for Forensic Science International

Elsevier Editorial System(tm) for Forensic Science International: Genetics Manuscript Draft Manuscript Number: FSIGEN-D-13-00243R2 Title: Developmental Validation of a Forensic Rapid DNA-STR Kit: Expressmarker 16 Article Type: Original Research Article Keywords: Forensic DNA analysis, Rapid identification Corresponding Author: Prof. Huaigu Zhou, Ph.D. Corresponding Author's Institution: Institute of Forensic Science, Shanghai Municipal Public Security Bureau/Key Laboratory of Forensic Evidence and Scene Technology, Ministry of Public Security/Shanghai Key Laboratory of Crime Scene Evidence First Author: Huaigu Zhou, Ph.D. Order of Authors: Huaigu Zhou, Ph.D.; Dan Wu, M.S.; Ronghua Chen, B.S.; Yan Xu, M.S.; Zifang Xia, M.S.; Yulin Guo, M.S.; Fan Zhang, M.S.; Weiguo Zheng, Ph.D. Abstract: DNA-STR analysis is widely used in the forensic science field and has important requirements on the analysis time to obtain faster inspections. The developed forensic STR Kit, referred to as Expressmarker 16 (EX16), could shorten the amplification time to a minimum of 35min. It enables 16 STR loci to be co-detected, including 13 CODIS loci, D2S1338, D6S1043 and Amelogenin loci. The kit is validated by a series of tests formed by DNA mixtures, stutter ratios, optimized PCR protocols based on annealing temperature research, species specificities, inhibitors, sensitivity, and parallel tests according to FBI QAS (2009/2011). The results demonstrated that EX16 was a useful tool for rapid criminal investigation. Suggested Reviewers:

Title Page (with Author Details)

Developmental Validation of a Forensic Rapid DNA-STR Kit: Expressmarker 16 Huaigu Zhou,1Ph.D.; Dan Wu, 1M.S.; Ronghua Chen, 1B.S.; Yan Xu, 1M.S.; Zifang Xia, 2M.S.; Yulin Guo, 2M.S.;Fan Zhang, 2M.S.; Weiguo Zheng , 2Ph.D.

1

Institute of Forensic Science, Shanghai Municipal Public Security Bureau/Key Laboratory of Forensic Evidence and Scene Technology, Ministry of Public Security/Shanghai Key Laboratory of Crime Scene Evidence, Shanghai 200083, China. 2 AGCU ScienTech Incorporation, Wuxi 214174, China.

*Detailed Response to Reviewers

Response to Reviewers Comments Dear Editor, This is the secondly revised manuscript (FSIGEN-D-13-00243) entitled “Developmental Validation of a Forensic Rapid DNA-STR Kit: Expressmarker 16” by Huaigu zhou, Wu Dan, etc. Thanks for your reviewing which makes our paper much more accurate and complete.We’ve corrected the errors and text editing mistakes word by word, and look forward to hearing from your favorable decision for this matter. Best wishes! Sincerely yours, Huai-gu Zhou Ministry of Public Security/Shanghai Key Laboratory of Crime Scene Evidence

1. Concerning the annealing study, the authors included in the materials and methods that they looked at annealing temperatures from 56°C to 63°C, which is indeed the short of range one would expect to study for a multiplex with an optimal annealing temperature of 60°C. However, the results section still only contains data on annealing temperatures from 59°C to 61.3°C over which optimal results are obtained. As the experiments have been performed it would benefit to show the data obtained at least at the 56°C and 63°C annealing temps so that the reader may be able to see how changes in annealing conditions (for example if their cycler was out of calibration) affects signals at specific loci. The authors do mention that vWA, D5S818, and FGA gradually decrease with increasing temperature but it would be good to show example data at least at 63°C in Figure 6. Also, the authors indicate that a non-specific amplification product appears in the TH01 locus at lower annealing temps, but there is no mention anywhere of the size of this peak in bases or the height of this peak relative to say the true TH01 allele. Again, including the electropherogram for the 56°C annealing temperature would help as a visual aid to the reader along with the text describing this peak. Answer: Thanks for the comments. The annealing temperature from 56°C to 63°C (56°C, 57°C, 58°C, 59°C, 61.3°C, 62°C, 63°C) was presented in supplemental figure 1 stored in another attach document named as supplemental figures, which didn’t show the genotyping profiles in the temperature ranges from 59 to 61.3, as they were not able to demonstrate some specific loci variation and figure 6 in the paper has listed the combine dyes as well. Figure 7 showed the gradual variation of THO1 as the annealing temperature rose. Besides, there was the additional description for the internal imbalance for CSF1P0, D7S820, please find the related sentences in line 2, paragraph 2 of section 3.5.2.

