DNA typing in populations of mule deer for forensic use in the Province of Alberta

Available online at www.sciencedirect.com

Forensic Science International: Genetics 2 (2008) 190–197 www.elsevier.com/locate/fsig

DNA typing in populations of mule deer for forensic use in the Province of Alberta Richard M. Jobin a,*, Denise Patterson 1, Youfang Zhang b a

Alberta Sustainable Resource Development, Fish and Wildlife Forensic Laboratory, 7th Floor OS Longman Building, 6909-116 Street, Edmonton, Alberta T6H 4P2, Canada b Department of Criminal Investigation, Zhejiang Police College, 555 Binwen Road, Hangzhou 310053, PR China Received 31 December 2007; accepted 25 January 2008

Abstract The present study involves the development of forensic DNA typing tests and databases for mule deer in the Province of Alberta. Two multiplex PCR reactions interrogating 10 loci were used to analyze samples from three populations of mule deer. Additionally, an amelogenin based sextyping marker was used to determine the gender of samples. Results show that the tests and databases are appropriate for use in forensic applications. Additionally, the results indicate that there is little population structure in mule deer in Alberta and that no changes to management of this game species are suggested. # 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Forensic science; Wildlife; Alberta; Mule deer; DNA typing; Microsatellite; Odocoileus hemionus; Poaching; Polymerase chain reaction; Sika amelogenin; BM4107; T7; Rt30; Rt7; Rt5; BM1225; OheN; BM4208; OheQ

The mule deer (Odocoileus hemionus) is a highly prized North American big game animal that is hunted both as a trophy and also for its meat. Its range covers the Western portion of North America [1]. In Alberta, illegal taking of mule deer is an ethical and public safety concern. However, no estimates of how many deer are taken illegally have been made. In other areas in North America, radiocollar studies performed on deer indicate that the number of illegally taken deer is between 9 and 61% of the legal harvest [2–4]. These studies indicate that illegally taken deer represent a significant source of deer mortality, especially in areas that experience considerable hunting pressure. Protection of big game through enforcement activities presents some significant challenges. These difficulties stem from the fact that wildlife officers are typically responsible for patrolling large areas of undeveloped land. When deer are illegally taken, it usually takes place in remote areas where it is unlikely that there will be any witnesses who are not party to the offence. * Corresponding author. Tel.: +1 780 422 3194; fax: +1 780 422 9685. E-mail address: [email protected] (R.M. Jobin). 1 Present address: Molecular Diagnostics Laboratory, Department of Medical Genetics, 8-26 Medical Sciences Building, University of Alberta, Edmonton, Alberta T6G 2H7, Canada. 1872-4973/$ – see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.fsigen.2008.01.003

These circumstances make it quite difficult to detect these offences and make it even more difficult to convict individuals who commit them. However, the advent of forensic DNA typing has made it possible for officers to collect evidence from illegal kill sites and link it back to biological material that is associated with the individual who committed the offence. In the present study, forensic DNA-typing tests and databases have been developed to protect mule deer in the Province of Alberta and surrounding territory. The goal of this project is to provide the court with acceptable evidence to assist in the prosecution of individuals who violate statutes designed to protect wildlife. DNA-based evidence has been used in the prosecution of human crime for over 20 years [5]. Currently, the analysis of microsatellite DNA via the polymerase chain reaction is the most commonly used forensic technique. This technology has been available to human forensics for over a decade [6,7]. DNA analysis has also been used in a forensic capacity in deer [8–10]. However, a majority of the published work on microsatellite analysis in deer has been aimed at answering questions regarding population structure and parentage [11–13]. The present study is the first in which DNA markers and mule deer populations are selected specifically to be used in a forensic capacity to protect mule deer in a defined geographic area. In addition to their use in a

