STR Analysis of Artificially Degraded DNA—Results of a Collaborative

Forensic Science International 139 (2004) 123–134

STR analysis of artificially degraded DNA—results of a collaborative European exercise Peter M. Schneidera,*, Klaus Bendera, Wolfgang R. Mayrb, Walther Parsonc, Bernadette Hosted, Ronny Decortee, Jan Cordonnierf, Daniel Vanekg, Niels Morlingh, Matti Karjalaineni, C. Marie-Paule Carlottij, Myriam Sabatierk, Carsten Hohoffl, Hermann Schmitterm, Werner Pflugn, Rainer Wenzelo, Dieter Patzeltp,Ru¨diger Lessigq, Peter Dobrowolskir, Geraldine O’Donnells, Luciano Garafanot, Marina Doboszu, Peter de Knijffv, Bente Mevagw, Ryszard Pawlowskix, Leonor Gusmaõy, Maria Conceicao Videz, Antonio Alonso Alonsoa1, Oscar Garcıá Fernańdeza2, Pilar Sanz Nicola´sa3, Ann Kihlgreena4, Walter Ba¨ra5, Verena Meiera6, Anne Teyssiera7, Raphael Coquoza8, Conxita Brandta9, Ursula Germanna10, Peter Gilla11, Justine Halletta12, Matthew Greenhalgha13 aInstitut fu¨r Rechtsmedizin, Universita¨t Mainz, Mainz, Germany bKlinische Abteilung fu¨r Blutgruppenserologie, Universita¨t Wien, Wien, Austria cInstitut fu¨r Gerichtliche Medizin, Universita¨t Innsbruck, Innsbruck, Austria dInstitut National de Criminalistique, Bruxelles, Belgium eLaboratory for Forensic Genetics and Molecular Archaeology, Leuven, Belgium fChemiphar, Brugge, Belgium gInstitute of Criminalistics, Prague, Czech Republic hDepartment of Forensic Genetics, Institute of Forensic Medicine, University of Copenhagen, Copenhagen, Denmark iNational Bureau of Investigation, Vantaa, Finland jInstitut de Recherche Criminelle, Rosny-Sous-Bois, France kLaboratoire de Police Scientifique, Toulouse, France lInstitut fu¨r Rechtsmedizin, Universita¨tMu¨nster, Germany mBundeskriminalamt, Wiesbaden, Germany nLKA Baden-Wu¨rttemberg, Stuttgart, Germany oLKA Rheinland-Pfalz, Mainz, Germany pInstitut fu¨r Rechtsmedizin, Universita¨tWu¨rzburg, Wu¨rzburg, Germany qInstitut fu¨r Rechtsmedizin, Universita¨t Leipzig, Leipzig, Germany rLKA Sachsen, Dez. DNA-Analytik, Dresden, Germany sForensic Science Laboratory, Dublin, Ireland tReparto Carabinieri Investigazioni Scientifiche, Parma, Italy uIstituto di Medicina Legale Universita` Cattolica, Roma, Italy vForensics Laboratory, University of Leiden, Leiden, The Netherlands wRettsmedisinsk Institutt, University of Oslo, Oslo, Norway xInstitute of Forensic Medicine, Medical University of Gdansk, Gdansk, Poland yIPATIMUP, Universidade do Oporto, Oporto, Portugal zInstituto de Medicina Legal, Universidade de Coimbra, Coimbra, Portugal a1Instituto National de Toxicologia, Madrid, Spain a2Area de Laboratorio Ertzaina, Bilbao, Spain a3Instituto National de Toxicologia, Sevilla, Spain

* Corresponding author. Present address: Institute of Legal Medicine, Am Pulverturm 3, D-55131 Mainz, Germany. Tel.: þ49-6131-3932687; fax: þ49-6131-3933183. E-mail address: [email protected] (P.M. Schneider).

a4National Laboratory of Forensic Science, Linko¨ping, Sweden a5Institut fu¨r Rechtsmedizin, Universita¨tZu¨rich, Zu¨rich, Switzerland a6Instiut fu¨r Rechtsmedizin, Universita¨t Basel, Basel, Switzerland a7Institut Universitaire de Me´decine Le´gale, Universite´ de Gene`ve, Gene`ve, Switzerland a8Laboratoire AMS, Lausanne, Switzerland a9Institut Universitaire de Me´decine Le´gale, Lausanne, Switzerland a10Institut fu¨r Rechtsmedizin, Universita¨t St. Gallen, St. Gallen, Switzerland a11Forensic Science Service Headquarters, Birmingham, UK a12LGC, Teddington Middlesex, UK a13Orchid Bioscience Europe, Abingdon, UK Received 21 May 2003; received in revised form 2 October 2003; accepted 3 October 2003

