the pattern varies from person to person, enabling the characterization of an individual's DNA; and the patterns are inherited, providing information about family relationships.

In the spring of 1985, Jeffreys published the first report on his DNA fingerprinting method. The study came to the attention of Sheona York, a lawyer who represented a Ghanaian family of UK citizens entangled in an immigration dispute. The family's youngest son had recently returned from a trip to Ghana, but the Home Office detained him, alleging that he held a forged passport. York asked Jeffreys for his help. Jeffreys compared DNA from the boy, the woman who claimed to be his mother, and her other children. In this first application of DNA fingerprinting, the technique revealed that the boy was the womanís son and that all of her children shared the same father.

A year later, the Leicestershire Constabulary asked Jeffreys to analyze DNA evidence from two rape-homicides. Jeffreys knew, however, that he would need to find a different strategy to scrutinize DNA.

Why not apply DNA fingerprinting to the murder investigation? Among other things, the technique required large amounts of DNA, a significant constraint when dealing with crime scene samples. Also, DNA fingerprinting relies on multi-locus probes that detect multiple nucleotide sequences throughout the genome and on numerous chromosomes. While the complex pattern produced by multi-locus probes is ideal for characterizing a pure DNA sample, DNA fingerprint results are difficult to interpret if an evidence sample contains DNA from two or more individuals.

To overcome the limitations of the multi-locus probe approach, Jeffries devised a panel of single locus probes. When a single locus probe attaches to a target minisatellite region, it produces a pattern of one or two bands, representing nucleotide sequences inherited from the mother and father. The simple pattern created by a single locus probe is no longer specific to an individual. However, the probability that DNA from two people would produce the same DNA profile drastically shrinks if an investigator combines results from numerous single locus probes. For a criminal investigation, the single locus probe tactic also offered the advantage of requiring about 10 to 100 nanograms of DNA, or one-tenth the amount needed for a multi-locus probe examination.

During the course of the Leicester murder investigation, 5,000 men provided blood or saliva samples. The Forensic Science Service analyzed DNA from the ten percent who had the same blood type as the killer. Meanwhile, two men confessed to the crimes; DNA profiling exonerated one and confirmed the guilt of the other.

A Rising STR
The RFLP technique became the first scientifically accepted forensic DNA analysis method in the United States. But it had a few drawbacks: the single locus probe strategy still required relatively large amounts of DNA, and that DNA had to be recovered from a crime scene in a high quality form. DNA degrades when exposed to the environment, breaking into fragments that may be too small for RFLP analysis.

Researchers overcame these limitations in two stages. In the first, scientists applied the new polymerase chain reaction (PCR) technique to forensic analysis. The PCR method can duplicate, or amplify, a billion copies of a target nucleotide sequence, even a target sequence buried in a mixture of millions of other sequences. And PCR duplication can be performed with a nanogram of DNA.

But there was a hitch. PCR could duplicate DNA segments of about 1000 to 2000 nucleotides, a length much shorter than numerous minisatellite targets. Incorporation of PCR amplification into forensic DNA analysis required a switch to small targets that offered the same type of tandem repeat variation. And hereís where stage two comes in: the use of microsatellites. While minisatellites have sequences of six to 100 nucleotides that repeat two to several hundred times, microsatellites have repeat unit lengths that typically peak at seven nucleotides, repeated five to 100 times.

In 1991, Baylor College of Medicine's Thomas Caskey suggested using a type of microsatellite called short tandem repeats (STRs). Peter Gill, then a researcher at Britainís Forensic Science Service, devised the PCR-STR method, which enabled the simultaneous analysis of multiple STRs, a tactic called multiplexing. With this technique at hand, investigators could analyze DNA recovered from minute and even partially degraded samples. Forensic DNA analysis no longer depended upon the availability of tissues and fluids, such as blood and semen. Now, DNA analysis could be performed with PCR-amplified biological traces acquired from dental molds, cigarette butts, eating utensils, chewing gum, postage stamps, ski masks, licked envelopes, toothbrushes, razor shavings, band aids, and fingerprints.

The FBI favored the STR approach, and in 1997 the agency devised a standard series for DNA typing that consisted of 13 STRs, each with a sequence of four nucleotides. For example, the STR called D16S539 has the sequence "GATA." In 2000 the FBIís Bruce Budowle and his colleagues published their conclusion that the probability that DNA from unrelated individuals would generate the same 13 STR profile would be less than one in a trillion. That same year, the FBI discontinued RFLP analysis in favor of STR multiplexing.

