Extended PCR conditions to reduce drop-out frequencies in low template STR typing including unequal mixtures

Article Outline

• Abstract
• 1. Introduction
• 2. Methods and materials
  • 2.1. DNA samples
  • 2.2. Amplification
  • 2.3. Detection of STR alleles
• 3. Results and discussion
  • 3.1. Drop-out frequency and various thermal cycling conditions
  • 3.2. Reliability of the extended PCR method
  • 3.3. The effect of extended PCR on mock casework samples and consensus profiles
  • 3.4. Enhanced detection of minor component in 10 to 1 mixtures for various STR kits
  • 3.5. Y-specific analysis with extended PCR
  • 3.6. Increased tolerance to EDTA in extended PCR
• 4. Concluding remarks
• Acknowledgments
• Appendix A. Supplementary data
• References
• Copyright

Abstract 

Forensic laboratories employ various approaches to obtain short tandem repeat (STR) profiles from minimal traces (<100pg DNA input). Most approaches aim to sensitize DNA profiling by increasing the amplification level by a higher cycle number or enlarging the amount of PCR products analyzed during capillary electrophoresis. These methods have limitations when unequal mixtures are genotyped, since the major component will be over-amplified or over-loaded. This study explores an alternative strategy for improved detection of the minor components in low template (LT) DNA typing that may be better suited for the detection of the minor component in mixtures. The strategy increases the PCR amplification efficiency by extending the primer annealing time several folds. When the AmpFℓSTR^® Identifiler^® amplification parameters are changed to an annealing time of 20min during all 28 cycles, the drop-out frequency is reduced for both pristine DNA and single or multiple donor mock case work samples. In addition, increased peak heights and slightly more drop-ins are observed while the heterozygous peak balance remains similar as with the conventional Identifiler protocol. By this extended protocol, full DNA profiles were obtained from only 12 sperm heads (which corresponds to 36pg of DNA) that were collected by laser micro dissection. Notwithstanding the improved detection, allele drop-outs do persist, albeit in lower frequencies. Thus a LT interpretation strategy such as deducing consensus profiles from multiple independent amplifications is appropriate. The use of extended PCR conditions represents a general approach to improve detection of unequal mixtures as shown using four commercially available kits (AmpFℓSTR^® Identifiler, SEfiler Plus, NGM and Yfiler). The extended PCR protocol seems to amplify more of the molecules in LT samples during PCR, which results in a lower drop-out frequency.

Keywords: Low template DNA, Allele drop-out, Short tandem repeat (STR), Genotyping, Annealing time, Consensus

Back to Article Outline

1. Introduction 

Obtaining a reliable short tandem repeat (STR) profile from low template (LT) DNA samples is a great challenge in forensic DNA typing. Generally, amplifications with less than 100pg of DNA are considered LT samples [1], [2], [3]. Internal validation studies for various commercially available STR kits have shown that with 63pg amounts of template DNA allele drop-out starts to occur in our laboratory setup. Hence, we need 30 spermatozoa or 20 somatic cells (which corresponds to 90pg and 120pg of DNA respectively) to obtain a full profile after laser micro dissection (LMD) [4], [5]. To increase the sensitivity of STR profiling, a number of LT techniques can be performed. These include increasing the PCR cycle number, nested PCR, molecular displacement amplification possibly in combination with molecular crowding, whole genome amplification, post-PCR sample clean-up prior to genotyping analysis and/or increased injection time or voltage during capillary electrophoresis (CE) [6], [7], [8], [9]. Profile interpretation can be challenging for LT samples as these may contain stochastic amplification effects such as heterozygote imbalance, allele drop-out, allele drop-in, increased stutter peaks, near threshold peaks and masked minor alleles in mixtures [2], [3], [8], [10], [11], [12]. A common strategy for the interpretation of reliable LT profiles involves the generation of a consensus profile from multiple independent amplifications of a sample, whereby only alleles that are replicated are included in the consensus profile [2], [8], [13].