Abstract

Degradation of human DNA extracted from forensic stains is, in most cases, the result of a natural process due to the exposure of the stain samples to the environment. Experiences with degraded DNA from casework samples show that every sample may exhibit different properties in this respect, and that it is difficult to systematically assess the performance of routinely used typing systems for the analysis of degraded DNA samples. Using a batch of artificially degraded DNA with an average fragment size of approx. 200 bp a collaborative exercise was carried out among 38 forensic laboratories from 17 European countries. The results were assessed according to correct allele detection, peak height and balance as well as the occurrence of artefacts. A number of common problems were identified based on these results such as strong peak imbalance in heterozygous genotypes for the larger short tandem repeat (STR) fragments after increased PCR cycle numbers, artefact signals and allelic drop-out. Based on the observations, strategies are discussed to overcome these problems. The strategies include careful balancing of the amount of template DNA and the PCR cycle numbers, the reaction volume and the amount of Taq polymerase. Furthermore, a careful evaluation of the results of the fragment analysis and of automated allele calling is necessary to identify the correct alleles and avoid artefacts. # 2003 Elsevier Ireland Ltd. All rights reserved.

Keywords: Short tandem repeat; Alleles; PCR degradation

1. Introduction Experience with degraded DNA from casework samples shows that every sample may exhibit different properties in Genetic analysis of short tandem repeat (STR) loci is this respect, and that it is difficult to systematically assess the currently the most powerful and widely used method to performance of routinely used typing systems to analyze identify the contributor of biological samples collected in the degraded DNA samples. To learn more about the efficiency context of criminal investigations, and the identification of of STR systems, a standardized reference sample of human remains. The successful application includes a vari- degraded DNA in sufficient amounts would be quite helpful, ety of substrates such as blood, saliva, epithelial cells, hair and could also be used for validation studies of new STR roots and even compact bone samples [1,2]. However, when typing systems [5–7]. Therefore, a collaborative exercise on more problematic samples such as old stains or decomposed degraded DNA was planned in the process of the ‘‘Standar- tissue are subjected to STR typing, failure to obtain repro- dization of DNA Profiling Methods in the European Union’’ ducible results may occur due to degradation of high mole- (STADNAP) network, and participation was offered to 50 cular weight DNA [3]. forensic DNA laboratories across Europe. The intention of In crime case investigations, degradation of human this exercise was to better understand the different para- DNA usually is the result of a natural process resulting meters influencing STR typing results obtained from from the exposure of the stain or tissue samples to the degraded DNA as well as to learn more about the strategies environment. Light, humidity, elevated temperatures as applied by the participating laboratories to optimize the well as bacterial and fungal contaminations followed by typing of problematic samples. the growth of these microorganisms lead to physical, chemical and biochemical degradation of genomic 2. Materials and methods DNA. Once the average DNA fragment length is reduced to sizes smaller than 300 bp, loss of genetic information 2.1. Production of degraded DNA samples may occur due to the lack of suitable template DNA and the subsequent failure of frequently used STR typing Batches of high molecular weight genomic DNA were methods [4]. prepared from two human cell lines, HepG2 and P118, with P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134 125 male and female genotypes, respectively. The DNA samples 2.3. Data analysis were degraded under standardized conditions to an average fragment length of less than 200 bp using a combination of The submitted data were recorded for each STR locus physical and biochemical methods, i.e. sonication and treat- based on the peak height (in relative fluorescence units, rfu) ment with DNAse I. Based on previous experiences from a for each allele observed. To allow a comparison between smaller exercise among the STADNAP partner laboratories laboratories, the peak heights were grouped into three using a single degraded DNA sample (data not shown) it was categories: strong signal: >150 rfu, low signal: 150– intended to achieve a slightly higher degree of degradation in 30 rfu, very low or no signal: <30 rfu. Furthermore, two sample B compared to sample A. The degradation process was categories of data were defined based on the exercise closely monitored to control the resulting fragment sizes as instructions (see above): ‘‘standard’’ PCR conditions using well as the suitability to obtain partial results by multiplex 0.25–2.5 ng degraded DNA and 28 PCR cycles, and STR typing. It was observed that the final fragment size ‘‘enhanced’’ PCR conditions using 0.5–5 ng DNA and distribution heavily depends on the DNA quality and con- 28–36 PCR cycles. Unexpected results, artefacts or incorrect centration, the reaction conditions and volumes, as well as the alleles were recorded separately for each locus and were not properties and the concentration of the DNAse I digestion. included in the peak height analysis. Batch to batch reproducibility could only be achieved by repeatedly testing the degraded DNA using STR analysis [8]. 3. Results and discussion 2.2. Exercise design Results were received from 38 laboratories from 17 Two aliquots of 20 ml each were shipped to the 50 European countries (Fig. 1). All laboratories submitted data participating laboratories with the following instructions: for at least one multiplex STR kit. The total number of datasets for each STR kit included in the study are summar- Apply your standard combination of forensic STR typing ized in Fig. 2. The majority of laboratories (63%) used the systems using regular PCR conditions. Profiler SGM Plus kit from Applied Biosystems (ABI). A If your first amplification fails, try to modify these con- total of 26% of the laboratories used the PowerPlex 16 kit ditions as you would do in casework, or use singleplex from Promega, and another 29% used the Profiler, Profiler systems. Plus and Cofiler multiplexes from ABI. In addition, 29% of Record the results for each locus with successful ampli- the laboratories submitted data for other commercial kits or fication by allele designation and peak height in a tabu- ‘‘home made’’ singleplex STR systems, and 21% of the lated form, and enclose prints showing the DNA profiles laboratories performed a DNA sequence analysis of the using the GeneScan or GenoTyper format. hypervariable D-loop regions of mitochondrial DNA. The If you routinely perform mtDNA typing, please try to majority of laboratories (58%) carried out capillary electro- sequence HV-I and HV-II. phoresis using the ABI Prism 310 Analyzer (Fig. 2, columns In addition, the participating laboratories were asked to on the right). provide further information about equipment, routine typing The correct typing results for the DNA samples A and B procedures, PCR conditions and rules for data analysis, e.g. are listed in Table 1 for all STR loci comprised in the SGM the threshold for allele calling. Plus, Profiler Plus and PowerPlex 16 multiplexes, including

