If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Performance comparison of Copan, ForensiX and Sarstedt forensic swabs.
•
Sarstedt swab was excluded due to the strikingly poor performance.
•
Higher DNA yields for ForensiX cotton swabs than for Copan nylon flocked swabs.
•
Significant differences between different police and laboratory operators.
•
Volume of sampling solution significantly influenced DNA recovery efficiency.
Abstract
The collection of DNA traces marks the first step determining the success of genetic analysis. This study aimed to identify and validate a suitable alternative to the currently used ForensiX Evidence Collection Kit containing a cardboard box for swab storage. This box has to be folded at the crime scene, which is time-consuming and carries the risk of potential contamination and handling difficulties.
A collaboration study involving three police departments and one laboratory for forensic genetics was performed to compare the currently used swab against three challenger swabs: ForensiX SafeDry, Copan 4N6FLOQSwab™ Genetics and Copan 4N6FLOQSwab™ Crime Scene. Mock samples consisted mainly of touch DNA, but also blood, saliva and semen were applied to twelve items with different surfaces. Every organisation contributed with three DNA collectors, whose individual collection efficiencies were investigated. The challenge of preparing homogenous traces, especially touch DNA, was addressed by enhancing hand contact frequency and sampling area. As a further part of the swab comparison study, we describe for the first time the influence of different swabbing solution volumes on the sampling efficiency of the different swabs.
The application of touch DNA was also tested for a further swab type, the Sarstedt Forensic Swab, which yielded such low DNA concentrations that it was excluded from the collaboration study. The Copan Genetics and Copan Crime Scene swabs yielded significantly lower DNA concentrations than the currently used ForensiX Evidence Collection Kit and ForensiX SafeDry swab. The inter-individual performance results of the operators revealed significant differences in sampling skills. Comparing different swabbing solution volumes showed higher DNA yields or no significant difference for the ForensiX Evidence Collection Kit and ForensiX SafeDry than the Copan Genetics, depending on the item or trace type swabbed.
Our results highlight the importance of validating first-step components that are decisive to the success of DNA typing in the context of specific sampling procedures and laboratory methods. Also, the significance of individuals' securing variations, principally unknown for crime scene investigation and laboratory teams, is emphasised for the first time, offering a practical approach for improving and training DNA collecting activities and ensuring the optimal securing evidence process. These findings increase the knowledge of impacts on DNA collection and, thus, benefit other laboratories and forensic services, particularly when using the same extraction methods.
Besides others, DNA evidence collection plays a vital role in crime scene investigation and the detection rate of crimes. In forensic genetics, methods, technologies and knowledge have evolved tremendously, but Locard's exchange principle remains the same: every contact leaves a trace [
]. At a crime scene, biological material left behind by the perpetrator is collected and analysed by crime scene investigators and forensic experts, respectively. The interpretation of a person's trace contribution finalises in a written report for the public prosecutor and the judicial authority, respectively. Therefore, the effective collection of target DNA left by an individual marks the initial step of genetic analysis and fundamentally determines the success of DNA profiling.
Various sampling techniques are known for biological traces, and several studies have addressed the performance as a function of swab types [
]. This continuing discussion about the optimal securing approach for forensic evidence probably reflects only a part of diverse sampling strategies, such as how to hold the swab or how much pressure to apply [
] have become commercially available over the last decades. According to the manufacturers and suppliers, the unique characteristics of each given swab promise to improve swab collection performance. In addition, the secured biological material is further exposed to, e.g., the swab solutions used [
]. Moreover, these factors impact the quality of DNA, not only in the short term but also over a more extended storage interval. However, to date, only a limited number of studies have been published addressing the DNA preservation for various swabs, and none of them includes a storage period longer than one year [
Next, the extraction platforms or laboratory-specific methods significantly influence the DNA recovery efficiency from the swabs' material. For example, swab head characteristics can negatively affect downstream DNA analysis [
]. Therefore, the right combination of swab type, laboratory-specific methods and the extraction platform as an entire system is key to better performance. Thus, forensic laboratories should consider and carefully align their preferred swab type with their sampling and laboratory processes [
] and the sampling procedure used by the crime scene investigators. With regard to the laboratories, however, many of them, including us, use swabs for historical or financial reasons without being aware of the given implications of the chosen swab or potentially more efficient alternatives.