The most prevalent stochastic effect is allele drop-out which can be expressed as allelic or locus drop-out. We hypothesized that these drop-outs are not due to complete absence of template DNA, but are rather missed during the (first rounds of) amplification as a consequence of ineffective primer binding. In order to develop a method with increased profiling success for LT DNA samples, the primer annealing time was increased from 1min up to 20min during a subset or all of the amplification cycles for AmpFℓSTR^® Identifiler^® (Identifiler) profiling. The frequency of drop-outs was determined for pristine and mock case work samples and in consensus profiles deduced from replicated amplifications of the samples. The occurrence of allele drop-ins and effects on peak height and heterozygote balance were determined. The general applicability of the approach was tested using the AmpFℓSTR^® SEfiler plus™ (SEfiler+), AmpFℓSTR^® NGM™ (NGM) and AmpFℓSTR^® Yfiler^® (Yfiler) PCR amplification kits. This method was able to improve the detection of the minor LT component in an unequal mixture. Applying conventional LT techniques would produce over-amplified or over-loaded profiles for this type of sample. Therefore, this approach has additional value in the analysis of forensic LT samples.

Back to Article Outline

2. Methods and materials 

2.1. DNA samples 

Several sample sources were used in this study. These include three pristine DNAs, being DNA007: the reference DNA in the Yfiler kit, DNA9947A: the reference DNA in the Identifiler kit and human DNA (hDNA): the reference DNA in the Quantifiler™ kit, all supplied by Applied Biosystems ((AB), Nieuwerkerk aan de IJssel, The Netherlands). DNA samples were diluted to the required concentrations with dH₂O, with the exception to the samples which were specifically diluted in EDTA-containing buffers (100% AE buffer (QIAgen, Venlo, the Netherlands)). In addition, DNA was extracted from vaginal epithelial cells of two female donors and several dilutions were made to obtain LT samples with a maximum concentration of 0.12ng/μl. Three other mock casework samples consisted of skin epithelial cells and were obtained by mimicked strangulation (by rubbing the hands of one donor on the arm of a second donor). These samples had a mixture ratio of approximately 1:2, and the concentrations of total human DNA were below 0.024ng/μl. For the Yfiler experiments, DNA mixtures consisted of male pristine DNA at a constant concentration of 24pg/10μl and female DNA extracted from vaginal samples at variable concentrations ranging from 2.4ng/10μl, 6ng/10μl, 12ng/10μl to 24ng/10μl so that the male:female ratio varied from 1:100, 1:250, 1:500 to 1:1000.

In the LMD experiment, detected spermatozoa were collected using a pulsed nitrogen UV-A laser on the PALM^® MicroBeam C HT robomover (PALM Microlaser Technologies GmbH, Bernried, Germany) from the surface of a Poly-Ethylen-Teraphtalate (PET) slide (PALM). The selected cells were catapulted into 1μl cell extraction mix (2% lysis enzyme, 10% 10× lysis buffer, 10% 0.9mM DTT, 7% AdvaGold PCR dye (in dH₂O)) on an AmpliGrid AG480F DNA free micro liter reaction slide (all Advalytix, Munich, Germany). Immediately after catapultation the cell extraction mix was covered with 5μl sealing solution A (Advalytix) and lysed for 10min at 75°C followed by 2min at 95°C.

All donors were consented volunteers with known STR profiles, and all profiles generated were compared and verified.

2.2. Amplification 

All reactions were set up with the Identifiler, SEfiler+, NGM or Yfiler PCR amplification kit (AB). Amplification was performed in a 9700 thermal cycler (AB) using various cycling parameters (Table 1). Cycling conditions for the conventional method were in accordance with the manufacturer's specifications with the exception that the number of cycles for the SEfiler+ kit was reduced to 28. For every extended protocol an additional amount of 0.5μl AmpliTaq Gold^® DNA Polymerase (AmpliTaq) (AB) was added to the reaction mixture. All amplifications were carried out in a 25μl PCR volume, except for the LMD samples for which the full DNA extract (1μl) was amplified in a reaction volume of 15μl.

Table 1. Tested cycling parameters for the various STR kits.