Fig. 1. Number and country of origin of the participating laboratories. 126 P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134

Fig. 2. STR multiplex kits and sequencing equipment used by exercise participants.

data on the maximal fragment size and fluorescent dye label the blue and yellow systems from Profiler (Fig. 4), and color for each system. Fragment sizes for each locus are PowerPlex 16 (Fig. 5). In each histogram, the STR systems consistent among the kits manufactured by ABI, although are arranged with increasing fragment size from left to right some have been labeled with different colors as shown. In (see also Table 1). In each figure, the results for sample A (a, the Promega multiplex kit, some STR systems are located in b) and sample B (c, d) are shown for standard (a, c) and a different size range in comparison to the ABI kits. enhanced (b, d) PCR conditions. The multiplex typing results based on the recorded peak The amount of genomic DNA and the number of cycles heights (in rfu) are depicted for the three most commonly varied significantly between laboratories. Therefore, the used multiplexes SGM Plus (Fig. 3), Profiler Plus including decision to assign the results into the two categories

Table 1 STR loci, fragment sizes and dye label colors of three multiplex kits used by the exercise participants

STR system Genotypes SGM Plus Profiler/Plus Power Plex

Sample A Sample B Allele size Ba Ga Ya Allele size B G Y Allele size BGY range (bp) range (bp) range (bp)

D3S1358 15, 16 16, 17 <150 þ <150 þ <150 þ VWA 17 16, 18 <210 þ <210 þ <170 þ FGA 22, 25 21, 22 <360 þ <360 þ <450 þ Amelogenin XY XX <112 þ <112 þ <112 þ D8S1179 15, 16 14 <170 þ <170 þ <250 þ D21S11 29, 31 30 <250 þ <250 þ <260 þ D18S51 13, 14 13 <350 þ <350 þ <370 þ D5S818 11, 12 12, 14 <180 þ <160 þ D13S317 9, 13 11, 12 <240 þ <200 þ D7S820 10 8, 10 <300 þ <250 þ D19S433 15.2 14, 15 <140 þ THO1 9 6 <200 þ <200 þ TPOX 8, 9 8, 12 <290 þ CSF1PO 10, 11 10, 11 <360 þ D16S539 12, 13 9, 11 <270 þ <305 þ D2S1338 19, 20 16, 17 <350 þ Penta E 15, 20 5 <480 þ Penta D 9, 13 11, 13 <440 þ a B: blue; G: green; Y: yellow. P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134 127

