Validation of a 21-locus autosomal SNP multiplex for forensic identification purposes.

A single nucleotide polymorphism (SNP) multiplex has been developed to analyse highly degraded and low copy number (LCN) DNA template, i.e. <100 pg, for scenarios including mass disaster identification. The multiplex consists of 20 autosomal non-coding loci plus Amelogenin for sex determination, amplified in a single tube PCR reaction and visualised on the Applied Biosystems 3100 capillary electrophoresis (CE) system. Allele-specific primers tailed with shared universal tag sequences were designed to speed multiplex design and balance the amplification efficiencies of all loci through the use of a single reverse and two differentially labelled allele denoting forward universal primers. As the multiplex is intended for use with samples too degraded for conventional profiling, a computer program was specifically developed to aid interpretation. Critical factors taken into account by the software include empirically determined extremes of heterozygous imbalance (Hb) and the drop-out threshold (Ht) defined as the maximum peak height of a surviving heterozygous allele, where its partner may have dropped out. The discrimination power of the system is estimated at 1 in 4.5 million, using a White Caucasian population database. Comparisons using artificially degraded samples profiled with both the SNP multiplex and AMPFISTR SGM plus (Applied Biosystems) demonstrated a greater likelihood of obtaining a profile using SNPs for certain sample types. Saliva stains degraded for 147 days generated an 81% complete SNP profile whilst short tandem repeats (STRs) were only 18% complete; similarly blood degraded for 243 days produced full SNP profiles but only 9% with STRs. Reproducibility studies showed concordance between SNP profiles for different sample types, such as blood, saliva, semen and hairs, for the same individual, both within and between different DNA extracts.

Sizing short tandem repeat alleles in capillary array gel electrophoresis instruments.

A 96-capillary array gel electrophoresis Applied Biosystems 3700 instrument has been used to analyse AMPF/STR SGM Plus short tandem repeat (STR) loci for forensic applications. This multiplex consists of ten STR loci plus the Amelogenin locus and currently forms the basis of the UK National DNA database that currently holds more than 1 million profiles. Of particular interest is the accuracy of allele designation that is determined by comparison with standard control allelic ladder markers. Some loci have higher standard deviations than others. In particular the high-molecular-weight HUMFIBRA alleles have high standard deviations of the order of 0.15 and it is these alleles that are most likely to be misdesignated. However, this risk is minimised by the analysis of at least five different allelic ladders across the array to estimate the mean size of each allele. In conjunction with this, a series of guidelines that can be programmed into expert systems are used to minimise risks of misdesignation. The efficacy of the procedures utilised are tested by computer simulation and demonstrated to be robust.

Validation of the AMPFlSTR SGM plus system for use in forensic casework.

The AMPFlSTR((R)) SGM Plus system is a commercially available STR multiplex produced by Applied Biosystems, a division of Perkin Elmer, Foster City, California, USA that supersedes SGM. The multiplex contains the six SGM loci, amelogenin and four additional loci. These additional loci are D3S1358, D19S433, D16S539 and D2S1338. Consequently, the match probability is significantly improved (conservatively quoted as 1 in 10(9) for reporting a full profile match). The system was subjected to validation. For example, ageing and degradation studies demonstrated semen stains to be the most stable evidence type, whereas buccal scrapes were the least stable. An apparent rise in the sensitivity increases the chance of obtaining successful results from the more difficult samples submitted for analysis. Two of the new loci (D3S1358 and D19S433) are low molecular weight (between 100 and 150 base pairs); this improved the success rates of the degraded samples where high molecular weight loci may drop out. Of 26 non-primates tested, four gave results that appeared as single peaks and were unlikely to cause interpretation problems. None of the 19 micro-organisms tested gave discernible results. Extensive casework and simulated casework studies demonstrated that SGM and SGM plus results were comparable. There was one example of a null allele (primer binding site mutation) recorded at the HUMFIBRA locus (in both systems). However, a concordance study of 1000 samples using both SGM and SGM plus did not demonstrate any differences in typing.