STR kit	Denaturation time (95°C)a		Annealing time (59°Cc)a		Elongation time (72°C)a		Final extension time (60°C)a		# cycles
	Conventional	Extendedb	Conventional	Extendedb	Conventional	Extendedb	Conventional	Extendedb
Identifiler	1′	1′ or 20″	1′	5′ to 20′	1′	1′ or 30″	60′	10′	28
SEfiler+	20″	20″	2′	20′	1′	30″	60′	10′	28
NGM	20″	20″	3′	20′	–	–	10′	10′	29
Yfiler	1′	20″	1′	3′–20′	1′	30′́	80′	10′	30

2.3. Detection of STR alleles 

PCR products were separated and detected using standard procedures on a 3130 XL Genetic Analyzer (AB) with POP-4™ polymer (AB) following a 15s injection time at 3kV, for Yfiler analysis injection time was set at 18s at 1.2kV. Samples were prepared for CE by adding 1μl PCR product to 9μl loading mixture consisting of the suitable amount of Hi-Di™ Formamide (AB) and the appropriate amount and type of size standard which is 0.3μl GeneScan-500 LIZ (AB) for Identifiler and NGM, 0.3μl GeneScan-600 LIZ (AB) for SEfiler+ and 0.25μl GeneScan-500 LIZ for Yfiler analysis. Data were analyzed with either Genemapper ID^® v3.2.1 software or Genemapper ID-X v1.1.1 (AB) using a detection threshold of 50 relative fluorescence units (rfu). For each STR kit, locus-specific −1 stutter filters were applied following the instructions of the manufacturer (AB). For NGM, in addition +1 stutter filters were used that resulted from an internal validation study; for locus D22S1045 a filter of 7.36% was applied and for all other NGM loci a filter of 2.5% was set. When determining the percentage of detected alleles present in the STR profiles, homozygous alleles were counted as two. Peak height ratios (PHR) at heterozygous loci were calculated by dividing the height of the lower peak by that of the higher peak (PHR=PH_low/PH_high). For generating consensus profiles, all alleles detected in half (n/2) of the independent amplifications were included as described in Benschop et al. [13]. The software tool LCstat [14] was used for mock casework samples to deduce and analyse consensus profiles from the Genemapper data.

Back to Article Outline

3. Results and discussion 

3.1. Drop-out frequency and various thermal cycling conditions 

When low amounts of DNA are amplified, some alleles drop-out from amplification while others are amplified significantly above detection level. Although it is conceivable that the non-amplified alleles are fully absent in the DNA extract added to the PCR, it is also possible that these alleles were present but not picked up for amplification. To increase the chance that primers bind to the template molecules present in a PCR mixture, we increased the annealing time in the thermal cycling program. Several conditions were tested by varying two aspects in the Identifiler protocol for this extended PCR approach: (1) the number of cycles undergoing extended annealing time which differed from the first 5, 10 and 15 cycles to all 28 cycles and (2) the duration of the extended annealing time which varied from 5, 15 to 20min. To ensure that sufficient polymerase is present up to the last cycle, all extended PCRs were carried out with 0.5μl additional AmpliTaq. We determined the average percentage of allele drop-out for the amplification of 30pg of pristine DNA007 over two independent experimental series (in the first experiment, 28 replicates were performed and in the second experiment seven replicates). Fig. 1 shows the average percentage of allele drop-out for the various extended PCR procedures. We found the lowest percentage of drop-outs when the annealing time is extended to 20min for all 28 cycles as the average drop-out frequency decreased from 27% (during conventional PCR) to 13% (Fig. 1). We found that the denaturation and elongation times could be reduced from 1min to 20s and 30s, respectively (Fig. 1, last two bars) and these conditions were used for subsequent experiments. The positive results of the extended method upon amplification of 30pg DNA (14% more alleles detected) were confirmed when 18pg of DNA007 were amplified in 7-fold; in comparison to the conventional method 7% more alleles were detected (results not shown). By substantially increasing the PCR annealing time, more efficient genotyping of LT samples appears possible as less drop-outs occur.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Comparison of the average drop-out frequency by means of Identifiler genotyping, tested with 30pg pristine DNA007 using extended annealing times during a variable number of cycles and for variable durations (measured average over two series, one consisting of 28 replicates and the other of seven replicates). Time is indicated in minutes (′). (a) Per reaction an additional amount of 0.5μl AmpliTaq was added. (b) Denaturation time reduced to 20s; elongation time reduced to 30s.