‘‘standard’’ and ‘‘enhanced’’ PCR conditions could not be cycles, but rather increased the amount of DNA for enhance- based on clear-cut rules. In general, results were assigned to ment, or used 30 cycles already as their standard procedure. the ‘‘standard’’ category when 28 cycles were performed, Furthermore, other laboratories provided only one result for and to the ‘‘enhanced’’ category for 30 cycles or more. each sample, which had then to be assigned to one of the two However, some laboratories either did not use more than 28 categories for analysis. Thus, the ranges for the number of

Fig. 3. Results of STR typing for the SGM Plus multiplex kit. For each locus, the data are summarized for the three peak height categories: >150 rfu (black columns), 150–30 rfu (grey columns) and <30 rfu (white columns). The results for all STR loci are arranged in the order of observed fragment sizes from left (small fragments) to right (large fragments) for the three categories. (a) Results for degraded DNA sample A using standard PCR conditions, i.e. 0.25–2 ng of DNA and 28–30 PCR cycles according to manufacturer or established operational procedures (n ¼ 15 laboratories); (b) results for degraded DNA sample A using enhanced PCR conditions, i.e. up to 5 ng of DNA and up to 35 PCR cycles according to established operational procedures (n ¼ 19 laboratories); (c) results for degraded DNA sample B using standard PCR conditions, i.e. 0.5–2 ng of DNA and 28 PCR cycles according to manufacturer or established operational procedures (n ¼ 12 laboratories); (d) results for degraded DNA sample B using enhanced PCR conditions, i.e. up to 10 ng of DNA and up to 35 PCR cycles according to established operational procedures (n ¼ 17 laboratories). 128 P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134

Fig. 3. (Continued ). cycles and the amount of template DNA given in the figure more degraded sample B, the 40% failure rate was reached legends may overlap, and the number of laboratories con- as well using VWA and FGA for standard and enhanced tributing to the ‘‘standard’’ and ‘‘enhanced’’ category are not conditions, respectively, although the pattern was not so identical. straightforward (Fig. 3c and d). Some loci such as D8S1179, Using standard conditions for the SGM Plus kit, a sig- THO1 and D21S11 actually performed better with sample B nificant loss of information with 40% of negative results than with sample A. Also, the proportion of laboratories (<30 rfu) occurred for sample A around 180 bp (the size of reporting successful amplification for all loci from Amelo- the VWA alleles; Fig. 3a), whereas more than 60% success- genin to D2S1338 using enhanced conditions was almost ful amplification (>30 rfu) was achieved for the majority of 50% for sample B (Fig. 3d), but only 25% for sample A laboratories up to approx. 320 bp (FGA/FIBRA system; (Fig. 3b). However, it has to be kept in mind that successful Fig. 3b) using enhanced conditions. For the supposedly amplification does not necessarily mean that the typing P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134 129 results were correct for all loci. This aspect will be discussed Profiler kit as these have an identical locus composition in in the context of the results from all three multiplexes. both kits. When compared to the SGM Plus results, the The results for the Profiler Plus kit were supplemented Profiler Plus data exhibited a higher proportion of low with data from the ‘‘blue’’ and ‘‘yellow’’ systems from the molecular weight alleles with peak heights >150 rfu both

Fig. 4. Results of STR typing for the Proﬁler/Proﬁler-Plus multiplex kit. For each locus, the data are summarized for the three peak height categories: >150 rfu (black columns), 150–30 rfu (grey columns) and <30 rfu (white columns). The results for all STR loci are arranged in the order of observed fragment sizes from left (small fragments) to right (large fragments) for the three categories. (a) Results for degraded DNA sample A using standard PCR conditions, i.e. 0.25–1.25 ng of DNA and 28–32 PCR cycles according to manufacturer or established operational procedures (n ¼ 8 laboratories); (b) results for degraded DNA sample A using enhanced PCR conditions, i.e. up to 2.5 ng of DNA and up to 34 PCR cycles according to established operational procedures (n ¼ 5 laboratories); (c) results for degraded DNA sample B using standard PCR conditions, i.e. 0.25–1.25 ng of DNA and 28–32 PCR cycles according to manufacturer or established operational procedures (n ¼ 6 laboratories); (d) results for degraded DNA sample B using enhanced PCR conditions, i.e. up to 2.5 ng of DNA and up to 34 PCR cycles according to established operational procedures (n ¼ 5 laboratories). 130 P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134