Besides a given collection technique or model chosen, the biological material can further complicate effective recovery success. For example, in early forensic investigations, principally only blood, semen and saliva were secured. Nowadays, DNA from touched objects (known as touch DNA) has become increasingly important, with some reporting even a share of 85% being analysed in forensic laboratories [
]. Apart from the invisible nature of touch DNA, its composition is complex and consists of skin cells, cell-free DNA, fragment-associated residual DNA, endogenous and exogenous nucleated cells, and anucleated corneocytes [
]. In addition, the amount of deposited touch DNA depends on variable factors such as duration and frequency of contact, substrate properties, and environmental conditions [
]. Furthermore, the shedding status of an individual, which describes the tendency to lose high amounts (good shedders) or low amounts (poor shedders) of touch DNA [
The currently used ForensiX Evidence Collection Kit (Prionics AG, Schlieren-Zürich, Switzerland) includes a cotton swab head and cardboard box for transport. This sampling system has been used for more than two decades. However, folding the cardboard box at the crime scene is prone to contamination and highly time-consuming, which further motivated the search for a swab equivalent or more efficient in terms of DNA quantity and quality outcome but superior in handling. To our knowledge, there is no literature covering the drawbacks of cardboard boxes during sampling. Thus, the handling aspect was crucial in this study, as an efficient and straightforward securing activity is less prone to contamination when conditions outside a laboratory are not optimal (e.g., at night, poor lighting conditions, fluctuating temperatures, and weather conditions).
This paper reports the findings of a large-scale collaboration study on the DNA recovery efficiency of different swab types. For this study, swabs with the criteria of being DNA free [
], potentially more efficient and convenient in terms of handling as outlined in Section 2.1, were selected from the swab types commercially available in 2018. Given the many variables that can affect reliable DNA transfer [
], a suitable method was developed to apply sufficiently homogenous amounts of touch DNA to predefined substrates. In addition to the central focus on touch DNA, the biological fluids blood, saliva and semen were included. The aforementioned complexity for providing mock touch DNA samples was addressed by designing a large sample size setup and carefully selecting realistic objects typically encountered in routine casework. Trained crime scene investigators and laboratory operators were involved in the sampling process; thus, their inter-individual influence on DNA collection could be compared. Furthermore, different volumes of liquid solution used for sampling per swab type and their influence on DNA recovery were analysed.
Based on the research results, a change of the swab type with the same or higher DNA output but improved and safer handling conditions for the performing laboratory and the affiliated collaborators was aspired.
2. Materials and methods
2.1 Selection of swabs and handling criteria
Swabs challenging the currently used ForensiX Evidence Collection Kit were selected based on general swab properties and handling features during packaging, sampling and lab processing (Table 1). General criteria included swabs being DNA-free and having an appropriate drying system. For packaging, handling time, contamination risk and the effort of opening and inserting the swab into its encasement were evaluated. In addition, users rated the pressure that can be applied to the swab, the stability of the swab head material during sampling, if swabs can easily be held, and the swabs' absorption behaviour. For the laboratory process, the following criteria were examined: the elasticity of the handle; if present, the position of the predetermined breaking points suiting the height of the spin baskets (QIAGEN GmbH, Hilden, Germany); the smooth and clean breaking of the swab head; the easiness in cutting the swab head. Models were dismissed from the study if they had a cardboard box for transport because of the same drawbacks as experienced with our current swab. Each investigator evaluated criteria in packaging, sampling and lab processing.
Table 1Evaluated swabs. All swabs utilise plastic tubes except for the ForensiX Evidence Collection Kit and ForensiX Collection Swab BasicDry. Reference numbers refer to catalogue numbers. It is indicated under the column "Chapter" which swabs were used in which study part, chapter numbers refer to 3.1: Tests for application of DNA, 3.2: Handling and selection of swabs, 3.3: Police and laboratory collaboration study, 3.4: Variation of sampling solution volumes for DNA collection.
For evaluating the optimal application method of touch DNA, the swab models ForensiX SafeDry, Copan Genetics and Sarstedt Forensic swab were initially selected (Table 1). Based on the results, the Sarstedt Forensic swab was excluded and replaced by the Copan Crime Scene, resulting in four swab types to be compared in this collaboration study. Swabs will be referred to as assigned in the column “Study Name” given in Table 1. An overview of the study parts and the included swab types are presented in Fig. 1.
Fig. 1Overview of the swab types used in each study part.
Twelve selected items with three different types of traces used in the collaboration study are summarised in Table 2. The collection areas were labelled on the objects' surfaces before cleaning.
Table 2Overview of selected items and their corresponding trace type for the collaboration study.
The carpets and T-shirts were cleaned with sodium hypochlorite in a washing machine at 60 °C for 1 h. All other items were cleaned by swiping with ethanol (70%) and water, afterwards air-dried and UV-light radiated (UV Airclean Workstation UVC/T-M-AR, 25 W) for 20 min per side. The cleaning success was verified exemplarily for each object with control swabs, yielding no DNA profiles.