3.2. Reliability of the extended PCR method 

The performance of an amplification protocol can be assessed from various characteristics: frequency of allele drop-outs, number of allele drop-ins (partly from increased stutter ratios), peak height and heterozygote peak balance. As these parameters depend on the amount of DNA input, a series of DNA007 inputs ranging from 15pg to 480pg was amplified with the Identifiler kit in 7-fold using the conventional and extended protocol. These inputs represent both low level and standard DNA inputs as forensic samples may be mixtures in which only one of the components is LT. For LT inputs (15, 30 and 60pg), the percentage of drop-outs halved with the extended protocol (Table 2). Allele drop-ins occurred more frequently in the profiles generated by extended PCR, especially with higher inputs (120, 240 and 480pg) (Table 2). Most drop-in alleles represent +1 stutters (amplification products one repeat unit longer resulting from polymerase slippage [10], [15]), which is in agreement with previous findings regarding LT genotyping [13]. The average peak height for all detected alleles increased 1.4-fold on average with the extended PCR conditions. Furthermore, the peak height ratio (PHR) at heterozygous loci was calculated and heterozygote imbalance (PHR<0.8) was found for the LT samples for both PCR approaches (Table 2). In these LT samples, PHRs were on average above 0.60 which is an acceptable value for single source samples [2]. Overall, with extended PCR conditions the number of alleles and peak heights increased and slightly more drop-ins were observed, depending on the input.

Table 2. Occurrence of stochastic amplification artefacts for various amounts of pristine DNA comparing the conventional and the extended PCR method for Identifiler profiling (n=7).

Quantity pristine DNA (pg)	Average percentage allele drop-outs		Average no. drop-ins per profile		Average peak height (rfu)		Heterozygote balance
	Conventional	Extended	Conventional	Extended	Conventional	Extended	Conventional	Extended
15	13.6%±1.9%	9.0%±3.3%	0	0	83±8	111±13	0.64±0.16	0.64±0.16
30	5.7%±1.3%	2.7%±1.8%	0	0	122±11	158±17	0.69±0.06	0.65±0.08
60	2.2%±0.8%	0.9% ±0.5%	0.1	0	224±11	306±36	0.68±0.04	0.63±0.04
120	0	0	0	0.3	447±40	664±67	0.73±0.03	0.76±0.04
240	0	0	0	0.3	867±83	1634±121	0.78±0.02	0.78±0.02
480	0	0	0.1	0.4	1723±112	2206±151	0.86±0.02	0.84±0.02

3.3. The effect of extended PCR on mock casework samples and consensus profiles 

When analyzing LT samples during casework, many laboratories use a consensus strategy and perform multiple independent amplifications in order to include replicated alleles in a consensus profile [2], [8], [13]. Although the extended PCR protocol reduces the number of drop-outs, the samples remain LT and a consensus approach remains appropriate. The effects of the extended Identifiler PCR protocol in consensus profiles were assessed focusing on the percentage of detected alleles and the number of drop-ins. If drop-ins occur randomly it is expected that they are not included in the consensus profile. Three types of LT mock casework samples were used; (1) diluted single contributor samples (four samples each replicated in 6-fold), (2) a specific number of cells collected by LMD (12 spermatozoa or 36pg of DNA which were collected in 8-fold), and (3) two-donor samples obtained by mimicked strangulation (three samples each replicated in 6-fold) in which both components are low level (approximately in a 2:1 ratio). For each sample, consensus profiles were deduced by including the alleles detected in half of the amplifications (Table 3). The positive effect on the percentage of detected alleles using the extended protocol that is observed in the individual profiles is carried through to the consensus profiles (Table 3). In the individual profiles a limited number of drop-ins (mostly at +1 stutter position) was detected when using the extended PCR procedure; these do not get included in the consensus profiles (Table 3), as they apparently occur at variable positions. Notably, in three of the eight LMD samples amplified with the extended protocol, a full Identifiler profile was obtained from the 12 spermatozoa while no full profiles were obtained using the conventional protocol. When using the extended protocol on the LMD samples, 30% more alleles were present and the peak heights were 1.7 times higher (results not shown). For the mixed samples the extended protocol increases the percentage of detected alleles substantially; for the major contributor 48% and for the minor component 25% more alleles are detected on average. This increase is seen in both the individual and the consensus profiles (Table 3). Three drop-ins were present in the extended PCR data, but none of these were included in the consensus profiles (Table 3). These results show that the extended PCR protocol is compatible with the frequently used consensus approach for both single and two-donor LT samples.