2. In the materials and method section the authors add descriptions of additional experiments conducted to evaluate the effect of magnesium concentration, final reaction master mix and primer pair concentration on signal and balance of EX 16. However, there are no data presented in the results section (supplemental figures or

tables would suffice) only a couple of sentences in the text. It would be useful to have some presentation of the data to show how signal at the various loci in EX 16 are affected by magnesium concentration and variations in the final 1X concentrations of reaction master mix and primer pair mix. As the authors have already performed these experiments it should be simple to generate the appropriate figures. Similarly, the authors looked at cycle number but just have one sentence saying that peak height increases (expected result). It would be good to show how locus to locus balance is affected by changes in cycle number, especially as it is known that increasing cycle number can affect this balance. This is important as some labs do increase cycle number when working with low copy number template and so it would be useful to such labs that may use this kit to know how such an increase in cycle number would affect balance. Again, this should be relatively simple to put together a figure from existing data. Answer: Thanks for the comments, sorry for the omission of the important figures about the reaction mix, primer and magnesium concentration variations, which were put in the supplemental figure documents. Please find the corresponding figures there. Supplemental Figure 2 was the presentation for the minor changes of primer concentration, and supplemental Figure 3 showed the amplification results affected by reaction mix changes, while supplemental figure 4 showed the effects performed by magnesium concentration variations.We’ve extended the magnesium concentration range from 1.5mM to 3.5mM in order to show the more details about some specific alleles. Figure 8 showed the locus to locus balance data resulted from cycle number changes. 3. The authors put a lot of effort into indicating the percent minor alleles detected in their mixture study, going so far as to split the results out by locus. It is really not necessary to go into that level of detail. Table 3 could be simplified greatly by just presenting data for "Observed amounts of Unique allele (%)" for both mixture sets rather than splitting out by loci. Answer: Thanks for the advice. Table 3 has been simplified. 4. In section 3.8, the authors claim that EX 16 has a better detection rate than

SinoFiler with semen samples on the basis of an 89.5% detection rate with EX 16 vs 87.7% with Sinofiler. A difference of 1.8%. This is a very small difference, and without any statistical analysis it is not possible to sates whether or not EX 16 has a greater detection rate. In addition, with blood stain samples a detection rate of 96.7% is reported for Identifiler and 94.9% for EX 16. A difference of 1.8%. By the criteria used for the semen samples it would appear that Identifiler has a higher success rate than EX 16 with blood stain samples. However, I doubt that either is the case and that there is no statistically significant difference between detection rates for EX 16 and the Life Technologies' kit against which it is compared, in which case I would suggest removing the wording "…and even showed high advantage when processing the semen DNA testing". Answer: Sorry for the over determined description and it had been already removed. 5. In section 3.3 there are two percent values given for stutter peaks where the allele label has been removed by the stutter filter. It is confusing to the reader as to why there are two values. One at 99.54% and one at 98.87%. Why is this the case. Answer：Sorry for our mistakes. We’ve re-calculated those not labeled stutter peaks in the first modification and excluded several stutter peaks over the stutter threshold at D13S317 which was mistakenly calculated in the initial edition. Therefor the more accurate and precise number should be 98.87%. The first paragraph should have been removed in the first modification edition, yet still stayed for our mistakes. You could see the same expression in the initial original edition. We have deleted the first paragraph now. Apologize again for our mistakes. 6. Minor point. The spelling of Aspergillus in Figure 7 is incorrect. Answer: Thanks for the kind suggestion. The spelling mistake has been corrected.

*Manuscript Click here to view linked References

Developmental Validation of a Forensic Rapid DNA-STR Kit: Expressmarker 16 Abstract DNA-STR analysis is widely used in the forensic science field and has important requirements on the analysis time to obtain faster inspections. The developed forensic STR Kit, referred to as Expressmarker 16 (EX16), could shorten the amplification time to a minimum of 35 min. It enables 16 STR loci to be co-detected, including 13 CODIS loci, D2S1338, D6S1043 and Amelogenin loci. The kit is validated by a series of tests formed by DNA mixtures, stutter ratios, optimized PCR protocols based on annealing temperature research, species specificities, inhibitors, sensitivity, and parallel tests according to FBI QAS (2009/2011)[1-2]. The results demonstrated that EX16 was a useful tool for rapid criminal investigation. Keywords: Forensic DNA analysis, Rapid identification

1. Introduction Forensic DNA analysis technology has developed rapidly since British geneticist Alec Jeffreys first introduced DNA fingerprinting in 1985. Detection efficiency has been greatly improved; the current normal sample testing comprises DNA extraction, PCR, electrophoresis, and lastly the analysis, which wholly costs 6-8 hours compared to 6-8 weeks previously. On the other hand, with the development and perfection of PCR technology, fluorescent labeling as well as capillary electrophoresis, applications of DNA-STR genotyping in forensic human biological screening has been greatly expanded and its reliability has been significantly increased as well [3]. At present, the DNA-STR genotyping has become one of the most scientific and technological techniques in forensic science. It is the most practical form of technology in forensic science with the results receiving wide recognition. It is one of the most important techniques used in criminal investigation and court trials. DNA-STR genotype technology generally includes DNA extraction, PCR amplification and electrophoresis analysis. DNA extraction requires more than 1 hour, and PCR amplification 3 to 4 hours. Electrophoresis analysis takes more than 1 hour. Thus, the whole DNA-STR genotype process requires 6-8 hours [1]. In order to shorten the total time needed, faster methods of DNA analysis have become a new research focus[4-6]. At present, DNA analysis research mainly focuses on faster methods of DNA extraction and rapid electrophoresis analysis. For example, Mercier and Gaucher [7] and Bu Y et. al., [8] reported a direct PCR technology from whole blood samples without DNA extraction. On the other hand, Promega (Madison, WI, USA), Life Technologies (Carlsbad, CA, USA), and AGCU ScienTech (Wuxi, Jiangsu, China) have already developed forensic analysis kits that enable direct amplification of blood stains. This method eliminates the need for DNA extraction, cutting down amplification time. Research on faster electrophoresis methods focuses on microfluidic chip technology [9 10].