2.3 Application of traces
Three touch DNA application approaches were tested for five items (stone, wood, knife-handle, knife-blade and glass slide) with the swabs CopGen, ForSaf and SarFor to apply traces homogeneously around the whole item. Details are summarised in Table 3. Based on our interpretation guidelines, methods were rated as appropriate if DNA profiles had no more than three allelic drop-outs in the mean for at least one type of swab. Based on the results (see Section 3.1), touch DNA was applied by the coordinating laboratory according to method 3 (see Table 3) on all the items described in Table 2 for the collaboration study. Each item was divided into four swabbing areas; every area was randomly assigned to another swab. Each item type was assigned to only one DNA donor. The donor treated only one item a day to ensure that the biological material and the donor's hand had enough time to regenerate DNA. In total, five different laboratory staff members volunteered with consent as DNA donors.
Table 3Description of the methods which were tested to apply touch DNA onto substrates homogeneously. Method 3 was used to apply touch DNA for the collaboration study.
Description
Method 1
Holding and rubbing each item for 20 min. Hands were washed before contact. Predefined swabbing area of 4 cm2 for each swab.
Method 2
Holding and rubbing each item for 5 min every 30 min for a total of 8 cycles. Hand washing directly before contact was not permitted. Predefined swabbing area of 4 cm2 for each swab.
Method 3
Holding and rubbing each item for 5 min every 30 min for a total of 8 cycles. Hand washing directly before contact was not permitted. Predefined swabbing area of 12 cm2 for each swab.
Blood and saliva samples were also tested for CopGen, ForSaf and SarFor swabs to confirm that the application of fluids to their respective substrate (Table 2) provided enough DNA for typing complete profiles, see Section 2.6. For these tests and the collaboration study, saliva was applied to openings of glass bottles by licking around five times. Blood and semen (semen only for collaboration study) were applied by pipetting 30 µl of a 1:10 dilution with water (amplification grade ddH2O, Promega Corporation, Madison, WI, USA) onto their respective substrate directly after vortexing.
2.4 Collaboration study design
The collaborative study involved crime scene investigation units of three police departments and the conducting laboratory, each responsible for the DNA recovery from the artificially treated substrates. The coordinating laboratory prepared and analysed all four units' samples and performed the statistical analysis.
After applying DNA, the objects were put into plastic bags for transport (according to the routine procedure), delivered to each police department and laboratory unit and swabbed after 1–3 weeks of sample preparation. Every organisation selected three experienced operators for sampling DNA, resulting in twelve operators. They received instruction in order to perform the evidence collection similarly. Each investigator collected DNA from every item listed in Table 2, with all four swabs yielding a total of 576 samples. Data from the participating operators and police/laboratory units were anonymised, assigning letters A to L to the persons and numbers 1–4 to the units. Single swabbing was performed by dipping the swab heads quickly in amplification grade water, without further restrictions to allow individual DNA collecting techniques. DNA extraction was performed 1–2 weeks after sampling.
2.5 Study design for variation of swabbing solution volumes used for sampling
The ForEvi swab was compared to the CopGen and ForSaf swab on three different item types (stone, glass slide and cotton dish towel) to investigate the influence of varying volumes of amplification grade water for sampling. Touch DNA was applied to previously cleaned stones and glass slides according to method 3 (see Section 2.3). Blood (20 µl, 1:10 dilution with PBS) was applied on cotton dish towels and dried for 6 h. Three 12 cm2 areas were assigned to each item to swab every item with each swab type.
Sampling was performed in three different setups, varying the amount of swabbing solution volumes: (1) dry swabbing, (2) equal amounts of water (10 µl) for all swab types and (3) varying amounts of water adjusted to the absorption behaviour of each swab type, in order to achieve moist but not too wet swab heads and leaving no water residues on the sampled surface. For the evaluation of absorption, 10 µl, 20 µl and 30 µl were directly applied to the swab heads, respectively.
One experienced investigator collected DNA from all items. Stones and dish towels were sampled for 15 s per swab; glass slides were sampled until dry to absorb the entire sample. Items were prepared in either 5 or 10 replicates for every setup.
Additionally, after quickly dipping each swab tip into a 1.5 ml tube with amplification grade water, the swabs' liquid absorption capacity was measured by weighing (n = 10 per swab type, CopGen, CopCri, ForSaf, SarFor). One experienced operator performed this test.