Fig. 4. (Continued ). for samples A and B. A loss of information with >40% ful amplification) compared to the SGM Plus kit with a failure rate begins with D21S11 around 220 bp for standard yellow dye (complete failure; Fig. 4a and c). Similar obser- conditions (Fig. 4a and c), and, using enhanced conditions, vations were made in a recent study on the interpretation of with D7S820 for sample A (approx. 280 bp, Fig. 4b) and mixtures from forensic casework. It was found that the D13S317 for sample B which has a similar size as D21S11 detection of mixed profiles was more successful using (Fig. 4d). It might be speculated that some of these differ- blue-labeled STR systems [9]. On the other hand, successful ences can be related to the fact that some loci have changed amplification across all loci (>30 rfu) using enhanced con- colors when the SGM Plus kit was designed to accommodate ditions was not better than for the SGM Plus kit, i.e. 25% for the additional STR loci (Table 1). In particular, using both samples due to the low values for D18S51, and although standard conditions, FGA performed significantly better the equally large D2S1338 system is present in the SGM in the Profiler Plus kit carrying a blue dye (30–50% success- Plus kit. P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134 131

The PowerPlex 16 multiplex comprises the largest num- formed only poorly, and the overall pattern of successfully ber of loci and covers the widest fragment size range up to ampliﬁed loci showed more variation. Using standard con- approx. 480 bp for the Penta E system. Therefore, it is no ditions, successful ampliﬁcation of >60% was achieved for surprise that the systems with fragment sizes >300 bp per- D16S539 for sample A and D8S1179 for sample B, which

Fig. 5. Results of STR typing for the Powerplex 16 multiplex kit. For each locus, the data are summarized for the three peak height categories: >150 rfu (black columns), 150–30 rfu (grey columns) and <30 rfu (white columns). The results for all STR loci are arranged in the order of observed fragment sizes from left (small fragments) to right (large fragments) for the three categories. (a) Results for degraded DNA sample A using standard PCR conditions, i.e. 0.5–1.25 ng of DNA and 30–32 PCR cycles according to manufacturer or established operational procedures (n ¼ 8 laboratories); (b) results for degraded DNA sample A using enhanced PCR conditions, i.e. up to 2.5 ng of DNA and up to 36 PCR cycles according to established operational procedures (n ¼ 4 laboratories); (c) results for degraded DNA sample B using standard PCR conditions, i.e. 0.5–1.25 ng of DNA and 30–32 PCR cycles according to manufacturer or established operational procedures (n ¼ 8 laboratories); (d) results for degraded DNA sample B using enhanced PCR conditions, i.e. up to 2.5 ng of DNA and up to 36 PCR cycles according to established operational procedures (n ¼ 4 laboratories). 132 P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134

Fig. 5. (Continued ). have sizes up to 270 bp (Fig. 5a and c). The size range of observed. However, due to the heterogeneity of the submitted successful amplification was extended using enhanced con- data regarding the amount of DNA used, the PCR parameters ditions to CSF1PO for sample A and to D18S51 for sample B and other conditions it was not possible to compile a locus-by- (both approx. 340 bp). Using enhanced conditions, the pro- locus statistical analysis. Therefore, typical problems are portion of systems exhibiting peak heights of >150 rfu was summarized. A strong peak imbalance was found in hetero- clearly improved. In general, the performance of the Power- zygous genotypes in particular for the larger STR systems, and Plex 16 kit was comparable to those of the ABI kits based on for enhanced conditions applying more than 30 cycles. Arte- fragment length assessment, although the locus composition fact signals (‘‘pull-ups’’) occurred due to overamplification is quite different regarding STR size ranges and dye colors mimicking alleles not present in the sample. Allelic drop-out (Table 1). (i.e. the complete loss of one allele in a heterozygous geno- As already mentioned, not all alleles could be identified type) was observed more frequently, and sometimes the correctly for all loci. A number of common problems were smaller of two alleles at a given locus was affected. P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134 133