In this study, EX16 STR kit was designed to shorten the amplification time and was made up of 15 gene loci including 13 CODIS loci and Amelogenin loci. The developmental validation was carried on the EX16 STR kit according to the FBI QAS (2009/2011)[1,2].

2. Materials and Methods 2.1 DNA Samples

1103 DNA samples, stored in filter papers or FTA cards that could keep human DNA for almost 1 to 10 years; 1757 human samples collected from daily cases, including blood (stains), semen (stains), saliva (stains), bones, hair, muscles, exfoliated cell samples, etc.; Samples from dogs, pigs, ox, sheep, cats, chicken, rats, rabbits, fish and E. coli; Standard DNA 9947A and 9948 (Promega ,Madison, WI, USA) 2.2 Sample preparation Samples from the filter paper or FTA cards were punched using a 1.2mm BSD600- DUET stiletto instrument (BSD, Australia) and placed into 96-well plates for direct PCR amplification by EX16. Besides, biomagnetic seperation techniques (Changchun Bokun Biotech Co., Chuangchun,China) were applied in all the samples mentioned in 2.1, except the standard DNA 9947A and 9948, to obtain the pure DNA prepared for the amplification by Identifiler (Life technologies, USA), Sinofiler (Life technologies, USA), and Profiler plus (Life technologies, USA). 2.3 STR loci selection

Except for the amelogenin, EX 16 selected 13 CODIS loci, which was deeply studied and applied in the current commercial kits. Besides, the other two loci D6S1043, D2S1338 were complemented in this kit because of their high polymorphic features; especially D6S1043 presented a robust power of discrimination rate in Chinese Han [11-13]. 2.4 Primers

Primer design requires high specificity, and excludes analytical errors caused by homology. The set of designed primers should act according to unified parameters, with a consistent temperature in order to achieve a stable and balanced amplification sensitivity and efficient amplification in multiple PCR processes. Extensive search on the current studies were made to conduct the primer design as the powerful reference and strong evidence to prevent the primer binding to the existing and mutant region [14-16]. Generally the primers were designed at the surrounding conserved region of the STR loci and one of the primer pair was fluorescent labeled. Primer amplification specificity was based on BLAST function of Gene Bank, and primer species specificity of genetic loci was ensured through comparing with the genome database data of other species. As interference among different primer sets could exist in the multiple PCR amplification systems here, large numbers of primer

sets were designed according to the need of each test method. These primer sets were further analyzed using a professional primer design software Oligo 6.0 (Premier Biosoft International, Palo Alto CA, USA). The best primer combinations were finally chosen based on multiple primer interference and balance in multiple PCR systems as well as software analysis. The following factors are main considerations during the PCR primer design: primer length had better be in the range of 19-28bp; Tm value corresponding to the primers in different loci should keep in a fixed range (around 69℃); Strict primers design parameter should avoid primer dimmers and hairpin structure; No more than 3 base pairings on 3' end of primer; Sequence homology of other parts in the same sequence. Length of PCR fragments was less than 450bp. Primers were synthesized by Sangon Biotech (Shanghai, China). or AGCU ScienTech Incorporation (HPLC level). Primer purity is more than 99%, and 1.2 by HPLC and A260/A280 analysis respectively. 16 loci are divided into four groups in four-color dye (Figure 1). 2.5 PCR amplification

The Reaction master mix was made of 2.5 mM magnesium ion, 0.25 mM dNTPs, 10mM Tris-HCl, 50 mM KCl, 10 mM Betaine, 0.008% NaN3, and 0.3mg/ml Cyclodextrin, etc. Besides, the final reaction system 25μl combined by 10μl Reaction Mix, 5.0 U Hot start Q-Taq polymerase, 5μl Primer mix, and the amount of pure DNA for 0.5-2ng (the optimal amount was 1ng); for the FTA card or the filter paper, the amount was controlled by the punch of 1.2mm. Cycling parameters for PCR amplification were as follows: 94℃ for 1min; 98℃ for 5sec, 15 sec at annealing temperature 59℃, and a 10 sec extension at 72℃, 30 cycles for purified DNA, while 28 cycles for direct amplification; 72℃ 10min; hold at 4℃ for further analysis. 2.6 Sample electrophoresis and data analysis

Amplicons were analyzed with Genetic Analyzer 3130XL (ABI, USA) by running electrophoresis of the mixture combined with 1μl PCR product (amplicons/allelic ladder), 9.5μl Hi-Di formamid and 0.5μl internal size standard of the marker size 75bp, 100bp, 139bp, 150bp, 160bp, 200bp, 300bp, 340bp, 350bp, 400bp, 450bp, 480bp and 500bp, for the injection time of 12 seconds at 3KV, after denaturation process by heating the samples at 95℃ for 3minutes. The required data was imported into the GeneMapper software version 3.2, and analyzed by 150 relative fluorescent units (RFU) threshold.