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

forensic capacity, data collected in this study may also be of interest to individuals who study and/or manage mule deer. 1. Methods 1.1. Location of deer populations Three mule deer populations were used in this study to serve as forensic databases. The locations of these databases were selected so that they covered a majority of the mule deer range within the province and so that the number of samples available was adequate to constitute a forensic database. These databases included the following areas: (1) an area mostly to the south of Lethbridge, Alberta (STHMLD), (2) an area surrounding Lloydminster, Alberta (LYDMLD), and (3) an area surrounding Fort St. John, British Columbia (FSJMLD). The database samples were collected within a 100-km radius of the sites indicated in Fig. 1. Although some mule deer can be found north of the FSJMLD database, there are relatively few deer in the extreme northern part of the Province (unpublished data from government deer counts). This low density of animals and limited accessibility has resulted in too few samples being available for an adequately sized database to represent the northern limit of the species in Alberta. 1.2. DNA processing Tissue samples were collected from hunter kills, found dead animals and animals that were culled as part of a disease prevention program. Samples were extracted using a Kingfisher ML purification system (Thermo Electron Corp., Waltham, MA, USA) and Kingfisher, genomic DNA purification kit (Thermo Electron Corp., Waltham, MA, USA). These kits have been discontinued but can be replaced by the MagExtractor, Nucleic acid purification kit (Toyobo Co. Ltd, Osaka, Japan). The DNA in each sample was quantified using a Biophotometer (Eppindorf, Hamburg, Germany). DNA samples were diluted, to a concentration of 5 ng/ml, in water, which was filtered,

Fig. 1. An outline map of three Provinces in Western Canada (British Columbia, Alberta and Saskatchewan). This map also shows the locations of the mule deer populations studied, Fort St. John (FSJMLD), Southern Alberta (STHMLD) and Lloydminster (LYDMLD).

191

autoclaved and deionized. Different DNA extraction techniques were not compared in this study; however, the methods used in this study have been shown to be effective in extracting DNA from biological material commonly encountered in casework [14]. Primers for 10 microsatellite markers were utilized (Table 1). These markers were amplified in two multiplexed PCR reactions. Multiplex 1 consists of 1 concentration of GeneAmp1 PCR Buffer II (Roche Molecular Systems Inc., Almeda, CA, USA), 2.1 mM MgCl2 (Roche Molecular Systems Inc., Almeda, CA, USA), 20 mM GeneAmp1 dNTP mix with dTTP (Roche Molecular Systems Inc., Almeda, CA, USA), 3 U AmpliTaq Gold1 (Roche Molecular Systems Inc., Almeda, CA, USA), 0.10 mM BM4107 primers, 0.20 mM T7 primers, 0.20 mM Ovir A primers, 0.20 mM Rt30 primers, 0.16 mM Rt7 primers, 5 ng DNA template and enough filtered–autoclaved– deionized water to make a 25-ml reaction volume. Multiplex 2 consists of GeneAmp1 PCR Buffer II (Roche Molecular Systems Inc., Almeda, CA, USA), 1.8 mM MgCl2 (Roche Molecular Systems Inc., Almeda, CA, USA), 20 mM GeneAmp1 dNTP mix with dTTP (Roche Molecular Systems Inc., Almeda, CA, USA), 3 U AmpliTaq Gold1 (Roche Molecular Systems Inc., Almeda, CA, USA), 0.08 mM Rt5 primers, 0.40 mM BM1225 primers, 0.20 mM OheN primers, 0.80 mM BM4208 primers, 0.80 mM OheQ primers, 10 ng DNA template and enough filtered–autoclaved–deionized water to make a 25-ml reaction volume. Primers for Rt7 and OheQ were produced by Applied Biosystems, Foster City, CA, USA. All other primers were produced by Integrated DNA Technologies, Coralville, IA, USA. Both multiplexes use the same amplification conditions. The amplification conditions are as follows: (1) (hot start) 95 8C for 4 min, (2) (denaturation) 94 8C for 30 s, (3) (annealing) 54 8C for 30 s, (4) (extension) 72 8C for 60 s, (5) steps 1 through 4 repeated 30 times, (6) (final extension) 60 8C for 45 min. Sika deer amelogenin primers were not included in the multiplex reactions. This locus is amplified alone when required under the following conditions: 1 concentration of GeneAmp1 PCR Buffer II (Roche Molecular Systems Inc., Almeda, CA, USA), 1.3 mM MgCl2 (Roche Molecular Systems Inc., Almeda, CA, USA), 20 mM GeneAmp1 dNTP mix with dTTP (Roche Molecular Systems Inc., Almeda, CA, USA), 1 U AmpliTaq Gold1 (Roche Molecular Systems Inc., Almeda, CA, USA), 0.20 mM sika amelogenin primers, 10 ng DNA template and enough filtered– autoclaved–deionized water to make a 25-ml reaction volume. The amplification conditions are as follows: (1) (hot start) 95 8C for 4 min, (2) (denaturation) 94 8C for 30 s, (3) (annealing) 57 8C for 30 s, (4) (extension) 72 8C for 60 s, (5) steps 1 through 4 repeated 30 times, (6) (final extension) 60 8C for 45 min. Amplified DNA was stored at 20 8C. The amplifications were performed on an MJ Research PTC-200 DNA Engine1 (MJ Research Waltham, MA, USA). The amplified DNA fragments were diluted (60 times dilution) in Hi-DiTM formamide (Applied Biosystems, Foster City, CA, USA) and a 400 times dilution of GenScan1 400HD [ROX] size standard was added to each sample (Applied Biosystems, Foster City, CA, USA). This mixture was