Table 2 times prevented the detection of a regular allele due to Allele calling thresholds for STR systems based on peak heights in the threshold settings of peak height detection, although relative fluorescence units (rfu) this threshold also prevented the inclusion of artefact bands Threshold (rfu) No. of labs in some cases. As we have included a peak height category of 30–150 rfu for our analysis, we have of course picked up n % more alleles than the majority of participating laboratories 30 4 10.5 would do using their standard threshold. This was done 50 23 60.5 deliberately to learn more about the detection limits and 60–100 3 7.9 the occurrence of artefacts using the various multiplexes. >100 3 7.9 In a real casework scenario, some of the wrongly recorded Flexible 2 5.3 low level alleles as well as low level correct alleles would n.d. 3 7.9 have been left out in favor of a more reliable analysis. Only Total 38 100 two laboratories have reported results using a flexible n.d.: not defined. threshold for allele calling based on the overall quality of the results (Table 2). A number of laboratories have used ‘‘home-made’’ sin- Some STR loci repeatedly exhibited problems. FGA, gleplex STR systems such as THO1, VWA, FGA, D2S1338, D21S11 and D18S51 were most frequently affected by SE33, D8S1179, D18S51, and others. As these results are allelic dropout in the ABI multiplexes. This was more quite diverse regarding the loci studied, they could not be prominent using enhanced conditions for loci that did not subjected to a comparative analysis. However, all results amplify at all under standard conditions. A similar pattern submitted were correct. Most laboratories typically use these was found for the D21S11, D8S1179 and D16S539 systems singleplex systems for difficult samples such as degraded for the PowerPlex 16. The fragment sizes of these loci are DNA. In particular, several STR systems with redesigned around or slightly above the fragment size threshold of primer sequences to generate short amplicons gave complete successful amplification, and stochastic effects most prob- and correct results for both samples [10,11]. ably generate the ambiguous and unreliable results. A Eight laboratories reported their results for mitochon- special situation was found in the D19S433 system of the drial DNA sequence analysis (Table 3). Of these, four SGM Plus kit. Although this locus generates very small laboratories failed to obtain results due to poor amplifica- fragments <140 bp, two laboratories reported the presence tion. The other four laboratories reported either complete or of a 14.2 allele, and another two laboratories a 15 allele for partial results, as the HV-II region was not analyzed by all sample A which was homozygous for the 15.2 allele. The participants. The two laboratories with complete results peak heights of these extra alleles had almost the same applied an overlapping PCR amplification strategy with height as the correct allele. Whereas the presence of the 14.2 short fragments of approx. 200 bp. Our control experiments allele could be explained by an extreme stutter effect, this after preparation of the degraded DNA indicated that such cannot be the case for the false 15 allele, as it is separated by fragments can be amplified reliably [8]. This has also been only 2 bp from the true allele. shown in recent studies on the genetic identity of historical Most laboratories apply a lower threshold for allele samples [12]. detection to prevent the inclusion of false alleles or stutter From these results, some conclusions can be drawn: bands. A total of 60% of the laboratories use a peak height threshold of 50 rfu (see Table 2). Thus, the use of the To overcome the effects of template DNA degradation, a GenoTyper software for automatic data analysis some- PCR protocol optimized for all parameters such as the

Table 3 Results of HV-I and HV-II mitochondrial DNA sequence analysis

Sample HV-I Resultsa

16126 16182 16183 16189 16217 16294 16295 16304 Pos. Neg. n.t.

A – CCCC– T – 44 BC––––T – C3 4 1 HV-II 73 146 152 263 309.1 315.1 AGC– G – C251 BG– CGCC 2 4 2 a Pos.: correct result; Neg.: no result; n.t.: not tested. 134 P.M. Schneider et al. / Forensic Science International 139 (2004) 123–134