2.7 Sensitivity study

The kit’s sensitivity was tested by a series of dilutions of 9947A, with decreasing concentrations starting from the maximum 0.2ng per reaction. 9947A was diluted into several concentrations: 0.2ng, 0.125ng, 0.1ng and 0.05ng per PCR, systems, the PCR products were under the normal capillary electrophoresis after the PCR process. Three parallel tests for each dilution were necessary for the final calculation. The number of successful genotyping allele peaks was counted to show the percent of the full profile. 2.8 Precision study and Sizing accuracy

Precision study was performed to run the capillary electrophoresis of the 32 allelic ladder samples on 3130XL Genetic Analyzer to demonstrate the allelic precision. The sizing accuracy between the STR loci size and the corresponding allelic ladder was calculated from 42 sample results. 2.9 Stutter ratio calculation

Proportion of the stutter peaks relative to the main alleles was measured by dividing the stutter peaks by the main peak height. 1103 samples were used for the stutter ratio calculation. 2.10 Mixture studies

Forensic casework samples often include more than one individual’s DNA. Hence, the mixture study is made to determine the kit’s identification capability with DNA mixture samples. The commercial sold DNAs, 9947A (female) and 9948(male) were chosen for the mixture study samples. The major and the minor DNA varied by 9:1, 2:1, 1:1 and the reciprocals, with the final amounts 1ng. For example, the 9:1 mixture contained 0.9ng 9947A and 0.1 ng 9948 per reaction. At each mixture ratio, it was performed three reactions to ensure the accuracy of the results analysis. And the PCR thermal cycle value was 30 cycles. 2.11 PCR-based studies

The comparison between the rapid protocol and the normal was performed to demonstrate if the rapid protocol affected the amplification efficiency. The rapid protocol followed the thermal parameters provided by 2.5, while normal protocol extended the denaturing, annealing and extension period to 1min per cycle, and both protocols were 30 cycles, amplified with the standard DNA 9947A. 0.1ng 9947A was amplified with 0.75*, 0.9*, 1*, 1.1*, 1.25* Reaction Mix and Primer mix to simulate the minor changes in volume due to pipetting error.

To evaluate magnesium in the reaction mix, the concentration variations of magnesium 1.5mM, 2mM, 2.5mM, 3mM. 3.5mM were studied while other components in the reaction mix stayed. Cycle number varied from 28 to 32 was used to detect the signal and balance of the profile. Besides, the annealing temperature, ranging from 56ºC to 63ºC was studied to evaluate if the EX16 system has the capability to withstand the slight temperature variation. In addition, cycle number and annealing temperature study should be under the standard protocol, amplified with 1ng 9947A. 2.12 Species specificity

The species specificity of Expressmarker 16 (EX16) system was evaluated by a series of tests on non-human DNA to see if other biological sources mixed with the human DNA at the criminal spot would affect final amplification result. The animal and bacterial samples usually collected from the work case are tested under EX16 identification system. The other selected species DNA samples are extracted from dogs, pigs, oxen, sheep, cats, chicken, rats, rabbits, fish, E.coli, Aspergillus niger and Candida Albicans. 2.13 Case-type samples: inhibited samples

The samples containing various kinds of inhibitors were used to simulate casework samples to validate the anti-interference capability of the systems towards the common identified inhibitors in the criminal scene. Heme, hemoglobin, indigo, humic acid, and calcium ion, were co-amplified with the normal PCR reagents to identify EX16 tolerance towards the various inhibitors. The varying concentrations for all the inhibitors in the reaction systems were as follows: Heme (μM): 40, 60, 80, 90, 100 Hemaglobin (μM):200, 300, 400, 500, 600 Indigo(mM): 6, 7.5, 9, 10.5, 12, 12.5, 13, 15 Humic acid (ng/μl): 5, 10, 15, 20, 25 Calcium ion(mM): 0.75, 1, 1.25, 1.5, 2 2.14 Parallel tests

Parallel tests were carried out according to the instructions of Identifiler kit, Sinofiler kit, Profiler Plus Kit. In the concordance study, extracted DNAs from 1103 FTA or filter samples were amplified by Indentifiler kit, and EX 16 directly amplified the FTA cards of filter paper. The comparison for the detection rate between the kits were carried out by the 1757 work case samples, to validate the PCR efficiency of EX16 aimed at the work case samples.