2.6 Genetic analysis
Samples were extracted using the Casework Extraction Kit (lysis), DNA IQ™ Casework Kit (purification) and Maxwell® RSC extraction platform (Promega Corporation, Madison, WI, USA) to an end volume of 50 µl. Quantification was performed with the Plexor® HY Kit (Promega Corporation) on a 7500 RT-PCR system (Applied Biosystems™, Thermo Fisher Scientific, Carlsbad, CA, USA). Amplification was performed with a target of 0.5 ng DNA (if possible) using the PowerPlex® ESI 17 (Promega Corporation) kit or the NGM Detect™ kit (Applied Biosystems™) for the DNA application tests, and only the PowerPlex® ESI 17 kit for the collaboration study. Capillary electrophoresis was performed on a Genetic Analyser 3500xL (Applied Biosystems™) under the following conditions: 1.2 kV injection voltage, 24 s injection time, and 1210 s run time for the PowerPlex® ESI 17. The analytical threshold was set to 50 relative fluorescence units (RFU). GeneMapper™ ID-X 1.4 and 1.5 software (Applied Biosystems™) was used for data analysis. Complete DNA profiles presented all kit-specific alleles of 16 loci (D3S1358, D19S433, D2S1338, D22S1045, D16S539, D18S51, D1S1656, D10S1248, D2S441, TH01, vWA, D21S11, D12S391, D8S1179, FGA and SE33). DNA profiles with less than three drop-outs were considered "interpretable". Drop-ins or artefacts were documented. Results were verified by comparing with the DNA donors' profiles.
2.7 Statistical analysis
For normally distributed data, all quantified DNA concentrations were transformed to a logarithmic scale. Statistical analysis was performed with R [
] to perform a linear mixed analysis of the relationship between DNA concentration and swab type, with swabs and items/trace type entered as fixed effects (with interaction term) and operators as random intercept. A Kruskal-Wallis rank-sum test was calculated to compare the rate of allelic drop-outs per swab and the absorption capacity after quickly dipping swabs into water. The mean amount of DNA concentrations collected by the individual operators and the four units was compared using pairwise t-tests on centred data around the model-coefficients of the different items by subtracting the coefficient of each item from its respective data points. Centring was applied to reduce the variance introduced by different item/trace types on the mean DNA concentration per operator to approximately compare the operator performances across the entire experiment. The variation of sampling solution volumes was compared using Welch t-Tests. All plots were created using ggplot2 [
]. A significance level of α = 0.05 was set for all tests.
3. Results
3.1 Tests for application of DNA
DNA application method 1 (see Table 3) resulted in no complete profile and a mean drop-out number of 24.5 per DNA profile across all swabs. Method 2 was more effective and yielded a mean drop-out number of 8.3 per DNA profile. Items treated with method 3 gave a mean drop-out rate of 5.9 per DNA profile across all swabs. However, when excluding the SarFor swab due to its poor overall performance (see below), the mean drop-out rate decreases to 1.0 per DNA profile.
Detailed results of method 3 are shown in Fig. 2. For the SarFor swab, DNA concentrations were lower considering all touch DNA items given in Fig. 2 A (mean = 0.0026 ng/µl) in comparison to the CopGen (mean = 0.036 ng/µl) and the ForSaf swab (mean = 0.046 ng/µl). Except for one sample (diluted blood on carpet), this phenomenon was observed for items where biological fluids were applied (Fig. 2 B). In particular, when diluted blood was directly applied to the swab heads, the SarFor yielded a DNA concentration of less than half compared to the other swabs (1.57 ng/µl versus 5.29 ng/µl and 6.51 ng/µl). Mean DNA concentrations across touch and fluid samples are given in Fig. 2 C and show that the SarFor swab yielded the lowest mean DNA concentration.
Fig. 2Swab performance comparison for the application of touch DNA (method 3) and biological fluids. A) Quantification results of touch DNA samples per item. B) Quantification results of items with biological fluids. C) Mean DNA concentrations for all substrates and traces (touch and biological fluid) per swab. Error bars represent standard errors. D) Allelic drop-outs per touch DNA profile for each swab.
While complete profiles were generated with all fluid samples, not a single complete STR profile was typed for touch DNA samples swabbed with the SarFor swab (mean = 15.6 allelic drop-outs per sample). In comparison, the CopGen and ForSaf swab mostly showed complete profiles for touch DNA samples (mean drop-outs per sample for CopGen = 1.8, for ForSaf = 0.2). The number of allelic drop-outs per touch DNA profile is displayed in Fig. 2 D.