The U.S. forensic science community has adopted the FBIís 13 STR series. According to Dr. Richard Saferstein, a forensic science consultant and former chief forensic scientist of the New Jersey State Police, the 13 STR array will remain the standard. One reason for this is that the FBI selected the 13 STRs to serve as the core genetic markers for its national database structure, the Combined DNA Index System Program. Dr. Saferstein also says that "DNA STR technology has become a mature science that will yield little in the nature of significant scientific breakthroughs in the next 5-10 years."

For now, forensic DNA analyses will remain focused on particular STR targets. But the strategy for duplicating those targets may soon change.

Taking the Heat Out of DNA Amplification
In 2000, the U.S. Justice Department predicted that, within ten years, investigators will be using portable, miniaturized instrumentation that will enable DNA characterization at a crime scene. This prediction may soon be realized thanks to a new strategy for amplifying DNA.

The polymerase chain reaction is initiated by heating DNA to separate the two strands of double-stranded target DNA molecules. The sample is then cooled in the presence of DNA primers that bind with selected nucleotide sequences. After reheating the solution to activate a polymerase, the enzyme extends the primers, synthesizing complementary copies of each DNA strand. Copies of target nucleotide sequences are generated exponentially by repeating these cyclic heating steps, a process that also binds DNA amplification to a lab setting.

Huimin Kong and colleagues at New England Biolabs (Beverly, MA) published a report in July 2004 on a technique that may liberate DNA amplification from the laboratory. In their helicase-dependent amplification method, helicase, rather than high temperature, uncoils DNAís double helix, enabling investigators to duplicate a DNA target at a uniform 37ƒC. Dr. Mark Griep, a helicase expert at the University of Nebraska-Lincoln, says that the New England Biolabs researchers ìhave eliminated the need for the energy-hungry heating-plate-type PCR thermocyclersî and have taken "a major step in the development of hand-held PCR devices for use in the field."

Helicase-dependent amplification requires some fine-tuning before miniature DNA analyzers enter the marketplace. Dr. Griep notes that the helicase technique duplicated a 123 nucleotide sequence fragment 10 billion times in three hours. "A useful hand-held device," he says, "would be able to amplify such a fragment in minutes, not hours."


Getting SNPy: A Different Tactic for Identifying Variation
RFLP and STR techniques focus on length polymorphisms. That is, only size matters. Yet about 90 percent of human genetic variation arises from the substitution of one nucleotide for another. Single nucleotide polymorphism (SNP) analysis focuses on these differences in nucleotide sequence. While STR analysis requires duplication of DNA fragments containing about 200 to 400 nucleotides, SNP analysis requires DNA fragments in the 50 to 90 nucleotide range. Consequently, the SNP technique can be used to analyze DNA that may be too degraded for STR analysis.

Despite this advantage, the Forensic Science Service's Dr. Peter Gill says that North American and European forensic science groups have taken the position that SNPs are unlikely to replace STRs in the near to medium future. This reluctance, he says, "stems partly from the fact that SNPs ìindividually have limited discriminating power - this means large multiplexes (about 50 or so) need to be constructed. Since typical assays are limited to analyzing about 12 SNPs per reaction, ìup to 5 different reactions would be needed, which is not feasible when there is very limited sample available."

Another reason why SNP technology wonít dominate forensic DNA analysis in the near future is that national DNA databases store STR data. Since STR and SNP analyses examine different regions of the human genome, a SNP profile cannot be searched against an STR profile. Conversion to a SNP-based system would require re-analysis of all DNA samples. Not only would such an effort be cost-prohibitive, but DNA samples are not routinely retained after analysis. Nevertheless, SNP technology has a place as an adjunct to STR analysis.

SNP analysis is not new. Alec Jeffreys examined SNPs in the late 1970s as a means to study inherited variation. His search for DNA regions more variable than SNPs led him to tandem repeats and, eventually, DNA profiling. Since then, DNA analysis has become an indispensable tool for criminal investigations. As David Coleman, chief constable of the Derbyshire Constabulary, declared in the Operation Advance press release, "the clear message to criminals is that there is no hiding place from the power of forensic science."