3. Results 3.1 Sensitivity

Using 150RFU as the detection limit, a full profile was obtained after the electrophoresis with the Genetic Analyzer 3130XL and analyzed by GeneMapper software above the concentration of 0.1ng 9947A /reaction. TH01, one of the CODIS loci, failed to be fully detected by reaching the peak height above 150RFU at the amount of 0.05ng per reaction. Hence, the sensitivity was determined to be 0.1ng/25μl (Figure 2). 3.2 Precision and accuracy

Standard deviation and the average size for each allelic ladder was calculated and was found to be less than 0.2 bases(Figure 3). The sizing accuracy and precision of STR allele peaks were calculated on Genetic Analyzer 3130XL. 42 random samples were tested for the statics of all the STR allele sizing precision compared to the corresponding allelic ladder. The absolute value was kept for the calculation of each allele’s nucleotide difference. Sizing accuracy between the allelic ladder alleles and the corresponding alleles revealed that all the alleles were sized within +0.2 nt (Data was not shown), and fully complied with the QAS (2009/2011) requirements [1]. 3.3 Stutter ratio calculation

Stutter is the presence of a peak that is one repeat smaller than the true peak. It is an expected outcome of the PCR process, which is caused by slippage of the Taq Polymerase during the elongation step [17-19]. As stutter peaks are inevitable with any STR kit, it is better to do a series of experiments on stutter ratio to ensure that stutter peaks appearing in the forensic STR detection system would not affect the genotyping results, and in another consideration to provide an effective reference to set the recommended filter values of each STR loci. Stutter and its relative allele peak heights were characterized and calculated to form a table to show the stutter properties of more than 1,000 blood samples (Table 1). The meaningful data was selected from all the loci that had a peak heights in the range varied from 500RFU to 7000RFU. At the recommend stutter filter values showed by table 1, only D13S317 and VWA had the stutter peaks exceeded the labeled line, and approximately 98.87% stutter peaks were not labeled. 3.4 DNA Mixture Study

Forensic casework analysis usually encounters mixture samples that contain the

DNA of more than one individuals [20]. Hence, an artificial mixture combined by female DNA 9947A and 9948 was created to verify the ability to make a difference to the forensic study in this validation. The varying ratio for the two components directly affects the final profile (figure 4). The minor contributor DNA changed as the mixture ratio transformed. Besides, 9947A and 9948 had shared alleles of each locus, according to the results analyzed by EX 16 (Table 2). We separated the test samples into 2 set, set A and set B by the mixture ratio, i.e. the minor components of set A was 9948 while the major contributor was 9947A. In set A, The minor contributor 9948 showed the unique alleles at loci D3S1358, CSF1PO, D18S51, FGA, Amelogenion, TPOX, TH01, D5S818, D8S1179 and D21S11; Set B represented its minor contributor alleles different from the major DNA 9948 at loci D3S1358, CSF1PO. D18S51, FGA, D7S820, D16S539, D6S1043, D2S1338, and VWA (D13S317 was not selected for 9947A and 9948 were both homozygotes at this locus). The percentage of the minor contributor at each locus was calculated to prove that the supposed peak of the DNA mixture could be fully detected confined to the ratio of 4:1 (Table 3). 3.5 PCR based study 3.5.1 Comparison of the rapid protocol and the normal protocol

The peak comparison between the rapid and normal protocol revealed that the rapid protocol ensured the regular amplification of 9947A. Except for the locus D6S1043, the peak heights of other loci with rapid protocol were neither more or less higher than the normal ones and obviously some STR loci (D2S1338, D8S1179) even presented higher peaks (Figure 5). 3.5.2 Annealing temperature study

The appropriate annealing temperature was confined to the scope between 59.0ºC and 61.3ºC to ensure the stability and accuracy of genotyping results (Figure 6), as the precise and accurate profiles could be observed at this annealing temperature. The peak heights of VWA, D5S818, and FGA gradually decreased as the temperature rose after 61.3ºC, and dropped a lot after it turned to 62ºC. The obvious internal imbalance was discovered in several alleles at 56ºC. The imbalance rates were 68%, 62.9%, 62.5% (ratios of the peak heights of two alleles at one locus), with the corresponding alleles CSF1PO, D21S11, D7S820, respectively. And it was effectively improved after the annealing temperature rose to 57ºC (supplemental figure 1). TH01 would leave an unexpected allele with the fragment size 193.09 bases. Figure 7 presented the exact changes for locus TH01, which demonstrated that the non-specific allele was increasingly lower as the annealing temperature rose, and fully disappeared at 59ºC. 3.5.3 Reaction components

EX16 was reliable to handle the minor changes of the primers, which stayed almost the same as the primer volume varied (supplemental figure 2). However, D6S1043, CSF1PO, D18S51, the large amplicons, dropped when the amounts of reaction mix increased to 25% above (supplemental figure 3). During the magnesium variation, the larger amplicon peak of allele 18 for D6S1043 could not be recognized at the magnesium concentration 3mM and fully undetected at 3.5mM. Besides, D18S51 lost at the highest magnesium concentration 3.5mM. The ski-slope trend of the profiles became obvious as the magnesium concentration rose (supplemental figure 4). As the complicated reaction mix also contained magnesium, it might indicate that magnesium was significant to keep the locus to locus balance, The optimized concentration would be 2.5mM in the final reaction system. 3.5.4 Cycle number