3.2 Handling and selection of swabs
Swabs were evaluated given the criteria described in chapter 2.1., resulting in different scores (see Table 4). The ForEvi swab reached only 7 out of 15 scores, primarily due to packaging associated handling difficulties. The swabs ForSaf, ForBas, CopGen, CopCri, and SarFor were considered best in handling criteria, all reaching a score of 11 or 10. However, the ForBas swab was excluded because of its cardboard storage tube and the material's assumed proneness to environmental influences. The PurCap swab also reached only a score of 7. Plus, its protective cap attached to the swab head raised doubts about the associated contamination risk during sampling, so the type was dismissed. Despite the SarFor swab’s good score, the model was excluded due to the strikingly poor performance during DNA application tests. Thus, the remaining three challengers, CopGen, CopCri and ForSaf were further compared to the currently used ForEvi swab in the collaboration study.
Table 4Comparison of criteria for swab handling with the following ratings: Positive (+), neutral (o), negative (-). CopGen and CopCri swabs’ handling characteristics are identical. Bold properties were considered as most important.
Category
Property
ForEvi
ForSaf
ForBas
CopGen/ CopCri
SarFor
PurCap
Packaging
Time/handling
–
+
+
+
+
–
Contamination risk
–
+
+
+
+
–
Transport / Protection against environment
–
+
–
+
+
–
Easy to open
–
+
+
–
+
–
Easy to insert swab and close tube/box
–
+
+
+
+
–
Easy to seal
+
+
+
+
+
+
Sampling
Pressure that can be applied with the swab
+
+
+
+
+
+
Swab head stability during sampling
–
–
–
+
+
+
Rigidity of swab handle
+
+
+
+
–
+
Swab handle can easily be held
+
–
–
–
–
+
Absorption behaviour
+
+
+
–
–
+
Lab processing
Breaking point position
o
o
o
+
o
o
Swab head can easily be broken off
+
+
+
+
–
–
Clean breaking point
–
–
–
+
+
–
Swab head can easily be cut for, e.g., trace type tests
A linear mixed-effect analysis of the log-transformed DNA quantities was performed to compare the three challenger swabs with our current swab on different items. The item type significantly influenced DNA recovery across all swabs (ANOVA p < 0.0001, data not shown). The discrepancy is not surprising since different touch DNA donors with varying shedder status (data not shown) handled the different item types. In addition, biological fluids like blood or semen were applied to some item types, resulting in higher DNA concentrations than touch DNA samples. However, no significant interaction was observed between swab and item/trace types (ANOVA p = 0.86). Thus, the relationship between swab-type and DNA concentration obtained is predominantly independent of the items swabbed in this study section, indicating that all swabs performed proportionally similar per item/trace type.
According to the linear mixed model, the choice of swab type significantly affects the amount of yielded DNA (ANOVA p < 0.0001). Both Copan swabs yielded significantly lower amounts of DNA when compared to our current swab ForEvi and the ForSaf swab (p < 0.001, respectively). Other post hoc comparisons of swabs were not significant. The distribution of DNA concentration per swab and post hoc comparisons of swabs are displayed in Fig. 3 A and B, respectively.
Fig. 3A) DNA concentrations obtained per swab. The line in the boxes represents the median. B) Post hoc comparisons of the linear mixed effect model for the estimated values of swabs. Error bars represent 95% confidence intervals. NS = not significant.
All four swab models resulted in complete DNA profiles when swabbing biological fluids from samples. For touch DNA samples, drop-outs occurred; however, no statistical significance in-between swab types were observed.
3.3.2 Comparison of operators
The comparison of DNA yield per operator and police/laboratory units is shown in Fig. 4. While the pairwise t-test showed no significant difference when comparing the centred mean DNA concentration of most operators, some operators are very efficient in DNA uptake, while others are underperforming (see Table 5 for details).
Fig. 4Comparison of DNA concentrations achieved by operators (A) and police/laboratory units (B). Line in the boxes represents the median.
A pairwise t-test of the centred mean DNA concentrations revealed that unit 1 was significantly underperforming in contrast to units 2, 3 and 4 (p < 0.013). No significant difference was observed when comparing the other units.
3.4 Variation of sampling solution volumes for DNA collection
Swabs’ optimal liquid absorption behaviour resulted in 10 µl water for CopGen, 20 µl for ForSaf, and 30 µl for ForEvi. In more than one-third of the tested scenarios, the CopGen swab was significantly underperforming regarding the recollected DNA yield compared to one or both challenger swabs (Fig. 5; all cotton (blood) setups and an equal amount for stone (touch DNA)). The ForSaf swab achieved a significantly higher mean DNA concentration than the ForEvi swab for both wet swabbing experiments on cotton. All other tested combinations were insignificant.