Table 3. The average percentage of detected alleles and number of drop-ins in four single contributors (n=6), 12 sperm selected by LMD (n=8) and three mixed two donor mock casework samples (n=6) using the Identifiler system with the conventional and extended PCR method, expressed in individual and consensus profiles.

3.4. Enhanced detection of minor component in 10 to 1 mixtures for various STR kits 

One of the most difficult types of casework samples are mixtures in which the minor component is several folds less abundant than the major component. Samples with such a mixture ratio are not suited for sensitized DNA profiling by increased cycling or higher CE injection settings as the major component will become over-amplified or over-loaded [8], [9].

Admixed samples in which two pristine DNAs were present in a 10 to 1 ratio (300pg of hDNA and 30pg of DNA007) were analyzed using the conventional and extended PCR protocol. In addition to the Identifiler STR kit, SEfiler+ and NGM kits were used. The mixtures were analyzed in 27-fold for each kit. Full profiles were obtained for the major contributor (300pg hDNA) for both PCR protocols. For the minor components, a reduced drop-out frequency was obtained for all three autosomal STR kits (Table 4). The largest decrease in percentage of allele drop-out was observed for the Identifiler and SEfiler+ kits; the percentage of allele drop-outs decreased less for the NGM kit (Table 4). This limited positive effect of the extended protocol with NGM may arise from the fact that standard NGM amplification uses already an annealing time of 3min, but it is also conceivable that the additional 29th cycle during the NGM PCR has a role. Since the presence of the major component may stimulate the occurrence of allele drop-ins (Table 2), the number of allele drop-ins per profile was determined. For the extended PCR method a small increase was found for all three STR kits: the majority of allele drop-ins represented +1 stutters (52%) and the remaining occurred at −1 stutter (15%) or other (33%) positions (Table 4). We examined whether stutter peaks were more pronounced for the extended protocol by comparing the average stutter ratios at selected alleles for the conventional and extended method. No significant differences were found for the −1 and +1 stutter ratios between the conventional and the extended method (Supplementary Table 1), and the drop-ins at stutter positions seem to reflect stochastic events that are known to accompany low template DNA profiling. On the whole, the results imply that the extended PCR protocol can improve detection of the minor component in an unequal mixture and that this strategy can be applied to various STR kits.

Table 4. Genotyping results of the LT minor component in a 10:1 mixture of two pristine DNAs (hDNA:DNA007) amplified with different STR kits each in 27-fold. The percentage of detected alleles of the minor component was determined by considering non-shared alleles only (20 alleles for Identifiler, 15 alleles for SEfiler+, 20 alleles for NGM).

					Frequency of drop-in type (in %)
	Average % allele drop-out		# allele drop-ins per profile		Conventional			Extended
STR kit	Conventional	Extended	Conventional	Extended	Stutter +1	Stutter −1	Other	Stutter +1	Stutter −1	Other
Identifiler	29%±2%	12%±1%	0.4	0.7	27%	55%	18%	53%	26%	21%
SEfiler+	21%±1%	14%±1%	0.3	0.7	13%	13%	75%	60%	5%	35%
NGM	36%±2%	32%±2%	0.1	0.3	100%	0%	0%	43%	14%	43%

3.5. Y-specific analysis with extended PCR 

A specific type of LT profiling is the detection of a low level male component in samples with an excess amount of female DNA. We tested whether the extended PCR assists in male-specific detection in such mixtures using the Yfiler STR kit. First, various extended annealing times (20, 10 and 3min) were tested. An annealing time of 20min during all 30 cycles resulted in background peaks both in samples containing only male (24pg) or male and female (24pg and 1ng) DNA. Optimal results were obtained using a 3min annealing time during all 30 cycles; all peaks were fully adenylated and no background peaks were visible. Subsequently, mixtures of male and female DNA in 4 ratios namely 1:100, 1:250, 1:500 and 1:1000 with a constant input of 24pg male DNA were analyzed using the conventional or extended Yfiler PCR protocol (in 8-fold). For the extended method, allele drop-out frequency decreased significantly compared to the conventional method and the average peak height increased up to 2-fold (Table 5). The presence of an excess amount of female DNA (up to 24ng) had no negative effect on the detection of the male component and did not reduce peak heights, as was also described by Gross and co-workers [16]. For the 1:1000 mixture unspecific signals were observed, for both the conventional and the extended method, a phenomenon previously described by Prinz and co-workers [17]. These results show that the extended PCR approach is also applicable to non-autosomal STR typing.