The peak heights increased as the cycle number rose as expected, yet the locus to locus balance got worse, obviously presented in the loci TH01, D2S1338 and CSF1PO as their peak heights increased a lot (figure 8). Off-scale allele peaks occurred at D3S1358 when cycling number turned to 32. (Supplemental figure 4 showed the whole profiles of the cycle number variation). 3.6 Species Specificity

The species specificity was studied to testify that other biological sources didn’t interfere with the ability to obtain reliable results on samples recovered from crime scenes[20]. Complex samples might contain DNAs from dogs, pigs, oxen, sheep, cats, chicken, rats, rabbits, fish, or kinds of microorganisms (E.coli, Aspergillus niger, Candida Albicans). The most common species were used to test the kit’s specificity and the results didn’t show any allele peak (Figure 9). 3.7 Inhibited Samples

The inhibited samples had to simulate case-type samples usually rich in this kinds of inhibitors. The results indicated that EX16 had certain resistent capability to the representative inhibitors (Table 4). The tolerance of the kit decreased as the inhibitor concentrations increased, and full profile was observed in the inhibitor concentration ranges including heme 0-40μM, Hemoglobin 0-500μM, Indigo 0-12.5mM, Humic acid 0-20 ng/μl, and Ca2+ 0-750μM. Generally, the peak heights decreased as the inhibitor concentrations rose and mostly the large amplicons dropped first. 3.8 Parallel tests

To evaluate the concordance between EX16 Kit and the commercial kits, Identifiler was selected to perform the positive control. The two kits had the fully concordant

genotyping results. EX16 kit was compared with the commercial efficient forensic analysis kit Identifiler, Sinofiler, and Profiler Plus to see the differences of detection rate. As the detection rate indicated the amounts of full-profile-obtained samples, it revealed that EX16 had no difference with the other three kits regarding to the detection rate for various kinds of samples. 1000 genotyping statics chosen from EX16’s electrophoresis analysis for 1103 FTA cards or blood filter papers were used to obtain the Combined Matching Probability, Cumulative Probability of Exclusion, Combined Paternity Index(Table 6), and the calculation formula referred to Advanced Topics in Forensic DNA Typing Methodology[20]. 4. Discussion

EX16 detection system was designed for forensic criminal identification applied to workcase inspection for its robust discrimination rate and rapid detection process by shortening the amplification time to one hour. With globalization, international travels are easier and have made the world become smaller and more accessible. It is highly possible that a criminal may be shopping in country B in the afternoon after conducting a crime in country A in the morning. Therefore, fast detection and timely discovery of criminals has become an urgent need to protect people’s safety, and thus EX16 will play an important role in the case detection. In this study, the developmental validation was performed based on the FBI QAS (2009/2011). The STR kit was proved to be a useful tool after series of experimental studies in sensitivity, sizing accuracy, stutter ratios, DNA mixtures, inhibited samples, and species tests. In addition, the comparison about the protocol selection and parallel tests with sold commercial kits excluded the worrying about low detection rate and poor amplification efficiency caused by the fast PCR process. The selected STR loci of EX16 were high genetic deviations mainly pointed to Chinese Han People and would take great advantages in the Chinese criminal inspection. References [1] Quality Assurance Standards(QAS) for DNA Databasing Laboratories.(2009). Available at . [2] SWGDAM. (2010). SWGDAM interpretation guidelines for autosomal STR yuping by forensic DNA testing laboratories. Available at . [3] Vallone PM, Hill CR, Butler JM. Demonstration of rapid multiplex PCR amplification involving 16 genetic loci. Forensic Sci Int Genet. 2008 Dec;3(1):42-5. [4] Liu P, Seo TS, Beyor N, Shin KJ, Scherer JR, Mathies RA. Integrated portable

polymerase chain

reaction-capillary electrophoresis microsystem for rapid forensic short tandem repeat typing. Anal Chem. 2007 Mar 1;79(5):1881-9. [5] Liu P, Yeung SH, Crenshaw KA, Crouse CA, Scherer JR, Mathies RA. Real-time forensic DNA analysis at a crime scene using a portable microchip analyzer. Forensic Sci Int Genet. 2008 Sep;2(4):301-9. [6] Yeung SH, Greenspoon SA, McGuckian A, Crouse CA, Emrich CA, Ban J, et al. Rapid and high-throughput