Fig. 5DNA yield of blood swabbed from cotton or touch DNA swabbed from glass or stones. Equal water amounts represent 10 µl per each swab; adjusted water amounts represent 10 µl for the CopGen, 20 µl for ForSaf and 30 µl for ForEvi swab. Asterisks indicate significant differences in DNA quantity with * p < 0.05, *** p < 0.001, **** p < 0.0001. The line in the boxes represents the median.
DNA concentrations in dependence of different water volumes are displayed in Fig. 6, showing the highest or second-highest DNA concentration for touch DNA on glass and stone when sampling with 10 µl of water for all swabs. For touch DNA on stones, a further increase in DNA output could be observed when more water was applied to the swabs ForSaf and the ForEvi. For dried blood on cotton, again, 10 µl of water resulted in the highest DNA yield for the ForSaf swab, whereas the CopGen and ForEvi swabs performed best when dry.
Fig. 6Change of DNA concentration per swab under varying amounts of water used for trace collection. Water capacity limits were 10 µl for CopGen, 20 µl for ForSaf, and 30 µl water for ForEvi.
Additionally, after quickly dipping them into the water, swabs' absorption behaviour was tested, resulting in similar amounts of water for CopGen, CopCri and ForSaf swabs (data not shown). In contrast, the SarFor swab showed a significantly lower absorption behaviour than all other swabs (p < 0.05).
4. Discussion
This study aimed to identify a suitable alternative to the currently used ForEvi swab to simplify handling, save time, and reduce the contamination risk caused by folding the transport box at the crime scene. A variety of alternative commercial products is available, but swab performances can vary substantially. Surprisingly, prominent differences were observed between the SarFor and the other swabs used in this study, highlighting the importance of evaluating swab performance before introducing a new type into the laboratory.
4.1 Swab handling
Even though the securing activity is decisive for the overall success of forensic DNA analysis, surprisingly few considerations are given to the handling of the most widespread forensic equipment, the swab. In this study, we wanted to eliminate an intrinsically insecure condition associated with the currently used swab. Each time-intensive folding process of the cardboard box may provoke DNA contamination by the crime scene investigator. For example, "threading" the swab with the valuable DNA evidence into the designated openings risks touching the inside of the box or other compartments, in the worst case, the operator's gloved hand, which could potentially lead to secondary DNA transfer. From the police perspective, the new swab model was supposed to provide a safer and faster workflow that creates an environment with a lower likelihood of human handling error.
Furthermore, the tube packaging was considered superior to cardboard box packaging by all DNA collectors. At the same time, a present handle lid (all tube models: ForSaf, ForBas, CopGen, GopCri, SarFor) restricted an operator’s unique collection technique or the high elasticity hindering a practical application of pressure for sampling (SarFor). Therefore, the investigators’ feedback is quite important, especially considering that collection activities often occur at crime scenes with suboptimal environmental conditions or in public places, with high time pressure to preserve evidence, which can lead to an increased potential for failures [
]. Together with the "laboratory perspective", for which the swab needs to be compatible with its specific environment (such as the handle's breaking point), there is considerable room for improvement of the most commonly used forensic collection tool.
4.2 Preparation of homogenous traces
The increasing sensitivity of STR PCR Kits over the last decades demands a deeper understanding of touch DNA to collect traces. However, research in this field is challenging due to difficulties in preparing and comparing touch DNA traces [
]. Therefore, it was crucial to prepare traces as homogenous as possible to compare swab efficiency on the same item. The initial step was to create the basis for homogeneous traces by cleaning the objects to eliminate background DNA. When testing different methods of touch DNA application, it became evident that DNA yield was improved by increasing touch frequency, sampling area and not washing hands immediately before touching objects. Hand washing likely resulted in removing loose cells, fragments, and cell-free DNA from the hand surface [
During tests for DNA application, the CopGen and the ForSaf swabs achieved higher DNA quantities and fewer allelic drop-outs than the SarFor swab. In contrast to the active drying system of CopGen and the ForSaf tubings, the SarFor tube utilises a passive drying system in the form of a ventilation membrane. Garvin et al. observed that the active drying system of the ForSaf was superior to the ventilation membrane of the SarFor swab and led to a higher amount of quantified DNA [
]. However, further testing with the ForEvi swab, stored in a cardboard box equivalent to a passive drying system, yielded comparable results to the CopGen and ForSaf swabs. All swabs collected for DNA application tests were processed briefly after trace collection to minimise possible storage effects [
]. Thus, the drying system is most likely not a causal or influencing factor for the varying DNA quantity and quality results.