amount of template DNA and Taq polymerase concentra- [3] M. Graw, H.J. Weisser, S. Lutz, DNA typing of human tion relative to the number of PCR cycles has to be remains found in damp environments, Forensic Sci. Int. 113 established. (2000) 91–95. Flexible guidelines for the interpretation of the observed [4] J.P. Whitaker, T.M. Clayton, A.J. Urquhart, E.S. Millican, T.J. Downes, C.P. Kimpton, P. Gill, Short tandem repeat typing of DNA profiles have to be applied. The guidelines should bodies from a mass disaster: high success rate and include strategies to reanalyze the sample if evidence for characteristic amplification patterns in highly degraded low-level peak signals is obtained. To allow for this, a samples, Biotechniques 18 (1995) 670–677. careful human evaluation of all results in addition to a [5] R. Sparkes, C. Kimpton, S. Watson, N. Oldroyd, T. Clayton, computer-based analysis (e.g. using the GenoTyper soft- L. Barnett, J. Arnold, C. Thompson, R. Hale, J. Chapman, A. ware) is recommended. Urquhart, P. Gill, The validation of a seven-locus multiplex For mtDNA analysis of heavily degraded DNA, PCR STR test for use in forensic casework. Part I. Mixtures, primers covering short overlapping DNA regions (as ageing. degradation and species studies, Int. J. Legal Med. described e.g. in [13]) should be applied. 109 (1996) 186–194. [6] E.A. Cotton, R.F. Allsop, J.L. Guest, R.R. Frazier, P. Koumi, I.P. Callow, A. Seager, R.L. Sparkes, Validation of the Some of the characteristics of STR typing results from AMPFlSTR SGM plus system for use in forensic casework, degraded DNA samples from this study have also been Forensic Sci. Int. 112 (2000) 151–161. observed under ‘‘low copy number’’ (LCN) conditions, [7] B.E. Krenke, A. Tereba, S.J. Anderson, E. Buel, S. Culhane, where stochastic effects strongly influence the outcome of C.J. Finis, C.S. Tomsey, J.M. Zachetti, A. Masibay, D.R. PCR amplification. Thus, some of the guidelines developed Rabbach, E.A. Amiott, C.J. Sprecher, Validation of a 16-locus for interpretation of LCN STR profiles [14] may also be fluorescent multiplex system, J. Forensic Sci. 47 (2002) 773– applicable when highly degraded DNA has to be analyzed. 785. The use of standardized degraded control DNA is helpful [8] K. Bender, M.J. Farfan, P.M. Schneider, Preparation of to understand and optimize the parameters affecting the degraded human DNA under controlled conditions, Forensic Sci. Int., in press. success of STR typing. The availability of such a degraded [9] Y. Torres, I. Flores, V. Prieto, M. Lopez-Soto, M.J. Farfan, A. control DNA in sufficient amounts would enable the forensic Carracedo, P. Sanz, DNA mixtures in forensic casework: a 4- laboratories to simulate a situation of heavily decomposed year retrospective study, Forensic Sci. Int. 134 (2003) 180–186. DNA, and thus allow studies to enhance the efficiency of [10] A. Hellmann, U. Rohleder, H. Schmitter, M. Wittig, STR their typing methods. typing of human telogen hairs—a new approach, Int. J. Legal. Med. 114 (2001) 269–273. [11] P. Grubwieser, R. Muhlmann, W. Parson, New sensitive Acknowledgements amplification primers for the STR locus D2S1338 for degraded casework DNA, Int. J. Legal Med. 117 (2003) 185–188. The present study was supported by contract SMT 97- [12] E. Jehaes, K. Toprak, N. Vanderheyden, H. Pfeiffer, J.J. Cassiman, B. Brinkmann, R. Decorte, Pitfalls in the analysis of 7506 from the European Commission in the context of the mitochondrial DNA from ancient specimens and the con- ‘‘Standardization of DNA Profiling Methods in the European sequences for forensic DNA analysis: the historical case of the Union’’ (STADNAP) network initiative. putative heart of Louis XVII, Int. J. Legal. Med. 115 (2001) 135–141. [13] M.N. Gabriel, E.F. Huffine, J.H. Ryan, M.M. Holland, T.J. References Parsons, Improved mtDNA sequence analysis of forensic remains using a ‘‘mini-primer set’’ amplification strategy, J. [1] P.D. Martin, H. Schmitter, P.M. Schneider, A brief history of Forensic Sci. 46 (2001) 247–253. the formation of DNA databases in forensic science within [14] J.P. Whitaker, E.A. Cotton, P. Gill, A comparison of the Europe, Forensic Sci. Int. 119 (2001) 225–231. characteristics of profiles produced with the AMPFlSTR [2] P. Gill, Role of short tandem repeat DNA in forensic SGM Plus multiplex system for both standard and low copy casework in the UK—past, present, and future perspectives, number (LCN) STR DNA analysis, Forensic Sci. Int. 123 Biotechniques 32 (2002) 366–368. (2001) 215–223.