forensic short tandem repeat typing using a 96-lane microfabricated capillary array electrophoresis microdevice. J Forensic Sci. 2006 Jul;51(4):740-7. [7] Mercier B, Gaucher C, Feugeas O, Mazurier C. Direct PCR from whole blood, without DNA extraction. Nucleic Acids Res. 1990 Oct 11;18(19):5908. [8] Bu Y, Huang H, Wu HP, Zhang XD, Zhou GH, Cui YX, et al. Direct multiplex PCR from whole blood for rapid detection of Y chromosome microdeletions. Zhonghua Yi Xue Yi Chuan Xue Za Zhi. 2008 Aug;25(4):406-9. [9] Qin J, Fung Y, Lin B. DNA diagnosis by capillary electrophoresis and microfabricated electrophoretic devices. Expert Rev Mol Diagn. 2003 May;3(3):387-94. [10] Liu D, Zhou X, Zhong R, Ye N, Chang G, Xiong W, et al. Analysis of multiplex PCR fragments with PMMA microchip. Talanta. 2006 Jan 15;68(3):616-22. [11] Guo H, Lin Y, Liu Y, et al. application of D6S1043 and D12S391 loci in forensic paternity testing, Fa yi xue za zhi 2007;23(5):345-346. [12] Yuan l, Jiang C, Ye J, et al. Development of a 10-locus STR fluorescent-multiplex system for forensic purposes. China Journal Forensic MED, 2012; 27(4): 181-189 [13] Budowle, B., Collins, P.J., Dimsoski, P., Ganong, C. K., Hennessy, L.K., Leibelt, C., Rao-Coticone, S., Shadravan, F. and Dennis J. Reeder. (2001) Population data on the STR loci D2S1338 and D19S433. Forensic Science Communications. Vol 3(3):1-4. [14] Carolyn R. Hil, Margaret C. Kline, et al. Concordance Study Between the AmpF‘STR MiniFilerTM PCR Ampliﬁcation Kit and Conventional STR Typing Kits. J Forensic Sci, 2007: 870-3. [15] Chen WJ,Li Y,Wu XJ, et al. Analysis of allelic drop-out at short tandem repeat loci. Zhonghua Yi Xue Yi Chuan Xue Za Zhi,2012:360-3 [16] Hammond H;Jin L;Zhong Y Evaluation of 13 short tandem repeat loci for use in personal identification applications 1994(55) [17] Levinson G, Gutman GA. Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol Biol 1987;4:203–21. [18] Schlotterer C, Tauz D. Slippage synthesis of simple sequence DNA. Nucleic Acids Res 1992;20:211–5. [19] Walsh PS, Fildes NJ, Reynolds R. Sequence analysis and characterization of stutter products at the tetranucleotide repeat locus vWA. Nucleic Acids Res 1996;24:2807–12. [20] Jonh M. Bulter. Advanced topics in forensic DNA typing methodology. San Diego: Academic Press, 2011:182-7.

Figure

Figure1. The assignment for EX16 STR kit.

Figure 2. 9947A of varying concentration at the 25μl reaction system to detect the sensitivity. The data shown by vertical ordinate is the proportion of the meaningful alleles (the peak height above 150RFU) in the full profile, while the horizontal ordinate stands for the increasing 9947A concentration in one reaction.

Figure 3. Sizing variation of all allelic ladders on 3130XL Genetic Analyzer (n=32). The X axis stands for the size length of the allelic ladders for each locus, while the Y axis represents the variation of the standard deviation corresponding to each allelic ladder.

Figure 4. Representative EX16 Electropherogram of DNA mixture formed by 9947A and 9948 totally amount to 1ng. Panels(top to bottom) show the genotyping results in a varying proportion of 9947A and 9948, by 9:1, 2:1, 1:1, 1:2, 1:9.

i Figure 5.comparison of peak height for each locus between rapid protocol and normal protocol. The locus peak height is a sum of the heterozygote peaks and the right single peak for a homozygote site. The rapid protocol is the EX16 protocol presented in proceeding part, while the normal protocol was set as 95ºC 2min, 30cycles of 94ºC 1min, 59ºC 1min, 72ºC 1min, and the last extension of 60ºC for 60min. The reaction system is prepared at the volume of 25μl, with 1ng 9947A per reaction.

Figure 6. Genotyping profile of 9947A. Panels (from top to bottom) are the profiles where all the allele peaks were observed in a relative balanced condition with the annealing temperature 59.0ºC, 59.6ºC, 60.1ºC, 60.5ºC,

61.3ºC.

Figure 7. Allele variation of locus TH01. Panels from top to bottom was the corresponding locus profiles in the annealing temperature 56oC, 57oC, 58oC, 59oC.

Figure 8. Locus to locus peak heights changes under different cycles.

Figure 9. Species electropherograms panel. Electrophoresis results (from top to bottom)different species, dogs, pigs, ox, sheep, cats, chicken, rats, rabbits, fish, E.coli, Aspergillus niger and Candida albicans. The reaction system amplified 5ng DNA for kinds of species samples.

Table

Table 1. Observed stutter for the EX16 STR Loci. Stutter values are calculated in percentage terms by dividing stutter heights by its relative peak heights and get a stutter range from the minimum stutter value to the maximum. Stutter mean and SD values are acquired by the AVERAGE and STDEV format to the stutter data. Recommended filter values are established for each locus by taking the average value and adding three standard deviations (SD).