Besides the drying system, the swab heads are made of different materials. While the CopGen swab consists of nylon and the ForSaf and ForEvi swab head of cotton, the SarFor swab head is made of viscose. In our study, the SarFor swab showed unusual and unfavourable delayed absorption behaviour; in some cases, even blood droplets pearled off the surface instead of being absorbed. When comparing the amounts of liquid absorbed after rapid immersion of the swabs in water, the SarFor swab soaked up significantly lower amounts of water than the CopGen, CopCri and ForSaf swabs (data not shown). The inefficient absorption is not necessarily related to the swab material but may also be caused by the morphology, porosity, and swab head density [
], which could also explain the more unsatisfactory performance of the SarFor swab in these test runs, even though its head is made of viscose. Bruijns et al. observed that rayon (the same as or similar to viscose) swabs reached a lower DNA recovery efficiency than nylon and cotton swabs. In contrast, in their study, the highest recovery efficiency was found for Copan 4N6FLOQSwab™ regularly flocked swab (4520CS01) [
]. In a comparison study of Copan Genetics 4N6FLOQswabs™ and rayon swabs (RAYON), samples were extracted with the DNA IQ™ Casework Sample kit and Maxwell® instrument [
] which supports our findings for touch DNA with low input amounts of biological traces. Since touch DNA traces collected with the SarFor swab resulted in many allelic drop-outs, and touch DNA was the main focus of this study, this particular swab was excluded for further study.
During the collaboration study, the CopGen and CopCri swabs yielded significantly lower amounts of DNA than the currently used ForEvi and the newly tested ForSaf swab. No significant difference was observed in the latter two swabs. This result contrasts with other studies showing that Copan's flocked swabs release more DNA than cotton swabs [
]. For our study, an automated DNA IQ™ extraction procedure performed on a Maxwell® RSC was used, operating with a similar magnetic particle technique like the EZ1 system by QIAGEN. Several other studies also describe higher DNA yields for cotton than for nylon flocked swabs [
Another aspect is the water amount used for sampling, which may influence the overall profiling success. Different sampling solution volumes are used between studies [
] or within a study for different swabs. Comte et al., for example, investigated varying amounts of water for the different swabs from no water for collars, one drop of water for the CopGen swab, and three drops for the ForEvi swab for all items in their study [
]. In our collaboration study, operators were instructed to moisten swabs by dipping them quickly into amplification grade water tubes. Here, the swabs can absorb according to their properties, which could vary between different swab types. This moistening practice, however, could result in liquid oversaturation. In fact, the CopGen and CopCri swabs left more residual water on swabbed surfaces than the ForEvi and ForSaf swabs, thus it was first assumed that the Copan swabs absorb more water than the ForensiX swabs during dipping. Another observation was that the amount of absorbed water could be better controlled for ForEvi and ForSaf swabs than for CopGen and CopCri, giving the impression that the Copan swabs fully soak up during dipping, which could result in more water. However, the absorption differences could not be verified during tests where swabs were weighed after dipping into water (data not shown). CopGen, CopCri and ForSaf absorbed comparable amounts of water, suggesting that Copan swabs rather cannot quickly reabsorb water transferred to the swabbed surface, which matches the lower DNA concentrations obtained throughout the collaboration study. However, only one operator performed the weighing tests of water amounts. Hence, it cannot be excluded that other operators absorbed different amounts of water for the various swabs during the collaboration study. According to Castriciano et al., wetting Copan 4N6FLOQSwabs™ by dipping results in sample loss [
]. Therefore, the effect of water volume variation for the different swabs was further analysed and discussed in Section 4.5. Other potential effects, such as wetting the swab with a dispenser or manually shaking off excess water, were not evaluated in the collaboration study.
In addition, the ForEvi swab has been integrated into our police department's routine casework and laboratory for decades. Therefore, a potential bias due to the swab handling habituation could have positively influenced the DNA yield. Although the comparison of other collection methods besides swabbing was out of the scope of this study, it should also be noted that other sampling approaches can be even more suitable for some surfaces. For example, for textiles, good results are normally achieved by tape-lifting or scraping when looking for touch DNA, whereas cutting is often used for blood, semen or saliva traces [
4.4 Sampling differences between operators and forensic units
This study enabled a comparison of single operators' sampling skills and made it possible to recognise performances differences and improve DNA collection techniques. A Comparison of all twelve operators mostly showed comparable performances. However, two operators of one organisation secured significantly lower DNA concentrations, resulting in this unit significantly underperforming compared to the other three. However, one of the better performing participants was also a member of the same unit. This result showed that the individual operator's sampling technique could heavily influence the sampling efficiency, which was previously observed by Hedman et al. [
]. Thereby, the evaluation of sampling performances of different operators offers the chance to improve their sampling success. However, it should be noted that supervisors could also misuse knowledge of individual sampling efficiency. Thus, reasonable anonymisation of data is essential.
