STRs are the gold standard for human identification and as such they can also be applied to C. sativa for forensic purposes. Moreover, studies that included STRs C. sativa genotyping have reached promising results for forensic purposes. In 2016, a 13-locus cannabis STR method was developed and validated according to guidelines of the International Society of Forensic Genetics and the Scientific Working Group on DNA Analysis recommendations for the use of non-human DNA. Although this STR method cannot distinguish hemp from drug type, a previous study showed that groups of cannabis seizures could be associated using phylogenetic analysis demonstrating its application for intelligence purposes. Generally, when a new method is developed, interlaboratory tests were carried out to test its robustness, reliability and reproducibility. The objective of this work was to test this newly developed cannabis STR multiplex kit through an inter-laboratory exercise with different laboratories in Europe. Up to date, this is the first collaborative exercise on STR identification of C. sativa. After approval by the Italian Forensic Geneticists society, a collaborative exercise to test a 13-locus cannabis STR was organized. Twenty-one European laboratories from different institutions, using different DNA analysis platforms, participated in the exercise.Each box contained: two cooling tablets, a self-sealing pre-PCR bag and a self-sealing postPCR bag. The pre-PCR bag contained a PCR master mix tube , a primer mix tube and sample set tubes in three separate bags. The sample set contained three aliquots of C. sativa DNA . The PCR master mix and the primer mix were prepared according to Houston et al.. Lastly, an aliquot of 5 µL of allelic ladder was contained in the post-PCR self-sealing bag. Protocols for PCR amplification and DNA genotyping were also provided. Participants used their standard DNA genotyping platforms, as well as the interpretation and reporting guidelines. Some reagents used in this exercise were kindly provided by Dr.
Running conditions were chosen based on standard HID STR genotyping protocols used by each laboratory. The different genetic analyzers used by each laboratory are displayed in Table 1. Different run parameters were applied according to each laboratory protocol. For this reason, the majority of laboratories had to adjust their bin sets appropriately. Marker panels and allelic bins files for different versions of Genemapper® software were developed and provided by the core lab to each participant laboratory. Indeed, grow tent for sale online technical support for the calibration of bins and markers according to different electrophoretic conditions was also provided. Analytical and stochastic thresholds were set according to each laboratory’s protocols and interpretation guidelines. Participating laboratories provided a table with genotyping results and the raw data sample files with the printouts of the obtained electropherograms. Heterozygous peak height ratio was calculated for a given locus by dividing the peak height of an allele with a lower relative fluorescence units value by the peak height of an allele with a higher RFU value in a heterozygous pair, and then multiplying this value by 100 to express the PHR as a percentage. For the PHR, the mean, standard deviation , median, minimum and maximum were calculated. The inter-loci balance was calculated as the ratio between the mean peak height for each locus and the mean peak height across all loci multiplied by 100. Stutter percentage was calculated based on the peak height of the parent allele . Stutter peaks that were in-between two alleles on + 1/− 1 stutter position were regarded as −1 stutter of the longer allele. Stutter peak average was determined and the −1/+ 1 stutter mean obtained from each laboratory were estimated. As part of a new forensic kit in-house validation the assessment of −1 and + 1 stutter unit repeat filter application was required. To identify if a peak was a true allele or a stutter, we applied stutter ratio filters for different STR loci. Peaks below that filters were considered stutters. Results from positive control, sample 1 and sample 2, each in triplicate, were utilized to determine the −1 and + 1 stutter ratio thresholds percentage. Each sample replicate was analyzed by the software GeneMapper ID-X ; no stutter filters were applied and a detection threshold was set to 30 RFU. RFUs from 3300 to 8000 data point were collected and analyzed.
For the statistical analysis, GraphPad Prism software ver. 9.0.2 was used. Mean of Peak Height and Parent Allele Stutter , −1 and + 1 Stutter Percentage Mean were also calculated. PHLM, PHPASLM, MSLM and PSLM were used to calculatethe Average of laboratories means and peak height and the standard error of the mean for all laboratories. The minimum peak height of parental allele with a stutter was also estimated. To determine the limits of detection for a given analytical procedure, it is necessary to determine a Minimum Distinguishable Signal , the signal at which a peak can reliably be distinguished from noise. The MDS may be considered a relative fluorescence unit or an analytical threshold for forensic purposes. Furthermore, the detection limit at 99.7% confidence is based on the Gaussian distribution of noise peaks height and this does not consider a possible asymmetric distribution where, the correct 99.87% confidence should be applied. The background noise level detected of each instrument were compared, to evaluate their impact on the general background noise. Three negative samples were amplified, run through capillary electrophoresis and analyzed by setting the GeneMapper™ software at 1 RFU as detection threshold. RFU data from 3300 to 8000 data point were collected. The GeneMapper data were exported to a txt file and then imported to an excel tool or GraphPad Prism software v. 9.0.2 for the analysis. The background instrument peak heights observed in each dye channel of each laboratory were compared. The following parameters were calculated for each laboratory: Maximum Peak Height , Average Peak Height , Standard Deviation , Limit of Detection , Limit of Quantitation , Analytical Threshold . APH and AT means were used for each dye channel to calculate the Standard Error of the Mean for each laboratory. The percentage of genotyping success was estimated to be 96% . Genotyping errors may be due to partial DNA degradation , amplification errors, stutter calculation errors, bin error and PCR artifacts such as allelic drop-out. All of these interpretation errors and artifacts can be resolved with an optimization of this methodology through an internal validation and subsequent interpretation guidelines. A locus was labeled discordant if one or more alleles were miscalled, for example, ANUCS305 locus resulted in about 10% of incorrect allele calls for both samples . A locus was labeled concordant when all alleles were correctly called, for example 4910, 9043, B05, 1528, CS1, D02, C11 and H06 loci gave a 100% of consensus allele calls among the participant laboratories . For sample 1, laboratories 2, 6, 7, 9, 11 and 13 reported incorrect allele call for a heterozygous STR loci: 9269 , 5159 , ANUCS305 and 3735 .Moreover, laboratory 6 did not produced results for three loci: 9269, ANUCS305 and 3735.