Table 5. Amplification characteristics of Yfiler genotyping using the conventional or extended (3min) PCR protocol. 24pg male DNA and various amounts of female DNA were used (n=8).

Male:female ratio	Average % allele drop-out		# allele drop-ins per profile		Average peak height (rfu)
	Conventional	Extended	Conventional	Extended	Conventional	Extended
1:100	32±2	12±2	0	0.5	108±54	217±103
1:250	23±2	17±3	0.3	0	118±50	188±105
1:500	36±2	15±2	0	0	113±44	201±85
1:1000	19±2	10±1	0.8	0.4	122±56	223±115

3.6. Increased tolerance to EDTA in extended PCR 

To protect DNA from DNase degradation, extracts are often stored in EDTA-containing buffers. The presence of an excess amount of EDTA may, however, inhibit polymerase activity during PCR. Since more polymerase is added during the extended PCR protocol, increased tolerance to EDTA may occur. To address this issue, we amplified 5μl and10μl of a 10pg/μl DNA solution with an EDTA concentration of 0.5mM in 9-fold using three Identifiler PCR methods: conventional, conventional with additional AmpliTaq and extended PCR. EDTA-inhibition is clearly observed using conventional cycling conditions; for an input of 10μl (EDTA concentration in PCR 0.2mM) similar drop-out percentages and peak heights were observed as with an input of 5μl (EDTA concentration in PCR mixture 0.1mM), while twice the amount of DNA is present (Table 6). Adding additional AmpliTaq overcomes the negative effect of EDTA both with a conventional and an extended annealing time as peak heights increase and the percentage of drop-outs decreases when the PCR input is doubled (Table 6). These results show that the extended PCR protocol has good tolerance to increased EDTA concentrations (0.2mM). This is concordant with the results of the initial tests on the optimization of the extended PCR protocol described in Fig. 1.

Table 6. Resistance to EDTA inhibition of various Identifiler PCR conditions of a 10pg/μl DNA solution with an EDTA concentration of 0.5mM (n=9).

PCR input	Average percentage allele drop-out			Average no. allele drop-ins per profile			Average peak height detected alleles (rfu)a
	Conventional	Conventionalb	Extendedb	Conventional	Conventionalb	Extendedb	Conventional	Conventionalb	Extendedb
5μl	77%±4%	62%±8%	30%±5%	0	0	0	26±6	38±9	66±21
10μl	78%±4%	44%±14%	9%±3%	0	0	0	25±6	56±16	128±68

Back to Article Outline

4. Concluding remarks 

In this study we have shown that LT sample profiling can be efficiently improved by increasing the annealing time in the PCR. The total amplification time of the extended protocol is around ten hours, but it requires no hands-on time and can be incorporated in a flow through for LT samples. The approach is applicable to various STR kits; both autosomal and Y-specific. Improved genotyping is observed not only for single donor samples but also for mixtures in ratios up to 1:10. Other LT techniques, like increased cycling parameters or higher CE injection settings result in over-amplification or over-loaded profiles for the major component when unequal mixtures are used. DNA profiles obtained by the extended PCR protocol show less allele drop-outs and increased peak heights than profiles obtained by the conventional protocol. The heterozygote peak balance is not affected. Slightly more allele drop-ins are observed when genotyping with the extended PCR protocol, with the majority at +1 stutter position. Notwithstanding the increased sensitivity of the extended PCR method, a LT interpretation strategy is appropriate, like generating a consensus profile from multiple independent amplifications. This consensus strategy restricts the effects of the increased occurrence of drop-ins: the drop-ins are not included in the consensus profile, due to their occurrence at variable positions. In summary, by substantially extending the annealing time, the few DNA molecules present in a LT sample seem to be more effectively used as templates and therefore, less drop outs occur.

Back to Article Outline

Acknowledgments 

This study was supported by a grant from the Netherlands Genomics Initiative/Netherlands Organization for Scientific Research (NWO) within the framework of the Forensic Genomics Consortium Netherlands. The authors would like to thank Tanja de Blaeij, Toineke Westen and Bas de Jong for their comments and critical reading of the manuscript.