4.5 Variation of water amount during sampling
To our knowledge, this study investigated for the first time how the amount of water used for sampling affects DNA yields for different swab types. For blood on a cotton surface, our findings suggest that no water should be used for all swabs except for the ForSaf, where 10 µl of water even increased the DNA concentration. In contrast, adding water reduced the DNA amount for the CopGen significantly, while there was a minor effect on cotton swabs. This finding indicates that blood adheres better to dry than wet nylon and that the wetting effect is more moderate for cotton.
When different sampling solution volumes were applied to a given swab, no statistical differences were observed for glass surfaces with touch DNA. However, the highest DNA yield was obtained with 10 µl of water for all swab types. Comparable performance of cotton and nylon swabs on glass slides was also reported by Wood et al. [
]. These findings suggest that the uptake of touch DNA on glass is already at its maximum and cannot be significantly increased by varying the amount of water or the type of swab.
Swabbing stones with 10 µl of water resulted in a significantly higher mean DNA concentration for the ForEvi than for the CopGen swab. However, the highest DNA yields were achieved for both ForensiX swabs when even more water (20/30 µl) was added. Thus, the optimal volume of the swab solution is larger for a porous surface like stone than for smooth, non-absorbent surfaces like glass slides. This may be caused by small cavities present on the used stones’ surfaces which trapped the touch DNA unless it was washed out by an adjacent amount of water. This is supported by the results of Hedman et al., who found a positive effect of adding more swabbing solution to cotton swabs when sampling from a porous wooden cutting board surface [
This study aimed to identify the best-suited swab for our laboratory environment in collaboration with three crime scene investigation teams by comparing various challenger swabs to our currently used evidence collection kit. DNA evidence sources and sampling surfaces were chosen to be as realistic as possible. The challenge of preparing homogenous traces, especially touch DNA, was addressed by enhancing hand contact frequency and sampling area.
The collaboration study showed that the CopGen and CopCri swabs yielded significantly lower DNA concentrations than the currently used ForEvi swab. The latter and the ForSaf achieved similar mean DNA concentrations. Based on DNA recovery and handling evaluation, the ForSaf achieved the best overall result, motivating a transition to this swab type.
The additional analysis of varying water amounts for sampling shows higher DNA yields or no significant differences for the ForEvi and ForSaf than the CopGen in dependence of surface and trace type. Nevertheless, further research is recommended regarding the optimal amount of water on different surfaces and trace types to evaluate the best sampling method that may be substrate-specific.
A second objective dealt with comparing different operators in their collection efficiency, achieved by preparing homogenous traces and large sample sizes. Again, significant differences were found, which helped to evaluate sampling ability and allow improvements. We have shown that operators are an influential parameter that should be carefully considered when planning a DNA collection study. In fact, any swabbing study should include more than a single DNA collector in order to obtain valid results. Moreover, the observed inter-individual differences translate directly into the DNA sampling efficiency at a given crime scene. This opens the gate for best-in-class DNA sampling, training and tutoring to increase the overall securing performance of the entire team.
This study highlighted the importance of validating the best-suited swab type within the own laboratory environment as a crucial component for downstream DNA profiling success. The outcome of this collaboration research delivered our best-suited swab within our laboratory processes which has replaced our currently used evidence collection kit. Furthermore, best-performing individuals will now conduct training on DNA collection to ensure optimal practices.
Besides short-term DNA recovery efficiency, it is also essential to investigate how different swabs perform after more prolonged periods, especially since only a few studies are available, and none of them covers over one year [
]. Therefore, we have initiated a long-term storage study with biological fluids to assess the effects of different packaging- and swab compositions on DNA preservation.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Acknowledgements
We thank Peter Elsener, Isabelle Häusler, Alexander Heuri, Agata Knap, Till Meyer, Urs Nachbur, Lisa Richter, Mischa Sigrist, Rafael Studer, Daniel Vallat and Nuaim Wenger for their effort and fruitful discussions.
Conflict of interests
The authors declare they have no conflict of interest.
00057-6&id=gr1.jpg){kind=link}
00057-6&id=gr2.jpg){kind=link}
00057-6&id=gr3.jpg){kind=link}
00057-6&id=gr4.jpg){kind=link}
00057-6&id=gr5.jpg){kind=link}
00057-6&id=gr6.jpg){kind=link}