Laboratory 13 did not produce any result for loci ANUCS501 and 9269 . For sample 2, laboratories 3 and 13 encountered a problem with incorrect allele call of the 9259 heterozygous STR locus. Moreover, laboratories 5 and 6 did not produced results on loci: ANUCS501, 9269 and ANUCS305. Indeed, laboratory 6 did not produce any result on locus 3735 . The calculations of percentage of success were based on a total of 13 STR loci. Examples of typing errors are displayed in Fig. 2. The negative controls did not show any evidence of external contamination. In the case when a laboratory failed to produce results for a specific STR locus or if only one allele at a heterozygous locus was obtained, it was considered an error and not a partial result for the purpose of the study. Results with concordant calls with those of the organizers were considered correct . In summary, twenty laboratories submitted results for cannabis STRs. Eleven laboratories obtained full and concordant profiles for the two samples without errors, two laboratories obtained full and concordant profiles with one error, four laboratories completed the exercise with two errors, one laboratory committed three errors, while two laboratories completed the test with more than three errors. The success rates for cannabis STR typing ranged from 74.4% to 100% for sample 1 and from 69.2% to 100% for sample 2. Genotyping success calculations were based on the total number of loci tested by a laboratory in this study. An example of an electropherogram of this study is displayed in Fig. 3. Peak heights for the heterozygous markers were averaged and the intra locus peak height ratio was calculated. The final average PHR was calculate by the mean PHR for each heterozygous locus of the three samples, and resulted to be 78.2% . The mean inter-loci balance was estimated to be 100,00%. The present collaborative study of the Ge.F.I.-ISFG working group allowed to demonstrate the robustness and reproducibility of a 13-locus cannabis STR multiplex system for forensic DNA profiling. Overall, each participant, using various instrumentation, polymers, size standard and arrays, generated STR profiles for each of the 13 markers. Analysis of the electropherograms, by the core laboratory, showed a concordance of more than 96%. Data misinterpretation resulted in discordance in some markers; typing was more problematic with the analysis of the ANUCS501, 9269, ANUCS305 and 3735 loci. This problem will require further investigation and all of these discordant allele calls can be resolved with, an optimized bin set, more training and experience with this multiplex STR kit. An important limitation of this study is that the analysis was conducted with cannabis DNA extracts instead of cannabis plant tissue. An important source of variation may occur during DNA extraction of different cannabis plant material. It is important to note that this STR kit cannot differentiate hemp from drug type; therefore, chloroplast DNA markers should be used for this purpose. However, using STRs,indoor tent grow cannabis seizures can be genetically associated through phylogenetic analysis of a previously established database.
Lastly, a plant generated from clonal propagation can be genetically associated to its clones using this STR kit. Future plans include: a) a collaboration between laboratories to test this kit with a new synthetic allelic ladder , to increase allele coverage and to aid in more accurate allele calls, b) studies of variation based on different extraction methods, types of tissue and various storage conditions and c) comparison of different databases. A written consensus standard for C. sativa authentication would be useful for the forensic community to establish rules and interpretation guidelines. Points to be considered for a future consensus standard include: allelic ladder, stutter filter recommendations, DNA quantitation methods, a comparative C. sativa STRs database, and troubleshooting. Finally, this inter-laboratory exercise can be considered a milestone in the identification of C. sativa samples. For centuries, cannabis has been used with many different purposes, including medicinal use.The Shennong Ben Cao Jing encyclopedia, which dates back to 2900 BC in China, recommended the seeds as treatment for pain, constipation and malaria.Additionally, the plant was used along with wine to create an anesthetic effect for patients undergoing surgery.Around 1000 AC, cannabis flowers became popular in India, providing analgesia, hypnotic, antispasmodic, and anti-inflammatory effects.In the 21st century, cannabis began to be explored by Western medicine, however only plant extracts were used,and active ingredients, both from leaves and flowers were isolated.During the 20th century, the endocannabinoid system was further understood and in the 3rd edition of the US Pharmacopoeia, in 1851, cannabis was included as a treatment for gout, rheumatism, tetanus, cholera, hysteria, depression, delirium tremens, and uterine bleedings.Cannabis was available in the US pharmacies since 1845 and was available in British pharmacies for over a century,however, because of the rise of concerns by its psychotropic effects, it was removed from the US Pharmacopeia in 1941.In 1976, the United States Controlled Substances Act classified cannabis as a Schedule I drug, meaning it had no acceptable medical use and high potential for abuse.