Research and Technology - Forensic Science Communications - October 2004
October 2004 - Volume 6 - Number 4
Research and Technology
Automating the Forensic Analysis of Nuclear DNA: The FBI's Research and Development Initiative
In the United States, there is a substantial backlog of samples requiring short tandem repeat (STR) DNA marker analysis. Approximately 200,000 to 300,000 collected convicted offender samples (U.S. Department of Justice 2003) and more than 540,000 evidentiary samples where there is no suspect (Attorney General's Report on the DNA Evidence Backlog 2004) currently remain to be analyzed nationwide. Additionally, 500,000 to 1,000,000 authorized convicted offender samples have not yet been collected (U.S. Department of Justice 2003).
Figure 1. Backlog of Convicted Offender Samples
To address this issue, Public Law 106-546, DNA Analysis Backlog Elimination Act of 2000, was enacted on December 21, 2000. This law authorized $170,000,000 toward reducing the backlog (DNA Analysis Backlog Elimination Act of 2000).
- $15,000,000 was appropriated in 2001 through 2003 for DNA analysis of samples to be included in the Combined DNA Index System (CODIS), the nationwide DNA database.
- $25,000,000 to $50,000,000 was appropriated in 2001 through 2004 for DNA analysis of samples from crime scenes and to increase the capability of public laboratories to carry out DNA analyses.
The intent of this spending was to eliminate the backlog of samples requiring STR DNA analysis by providing additional manpower and instrumentation for use with existing nuclear DNA analysis technologies.
Although existing nuclear DNA analysis technologies are valid and accurate, they are also labor-intensive and time-consuming. Introducing automation into the process flow for analysis of forensic biological samples would overcome the backlog problem and prevent its recurrence.
Identification of probative biological samples, the technical steps for typing the 13 core CODIS STR loci, and the interpretation of STR-analytical results and associated data quality review could all be automated. This would assist in achieving process quality and reproducibility.
During 2000 in support of the backlog reduction efforts, the FBI's Counterterrorism and Forensic Science Research Unit designed a research and development plan for automating the forensic analysis of biological evidence, which was funded by Congress in 2001. The FBI automation initiative is divided into three main areas—serology, the STR-typing process, and online data interpretation and quality assessment tools.
The goals for the serology initiative are to develop methods for the definitive identification of all forensically relevant biological stains and to automate the execution of these methods. Currently, biological samples deposited at crime scenes are identified by visual inspection, chemical reactions, enzymatic reactions, and standard immunological methods.
Figure 2. Serological Examination of Forensic Evidence
Both a presumptive test and a confirmatory test are performed in the process, and these tests are conducted sequentially, requiring a new sample for each test (Ballantyne 2000). However, definitive tests do not exist for each of the frequently encountered body fluids (e.g., saliva or urine). Operational efficiency could be improved if a system existed whereby a complete panel of body fluid identification tests was performed simultaneously from a single sample (i.e., multiplexed analysis), and the identification system was amenable to automation.
The FBI recently initiated a project to develop a multiplexed immunoassay to identify forensically relevant body fluids. Novel antigens not previously used in forensic analysis will be interrogated using monoclonal antibodies, the most specific immunological reagents available.
The immunoassays will first be developed in a format suitable for individual high-throughput automation on a robotic liquid handler. Subsequently, the individual assays will be multiplexed and adapted to an as yet undecided detection platform. One possibility is a suspension array based on flow cytometry (Kellar and Iannone 2002; Nolan and Mandy 2001).
A second project to identify body fluid under development by the FBI is the probing of messenger RNA (mRNA) species that are selectively expressed in cells that collectively comprise a particular body fluid (Juusola and Ballantyne 2003). Each cell type in the human body has a unique pattern of gene expression that is manifested by the presence and relative abundance of specific mRNA species, the molecular intermediate between genic DNA, and expressed protein (Caron et al. 2001).
Figure 3. Visual Inspection of Forensic Evidence
Studies have shown that mRNA is stable in biological stains of forensic relevance and can be recovered in sufficient quantity and quality for analysis (Juusola and Ballantyne 2003). Candidate tissue-specific genes have already been identified for saliva (statherin, histatin 3, PRB1, PRB2, and PRB3), but tissue-specific genes for other body fluids of forensic importance have yet to be identified (Juusola and Ballantyne 2003). When a panel of tissue-specific genes has been identified for body fluids significant in forensic investigations, a parallel analysis method will be developed to probe these mRNA species. One possibility is a microarray of complementary DNAs, or sequence-specific oligonucleotides, capable of recognizing each of the candidate genes in a "chip" format (Juusola and Ballantyne 2003).