Locus D3S1358 D13S317 D7S820 D16S539 D6S1043 TPOX TH01 D2S1338 CSF1PO vWA D5S818 FGA D8S1179 D21S11 D18S51

Stutter range（%） 4.41-16.4 2.15-16.7 2.77-10.64 4.07-13.02 5.55-14.88 1.50-7.20 1.81-7.61 7.01-16.47 0.65-14.02 0.99-17.55 1.92-17.19 3.91-14.79 6.12-17.16 1.50-18.91 6.00-13.88

Stutter mean(%)

SD

Recommended filter

11.57 5.60 6.14 7.47 9.79 3.97 4.71 11.46 7.88 9.04 9.12 9.97 10.20 11.74 9.30

2.54 2.84 2.28 2.54 2.81 1.51 1.48 2.82 2.38 2.60 2.77 2.73 2.82 2.99 2.66

0.1919 0.1412 0.1298 0.1509 0.1822 0.0850 0.0915 0.1992 0.1502 0.1684 0.1743 0.1816 0.1866 0.2071 0.1728

Table 2. Genotyping results of 9947A, 9948 and DNA mixture combined by the two kinds of DNAs by EX 16

locus

Genotyping results

Amelogenin D3S1358 D13S317 D7S820 D16S539 D6S1043 TPOX TH01 D2S1338 CSF1PO VWA D5S818 FGA D8S1179 D21S11 D18S51

9947A

9948

DNA mixture of 9947A and 9948

X 14,15 11 10, 11 11, 12 12, 18 8 8, 9.3 19, 23 10, 12 17, 18 11 23, 24 13 30 15, 19

X, Y 15,17 11 11 11 12 8, 9 6, 9.3 23 10, 11 17 11, 13 24, 26 12, 13 29, 30 15, 18

X, Y 14, 15, 17 11 10, 11 11, 12 12, 18 8, 9 6, 8, 9.3 19, 23 10, 11, 12 17, 18 11, 13 23, 24, 26 12, 13 29, 30 15, 18, 19

Table 3. The percentage of unique minor contributor alleles detected at each mixture ratio.The results obeys this formular: Observed amount of unique allele(%) = observed unique allele loci counts/ ideally observed unique allele counts. Mixture ratio/set

Set A ( 9948 as the minor contributor)

Set B (9947A as the minor contributor)

9:1

40

33.33

4:1

100

100

1:1

100

100

Table 4. Highest concentration of PCR inhibitors where full profiles were observed. All were final concentrations in the PCR reaction systems with 1ng 9947A in total 25μl reaction systems.

Inhibitors

Concentration

Heme Hemoglobin Indigo Humic acid Ca2+

40μM 500μM 12.5mM 20ng/μl 0.75mM

Table 5. Rapid PCR group and parallel group tests. Typical samples from 899 bloodstains, 474 saliva, 57 semen, 317 exfoliated cells and 10 costal cartilage are performed under the rapid amplification protocol to do a comparison with ABI kit. (Detection rate was the amount of full profile obtained samples divided the number of whole samples. All the samples were extracted by magnetic bead purification techniques, including 899 blood stains, 474 Saliva, 57 Semen, 317 exfoliated cells, 10 costal cartilage)

Kit

Blood stains

Saliva

Semen

Exfoliated Cells

Costal cartilage

(899)

(474)

(57)

(317)

(10)

EX16

Identifil

EX16

er Detection

94.9%

96.7%

Rate

Sinof

EX16

iler 88.8

88.2

%

%

Sinof

EX16

iler 89.5%

87.7

Profiler

EX16

Identifiler

100%

100%

plus 53.0%

50.8%

%

Table 6. Calculated genetic parameters based on the genotyping results of 1000 samples of Chinese Han Population Genetic parameters (N=1000)

Data

Combined Matching Probability 7.52739×10-16 Cumulative Probability of Exclusion

0.999994372

Combined Paternity Index

603000.2

Additional Files Click here to download Additional Files: supplemental figures.doc

Supplemental figure 1. Panels 1.1-1.7 showed genotyping profile of 9947A by the varied temps 56oC, 57oC, 58oC, 59oC, 61.3oC, 62oC, 63oC. 1.1 56oC

1.2 57oC

1.3 58oC

1.4 59oC

1.5 61.3oC

1.6 62oC

1.7 63oC

Supplemental figure 2. The 9947A profiles for the varying primer concentrations.Panels 2.1-2.5 were under the increasing primer concentrations 0.75*, 0.9*, 1*, 1.1*, 1.25*. 2.1 0.75* primers

2.2 0.9* primers

2.3 1*primers

2.4 1.1*primers

2.5 1.25*primers

Supplemental figure 3. The 9947A profiles under the varying reaction mix volume in a total 25μl reaction system. Panels 3.1-3.5 were the presentation of the amplification results with the increasing reaction volume 0.75*, 0.9*, 1*, 1.1*, 1.25*. 3.1 0.75*reaction mix

3.2 0.9*reaction mix

3.3 1*reaction mix



Supplemental figure 4. The 9947A profiles for the magnesium concentration variation. Panels 4.1-4.5 showed the loci changes with the increased magnesium concentration 1.5mM, 2mM, 2.5mM, 3mM, 3.5mM in the PCR reaction systems. 4.1 1.5mM

1.2 2mM

1.3 2.5mM

1.4 3mM

1.5 3.5mM