192

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

Table 1 Primers sequences for loci that were studied LOCUS

Primer sequence (50 ! 30 )a

Reference

Hob

Total allelesc

BM4107

AGC CCC TGC TAT TGT GTG AG ATA GGC TTT GCA TTG TTC AGG TTT CAC TGT TTT CTC CTT CAG TGC CCA ATC AGA TGT TGT AG CAC AAA GAA TCA GAC GTG GT G TGC ATC TCA ACA TGA GTT AGG GTG TAA CCC AAA GGG CAA CT CTG GTG TAT GTA TGC ACA CT CCT GTT CTA CTC TTC TTC TC ACT TTT CAC GGG CAC TGG TT CAG CAT AAT TCT GAC AAG TG GTT GAG GGG ACT CGA CTG ACC CCT ATC ACC ATG CTC TG TTT CTC AAC AGA GGT GTC CAC GCA ACC AAT AGG ATA GGT CG GCT GGA TGG AAC TGA AAG TC TCA GTA CAC TGG CCA CCA TG CAC TGC ATG CTT TTC CAA AC TAT GGA CTT TGG GCG ATT AC ATC CAG GCA ACC ATC TAG GCC CAG CAG CCC TTC CAG TGG CCA AGC TTC CAG AGG CA

[15]

0.60

13

[9]d

0.67

7

[16]d

0.83

12

[17]

0.75

15

[17]

0.80

12

[17]

0.79

11

[15]

0.39

2

[9]

0.85

13

[15]

0.32

5

[9]d

0.36

12

[18]

NA

3

T7 OvirA Rt30 Rt7 Rt5 BM1225 OheN BM4208 OheQ Sika amelogenin a b c d

The top sequence is the labeled primer. Ho, observed heterozygosity. Total number of alleles in all populations. T7, OvirA and OheQ primers have been modified, to verify see gene bank accession numbers AF102240, L35576 and AF102241, respectively.

denatured at 95 8C for 9 min and then snap chilled on ice. The amplified fragments of DNA were sized using an ABI Prism1 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). The data was collected using ABI Prism1 310 Data

Collection (Version 2.0, 2000) and GeneScan1 software (Version 3.1.2, 2000) (Applied Biosystems, Foster City, CA, USA). An example of electropherograms from multiplex 1 and multiplex 2 are shown in Figs. 2 and 3, respectively.

Fig. 2. Electropherogram of multiplex 1 for mule deer. The X-axis measures size in base pairs. The Y-axis measures intensity of fluorescence in relative fluorescent units. The upper panel shows fragments containing the blue (FAM) fluorescent tag. Fragments from the BM4107 and T7 loci are represented. The second panel shows fragments containing the green (HEX) fluorescent tag. Fragments from the OvirA and Rt30 loci are represented. The third panel shows fragments containing the yellow (NED) fluorescent tag. Fragments from the Rt7 locus are represented. The bottom panel shows fragments from the GenScan1 400HD [ROX] size standard (Applied Biosystems, Foster City, CA, USA).