Two parallel strategies are planned for automating the technical aspects of forensic nuclear DNA analysis. First, some of the manual steps in the STR-typing process will be replaced by automated systems performing these tasks or by new procedures that are less labor-intensive. Second, a completely dedicated, automated system for forensic genotyping will be developed. Improvements in the steps of the STR-analysis procedure can be developed in the short-term and introduced into casework within two to three years. Developing an integrated STR-analysis system is a longer-term project and estimated to take between three and five years.
Automating Individual Steps in the STR-Typing Process
STR-analysis involves the following four steps:
- Extraction of DNA from biological samples
- Quantification of the obtained DNA
- Amplification of the CODIS-required STR loci by the polymerase chain reaction (PCR)
- Separation and detection of the amplification products by capillary electrophoresis
The PCR amplification of template and subsequent production of DNA patterns by capillary electrophoresis already are semiautomated, as they involve little user intervention. Extraction and quantification of DNA remain to be automated.
Figure 4. Set-up of the Polymerase Chain Reactions for Amplifying the CODIS-Required STR Loci
Extraction and purification of nucleic acids from biological samples is one of the most difficult and tedious tasks in forensic DNA typing. DNA extractions from bloodstains, semen stains, buccal swabs, or liquid blood samples entail separating DNA molecules from proteins and other cellular material through a manual process.
For mixtures of epithelial and sperm cells, as in the case of vaginal swabs, the cells must first be separated prior to extracting the DNA. This is accomplished by simple chemistry. The swab is incubated in a detergent/proteinase K solution to lyse the nonsperm cells. Then the sperm nuclei are lysed by incubation in a detergent/proteinase K/dithiothreitol solution and purification of the DNA follows from each of the cell lysates. This procedure is known as differential extraction (Gill et al. 1985).
One of the FBI's research projects in this area is to build an instrument with a microfluidic platform that will perform differential and nondifferential DNA extractions. The core of this system will be an ultrasonic module for selective cell lysis and a silicon chip module for rapid DNA extraction, purification, and concentration.
Another project involves evaluating several DNA extraction procedures that are amenable to execution on a robotic liquid handler and programming a liquid handler to perform some of those processes. At the same time, a robotic liquid handler will be programmed to execute a proprietary differential extraction procedure that is based on differential lysis and selective filtration.
The detection and quantification of human nuclear DNA is another labor- and time-intensive step in the forensic STR-analysis procedure. The most popular method for quantifying human DNA in forensic laboratories today involves hybridization with a primate-specific alpha satellite DNA sequence located on chromosome 17, D17Z1 (Walsh et al. 1992).
Figure 5. Quantification of DNA by Hybridization to Human DNA-Specific Probes
In this procedure, a slot-blot apparatus is used to capture genomic DNA and a serial dilution of a human DNA standard on a nylon membrane. The primate-specific probe is added to the membrane and hybridization of the probe to DNA on the membrane is detected by chemiluminescence.
In contrast to this labor-intensive method, new PCR-based quantification assays provide automated detection and quantification of nucleic acid sequences. A human-specific quantification assay is being developed based on real-time PCR amplification of Alu sequences using TaqMan fluorescent chemistry (Holland et al. 1991; Lyamichev et al. 1993). As part of this assay development, three existing human-specific, real-time PCR-based quantification systems will also be evaluated (Applied Biosystems 2003; Nicklas and Buel 2003; Richard et al. 2003).
Integrated, Fully Automated System for Forensic Genotyping
A complete system for forensic nuclear DNA analysis could enhance user convenience and relieve analysts from performing repetitive tasks. The FBI is taking two approaches toward a totally automated system. One is to implement robotic liquid handlers. The other is to develop a miniaturized analysis system for forensic STR typing.
Robotic Liquid Handlers
Robotic liquid handlers will be programmed to integrate all of the steps of forensic nuclear DNA analysis, from extraction of DNA to profile analysis of the 13 core STR loci. Two robotic workstations are planned for this task.
Figure 6. Robotic Liquid Handler for Performing the Manual Steps of STR Typing
One of the workstations will perform DNA extractions, both differential and nondifferential, and set up real-time quantitative PCR assays. Using values from a real-time PCR system, this same workstation will also prepare additional plates of DNA solutions at a common concentration and set up PCR multiplex genotyping assays.