Back to Article Outline

Appendix A. Supplementary data 

Back to Article Outline

References 

1. Smith PJ, Ballantyne J. Simplified low-copy-number DNA analysis by post-PCR purification. J. Forensic Sci. 2007;52:820–829
   • View In Article
   • MEDLINE
   • CrossRef
2. Petricevic S, Whitaker J, Buckleton J, Vintiner S, Patel J, Simon P, et al. Validation and development of interpretation guidelines for low copy number (LCN) DNA profiling in New Zealand using the AmpFlSTR^®SGM Plus™ multiplex. Forensic Sci. Int. Genet. 2010;4:305–310
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (227 KB)
   • CrossRef
3. Forster L, Thomson J, Kutranov S. Direct comparison of post-28-cycle PCR purification and modified capillary electrophoresis methods with the 34-cycle “low copy number” (LCN) method for analysis of trace forensic DNA samples. Forensic Sci. Int. Genet. 2008;2:318–328
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (601 KB)
   • CrossRef
4. Benschop CCG, Wiebosch DC, Kloosterman AD, Sijen T. Post-coital vaginal sampling with nylon flocked swabs improves DNA typing. Forensic Sci. Int. Genet. 2010;4:115–121
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (360 KB)
   • CrossRef
5. Vandewoestyne M, Van Hoofstat D, Van Nieuwerburgh F, Deforce D. Automatic detection of spermatozoa for laser capture microdissection. Int. J. Legal Med. 2007;123:169–175
   • View In Article
   • CrossRef
6. Ballantyne KN, van Oorschot RAH, Mitchell RJ. Comparison of two whole genome amplification methods for STR genotyping of LCN and degraded DNA samples. Forensic Sci. Int. 2007;166:35–41
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (485 KB)
   • CrossRef
7. Ballantyne KN, van Oorschot RAH, Mitchell RJ, Koukoulas I. Molecular crowding increases the amplification success of multiple displacement amplification and short tandem repeat genotyping. Anal. Biochem. 2006;355:298–303
   • View In Article
   • MEDLINE
   • CrossRef
8. Gill P, Whitaker J, Flaxman C, Brown N, Buckleton J. An investigation of the rigor of interpretations rules for STRs derived from less than 100pg of DNA. Forensic Sci. Int. 2000;112:17–40
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (1093 KB)
   • CrossRef
9. Westen AA, Nagel JHA, Benschop CCG, Weiler NEC, de Jong BJ, Sijen T. Higher capillary electrophoresis injection settings as an efficient approach to increase the sensitivity of STR typing. J. Forensic Sci. 2009;54:591–598
   • View In Article
   • CrossRef
10. Balding DJ, Buckleton J. Interpreting low template DNA profiles. Forensic Sci. Int. Genet. 2009;4:1–10
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (485 KB)
    • CrossRef
11. Gill P. Application of low copy number DNA profiling. Croat. Med. J. 2001;42:229–232
    • View In Article
    • MEDLINE
12. Buckleton J. Validation issues around DNA typing of low level DNA. Forensic Sci. Int. Genet. 2009;3:255–260
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (109 KB)
    • CrossRef
13. C.C.G. Benschop, C.P. van der Beek, H.C. Meiland, A.G.M. van Gorp, A.A. Westen, T. Sijen, Low template STR typing: effect of replicate number and consensus method on genotyping reliability and DNA database search results, Forensic Sci. Int. Genet., in press.
    • View In Article
14. http://www.liacs.nl/∼hmeiland/projects/lcstat/used (January 2010–May 2010).
    • View In Article
15. Clayton TM, Whitaker JP, Sparkes R, Gill P. Analysis and interpretation of mixed forensic stains using DNA STR profiling. Forensic Sci. Int. 1998;91:55–70
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (474 KB)
    • CrossRef
16. Gross AM, Liberty AA, Ulland MM, Kuriger JK. Internal validation of the AmpFISTR Yfiler™ amplification kit for use in forensic casework. J. Forensic Sci. 2008;53:125–134
    • View In Article
    • CrossRef
17. Prinz M, Boll K, Baum H, Shaler B. Multiplexing of Y chromosome specific STRs and performance for mixed samples. Forensic Sci. Int. 1997;85:209–218
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (839 KB)
    • CrossRef