Analysis and interpretation of short tandem repeat microvariants and three-banded allele patterns using multiple allele detection systems

The Palm Beach County Sheriffs Office (PBSO) Crime Laboratory and the Alaba ma Department of Forensic Sciences (ADFS) have validated and implemented an alysis of short tandem repeat (STR) sequences on casework using silver stai ning kit anti SYBR(R) Green I detection systems and are presently validatin g fluorescently tagged STR alleles using the Hitachi FMBIO 100 instrument. Concurrently, the Broward County Sheriffs Office (BSO) Crime Laboratory is validating the ABI Prism310 Genetic Analyzer capillary electrophoresis STR detection system (ABI CE310) from Perkin Elmer Applied BioSystems. During t he course of analyzing over 10,000 individuals for the STR loci CSF1PO, TPO X and THO1(CTT) using silver staining for allele detection, 42 samples demo nstrated alleles that were ''off ladder,'' contained three-banded patterns at a single locus, or exhibited an apparent THO1 "9.3,10'' allele pattern. PBSO, ADFS and BSO Crime Laboratories have collaborated on the verification of the allele patterns observed in these 42 samples using the following al lele detection systems: (1) manual silver staining, (2) SYBR(R) Green I sta ining, and/or (3) fluorescently tagged amplified products separated by poly acrylamide gel electrophoresis or capillary electrophoresis followed by las er detection. Regardless of the CTT allele detection system utilized, conco rdant results were obtained fur 41 of the 42 samples. The only exception wa s a sample in which a wide band within the THO1 locus was identified as a T HO1 "9.3, 10" genotype by silver staining lilt and SYBR(R) Green I staining but was verified to be a THO 1 "9.3" homozygote by all other allele detect ion systems. Manual allele detection could readily identify microvariants, as a visual assessment of stained gels clearly shows that alleles do not mi grate coincident with well-characterized allele size standards. As would be predicted, however, the manual detection systems did not provide adequate resolution to approximate the basepair size for off-ladder variants. All fl uorescent software program-systems were consistent in designating alleles " not in range'' or "off ladder,'' thereby indicating true microvariants. All single-locus three-banded patterns were detected using all of the STR mult iplex systems. In addition, individual locus-specific primers verified mult iplexed amplified products were specific for the locus in question.