The second robotic workstation will prepare genotyping amplification products for capillary electrophoresis analysis on a genetic analyzer. In order to be compliant with quality assurance standards for forensic DNA testing laboratories, the two robotic workstations will be placed in separate rooms—a preamplification room and a postamplification room, respectively (Federal Bureau of Investigation 2000A; Federal Bureau of Investigation 2000B). Although this is not a fully automated system, the robotic liquid handlers will dramatically reduce operator involvement compared to the semiautomated steps described in the previous section.
Micro-Total Analysis System for STR Typing
A micro-total analysis system, also known as a lab-on-a-chip, is a small (centimeter-sized) chip that contains micron-sized channels for fluid transport and other design elements, such as pumps, valves, and reactors that miniaturize, integrate, and automate complex, multistep chemical and biological processes (Effenhauser and Mana 1994; Freemantle 1999; Lab-on-a-chip 2002; Manz et al. 1993; Meldrum 1999).
Compared to conventional laboratory instrumentation, micro-total analysis systems are virtually hands-free, consume extremely small volumes of samples and reagents, provide short sample-to-answer times, and are inexpensive and small. The devices are constructed from glass, quartz, silicon, or plastic using microfabrication technologies from the semiconductor industry (photolithography, micropatterning, microjet printing, light-directed chemical synthesis, laser stereochemical etching, and microcontact printing), or casting, cutting, and stamping techniques.
The FBI has taken two approaches toward developing a micro-total analysis system for forensic STR typing. The first combines all of the functions required for forensic nuclear DNA analysis onto one disposable chip. In this case, printed circuit boards will control the on-chip heating and cooling required for thermal cycling, and off-chip instrumentation will be used to achieve multicolor fluorescence detection.
Figure 7. Prototype Micro-Total Analysis System for STR Typing
The second approach for a STR-typing micro-total analysis system is a modular system composed of three microchips—a cell separation/DNA extraction chip, a multiplexed PCR chip, and a multichannel electrophoresis chip. Microfluidic interconnects will transfer material from one chip to the next, and off-chip systems will be used to actuate thermal cycling and for multicolor fluorescence detection. In both micro-total analysis system approaches, all required quality controls for STR typing will be accommodated on the chip along with the samples.
Online Data Interpretation and Quality Assessment Tools
An enormous bottleneck in the process flow for forensic STR typing is the interpretation of results and the associated data quality review. Each forensic laboratory has established guidelines for how these reviews are conducted. The guidelines include criteria to determine whether the results are of sufficient intensity and quality for interpretation, how to identify artifacts, and how to interpret mixed DNA samples and partial profiles. Most laboratories perform these functions manually, although software solutions are emerging (Perlin 2000; Perlin et al. 2001).
The FBI is developing expert-system software for forensic STR applications that will address single source and mixed samples. The software will determine STR DNA fragment sizes relative to an internal size standard, generate allelic designations, and translate the DNA signals into useful information in accordance with a laboratory's interpretation guidelines. The expert system will also assess the quality of the data, note any problems, and put the data into a format compatible for uploading into CODIS.
Figure 8. AmpFlSTR Profiler Plus STR Data Collected on an ABI PRISM 310 Genetic Analyzer
Outsourced Research Projects
The FBI's research and development initiative to automate forensic analysis of nuclear DNA is being executed primarily through external contractors. Table 1 lists the projects that are part of this research and development program.
Table 1: Research and Development Projects
Research and development for the automated nuclear DNA analysis initiative began in July 2003, and deliverables from these projects are expected in 2005 or 2006. The first deliverables will increase automation of current STR-typing procedures. The later deliverables will be fully automated lab-on-a-chip STR-analysis devices.
Nineteen state and local forensic laboratories are participating in this research and development effort through the FBI's Research Partnership Program (Counterterrorism and Forensic Science Research Unit 2004). Their involvement will include assisting in the evaluation of prototype instruments, contributing data for training the expert-software system, and conducting validation studies on the robotic liquid handlers for STR typing, an automated sperm-searching protocol, and the expert-software system.
For additional information contact
Kerri A. Dugan
Stephen T. Homeyer
Counterterrorism and Forensic Science Research Unit
Federal Bureau of Investigation
Applied Biosystems. Quantifiler Human DNA Quantification Kit User's Manual. Applied Biosystems, Foster City, California, 2003.
Attorney General's Report on the DNA Evidence Backlog, April 2004. Available: http://www.ojp.usdoj.gov/nij/pdf/dna_studyreport_final.pdf.
Ballantyne, J. Serology: Overview. In: Encyclopedia of Forensic Sciences. J. A. Siegel, P. J. Saukko, and G. C. Knupfer, eds. Academic, London, 2000, pp. 1322-1331.
