Dr. Peter Vallone is a research chemist in the Biochemical Science Division of the National Institute of Standards and Technology (NIST). We asked him to tell us about some of his work developing and evaluating new technologies for human identity testing at NIST.
What is the main focus of your lab at NIST?
I work on a project for human identification in the Biochemical Science Division. Our project team is primarily focused on forensic DNA testing — forensic genetics. So we evaluate new technology platforms and new genetic markers. This is for the forensic community and companies or industries involved with forensic DNA typing. We don't do any casework. We're a research lab that's sort of between academia and the standard forensics labs that do routine testing. We do the research that standard forensics labs don't have the time to do, and then we pass that information along to the forensics community.
We have a team here at NIST, with John Butler as team leader — he's well-known in the community. It's fun and exciting to work with new technologies, new markers, and it's nice to feel we're doing something that gets back to the forensic community. A lot of forensics labs are so overburdened with casework that they don't have time to evaluate new assays and new techniques, and it's great to be able to do that and provide that information back to the forensics community. It's definitely a unique position.
Can you talk about some of the specific projects in the lab that you are doing?
One of the main things I do is help develop multiplex PCR testing methods for analyzing multiple genetic markers in a single reaction. One of the things we've done recently is working with miniSTRs — smaller amplicon size STRs for forensic identification.
Also SNPs: I'm looking at assays to analyze mitochondrial SNPs, Y chromosome SNPs, and autosomal SNPs. So we've been working on a lot of multiplex assays to develop those further in order to evaluate the markers that come out of our study.
I'm working with an array system for mitochondrial sequencing, and comparing that with dideoxy sequencing. I gave a presentation about that at the AAFS meeting (1). And we're also coming out with reference samples — quantitation standards for human DNA. That'll be useful for quantitative PCR assays that are out there. This would provide forensic scientists with a standard set of samples that have been quantitated here at NIST and that can be used for calibrating quantitative PCR.
But we also evaluate things like expert system software, which can be used for analyzing STR data. We're going to do some work evaluating DNA mixtures, creating different mixtures and see if we can come up with some guidelines for deconvoluting those mixtures because mixtures are important in forensic DNA analysis.
How do you go about making this information available to labs, to the public?
For the most part, like for genetic markers, it's published. So if somebody wants to reproduce it, that's fine. Or if a company wants to make a kit out of it, if there is a market for it, they can use those primers to start building their own testing method. There's a Web site that's maintained by our group called STRBase ( www.cstl.nist.gov/biotech/strbase/ ), and often we'll put information up there, like allele frequencies or information about the assays, as well as primer sequences and conditions for multiplex assays.
What do you see as the main challenges for your lab?
One of the main challenges now is multiplex PCR. Multiplexing is useful in reducing the complexity of a system: if you can look at 20 markers in a single reaction, obviously it saves you time, reagents, and sample, but it helps in terms of data analysis too — having all the data in one file in one format. But it's a challenge to build larger and larger multiplexes that are robust enough for genetic identity applications.
In more general terms though, forensic genetic testing is challenging because there are so many different aspects to it. You've got extraction, quantitation, PCR, different detection platforms, data analysis, new markers. So overall it's challenging just to keep up with advances in all of these areas. There's always new things being developed, and it's a challenge to keep up with each of those areas and give them our attention.
What do you see beyond STR testing as useful markers for genetic ID testing?
For the medium future, say 5 years or so, I don't really see anything supplanting STRs for human ID testing. But on the other hand, I think SNPs are important, and can be a powerful tool sometimes. For example, mitochondrial SNPs in the coding region can be useful for separating people who have the same control region sequence. In Caucasians, 7% have the same HV1, HV2 in the mitochondrial genome. We've done some work with the Armed Forces DNA Lab to develop coding-region mitochondrial SNPs to try and separate those individuals.
Another potential use of SNPs, that people are looking for now, is for eye color, hair color, pigmentation. And there are other more specialized applications. I know that there are panels of SNP markers for determining ancestry or population of origin. And that might be used more in the farther future, say 10 years or so. You could also use SNPs to derive kinship analyses. SNPs fill a role complementing the core STR tests that are being used now.
Besides DNA, what about non-DNA markers for human ID?
I am always interested in new forensic testing with RNA, like for tissue type designation. So if you get a sample you can perform an RNA test to tell what type of tissue it came from. That could be useful. Another non-DNA technology that is developing is new technology for facial recognition algorithms, which are becoming more and more important in analyzing biometrics.
References
1. Vallone, P.M., Jakupciak, J.J., and Coble, M.D. (2007) Abstract B184: Forensic application of the Affymetrix human mitochondrial resequencing array. Proc. Amer. Acad. Forensic Sci. 13, 140