Caron, H., van Schaik, B., van der Mee, M., Baas, F., Riggins, G., van Sluis, P., Hermus, M. C., van Asperen, R., Boon, K., Voute, P. A., Heisterkamp, S., van Kampen, A., and Versteeg, R. The human transcriptome map: Clustering of highly expressed genes in chromosomal domains, Science (2001) 291:1289-1292.
Counterterrorism and Forensic Science Research Unit. FBI visiting scientist program, Forensic Science Communicationshttp://www.fbi.gov/hq/lab/fsc/backissu/jan2004/shortcomm/ 2004_01_short01.htm. [Online]. (January 2004). Available:
DNA Analysis Backlog Elimination Act of 2000, Pub. L. No. 106-546, 114 Stat. 2726 (2000).
Effenhauser, C. S. and Mana, A. Miniaturizing a whole analytical laboratory down to chip size, American Laboratory (1994) 26:15-18.
Federal Bureau of Investigation. Quality assurance standards for forensic DNA testing laboratories, Forensic Science Communications [Online]. (July 2000A). Available : http://www.fbi.gov/hq/lab/fsc/backissu/july2000/codispre.htm.
Federal Bureau of Investigation. Quality assurance standards for convicted offender DNA databasing laboratories, Forensic Science Communications [Online]. (July 2000B). Available: http://www.fbi.gov/hq/lab/fsc/backissu/july2000/codispre.htm.
Freemantle, M. Downsizing chemistry: Chemical analysis and synthesis on microchips promise a variety of potential benefits, Chemical and Engineering News (1999) 77(8):27-36.
Gill, P., Jeffreys, A. J., and Werrett, D. J. Forensic application of DNA fingerprints, Nature (1985) 318:577-579.
Holland, P. M., Abramson, R. D., Watson, R., and Gelfand, D. H. Detection of specific polymerase chain reaction product by utilizing the 5' to 3' exonuclease activity of Thermus aquaticus DNA polymerase, Proceedings of the National Academy of Sciences of the USA (1991) 88:7276-7280.
Juusola, J. and Ballantyne, J. Messenger RNA profiling: A prototype method to supplant conventional methods for body fluid identification, Forensic Science International (2003) 135:85-96.
Kellar, K. L. and Iannone, M. A. Multiplexed microsphere-based flow cytometric assays, Experimental Hematology (2002) 30:1227-1237.
Lab-On-A-Chip: The Revolution in Portable Instrumentation. Technical Insights, Frost and Sullivan, San Antonio, Texas, 2002.
Lyamichev, V., Brow, M. A. D., and Dahlberg, J. E. Structure-specific endonucleo-lytic cleavage of nucleic acids by eubacterial DNA polymerases, Science (1993) 260:778-783.
Manz, A., Harrison, D. J., Verpoorte, E., and Widmer, H. M. Planar chips technology for separation systems: A developing perspective in chemical monitoring, Advances in Chromatography (1993) 33:1-66.
Meldrum, K. Microfluidics-based products for nucleic acid analysis, American Laboratory (1999) 31:20-22.
Nicklas, J. A. and Buel, E. Development of an Alu-based, real-time PCR method for quantitation of human DNA in forensic samples, Journal of Forensic Sciences (2003) 48:936-944.
Nolan, J. P. and Mandy, F. F. Suspension array technology: New tools for gene and protein analysis, Cellular and Molecular Biology (2001) 47(7):1241-1256.
Perlin, M. W. An expert system for scoring DNA database profiles. In: Proceedings of the Eleventh International Symposium on Human Identification 2000. Promega, Biloxi, Mississippi, October 2000. Available: http://www.promega.com/geneticidproc/ussymp11proc/content/perlin.pdf.
Perlin, M. W., Coffman, D., Crouse, C. A., Konotop, F., and Ban, J. D. Automated STR Data Analysis: Validation Studies. In: Proceedings of the Twelfth International Symposium on Human Identification 2001. Promega, Biloxi, Mississippi, October 2001. Available: http://www.promega.com/geneticidproc/ussymp12proc/contents/perlin.pdf.
Richard, M. L., Frappier, R. H., and Newsman, J. C. Developmental validation of a real-time quantitative PCR assay for automated quantification of human DNA, Journal of Forensic Sciences (2003) 48:1041-1046.
U.S. Department of Justice. The President's Initiative to Advance Justice Through DNA Technology, Fact Sheet, March 11, 2003.
Walsh, P. S., Varlaro, J., and Reynolds, R. A rapid chemiluminescent method for quantitation of human DNA, Nucleic Acids Research (1992) 20:5061-5065.