193

68 78 80 82 86 88 90 92 94 96 97 101

8.3 12.2 0.3 0.3 0.8 1.1 33.6 9.1 19.0 6.8 7.6 1.1

154 158 162 164 166 168 173 175 177 181 183 185 188 190 192

1.7 0.1 2.8 4.0 2.7 12.3 9.2 7.8 1.2 0.8 38.4 10.2 3.7 3.6 1.5

208 210 214 217 225 227 229 231 233 235 239 241

9.6 4.1 3.9 18.2 4.8 31.0 21.1 3.5 2.4 0.3 0.4 0.7

90 93 98 100 101 102 104 106 108 111 113

0.1 12.8 3.7 40.4 0.1 25.3 1.6 1.7 9.1 4.5 0.3

215 229

33.6 66.4

252 264 268 272 276 280 284 288 292 296 300 304 308

17.8 0.5 3.6 17.0 22.2 6.3 1.9 7.1 6.6 14.4 1.3 0.9 0.4

139 142 144 146 148

0.3 15.9 49.2 4.7 25.9

255 259 263 267 271 275 279 283 287 291 299 303

0.3 23.7 6.8 20.2 24.9 7.9 0.8 0.9 0.4 7.8 1.5 0.1

1.3. Data analysis Collected data were categorized into different allele sizes using Genotyper1 (Version 2.5.2, 2000) software (Applied Biosystems, Foster City, CA, USA). The Excel microsatellite tool kit [19] was used to calculate allele frequencies, heterzygosity, check for duplicate samples and format data for other programs. The data were tested for Hardy–Weinberg equilibrium and linkage disequilibrium using Genepop [20]. A Bonferroni correction for multiple comparisons was used on the data [21]. Genetic differences between populations were measured using F ST as calculated by Genepop [20]. The coancestry coefficient theta (u) was calculated with FSTAT Version 2.9.3 [22]. The program ‘Structure’ was used to independently assess the populations of mule deer selected as forensic databases [23]. Probability of identity was calculated using Gimlet Version 1.3.3 [24]. 1.4. Forensic calculations Human DNA evidence has gained general acceptance in courts in North America. For this reason, whenever it was reasonable and possible, the same methodology and data analysis used in human forensic DNA typing was adopted by our laboratory to calculate random match probabilities for mule deer. Our laboratory collected the recommended 100–150 individual samples per database [25,26]. The number of individuals contained in each database ranged from between 117 and 129 (Table 3). In human forensics, minimum allele frequencies are calculated as 5/2N (N = the number of samples in the population) [27]. In the present study, mule deer data show a maximum of 13 common alleles (those with frequencies above the minimum allele frequency, data not shown). According to Chakraborty [25], all alleles above the minimum allele frequencies in every population tested are represented with at least 95% confidence. To make this estimate even more conservative, for our calculations the minimum allele frequency calculation was increased to 8/2N so that common alleles could be represented with at least 99% confidence. Table 3 Variability measures for each population Population

Na

Heb

Hoc

Ad

STHMLD LYDMLD FSJ

129 128 117

0.72 0.69 0.71

0.67 0.65 0.64

9.40 7.90 8.00

0.7 28.1 0.1 49.7 15.9 4.1 1.3

a b c

Number of individuals (N). Expected heterozygosity (He). Expected heterozygosity (He). Number of alleles detected (A).

207 214 218 222 227 231 235

d

0.4 0.5 52.0 16.7 0.3 2.5 1.7 0.3 21.9 0.7 0.4 1.3 1.2

Table 4 FST values for each population pair

136 142 147 149 151 153 155 157 159 161 163 165 167

OheQ BM4208 OheN BM1225 Rt5 Rt7 Rt30 OvirA T7 BM4107

Table 2 Allelic frequencies of pooled data

Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency Alleles Frequency

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

Population

STHMLD

LYDMLD

LYDMLD FSJMLD

0.0222 0.0322

0.0125

194

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

Fig. 3. Electropherogram of multiplex 2 for mule deer. The X-axis measures size in base pairs. The Y-axis measures intensity of fluorescence in relative fluorescent units. The upper panel shows fragments containing the blue (FAM) fluorescent tag. Fragments from the Rt5, BM1225 and OheN loci are represented. The second panel shows fragments containing the green (HEX) fluorescent tag. Fragments from the BM4208 locus are represented. The third panel shows fragments containing the yellow (NED) fluorescent tag. Fragments from the OheQ locus are represented. The bottom panel shows fragments from the GenScan1 400HD [ROX] size standard (Applied Biosystems, Foster City, CA, USA).

Random match probabilities were calculated using a variation of the product rule. According to the product rule, the frequency of occurrence of a heterozygote is equal to twice the product of the frequency of the two alleles that are observed ( f = 2pq) while the frequency of occurrence of a homozygote is equal to the square of the frequency of the single allele observed ( f = p2). For our calculations, heterozygotes were treated as indicated by the product rule, however, the frequency of occurrence for homozygotes was calculated as the straight frequency of the observed allele ( f = p). This very conservative calculation for homozygotes was adopted for mule deer and all other species in our laboratory as a measure to offset the effects of possible null alleles [28]. The frequency of occurrence of the alleles at each locus was calculated using the above methodology. Results from each locus were then multiplied together. The inverse of this product was then presented as the random match probability. The coancestry coefficient was calculated using FSTAT Version 2.9.3 [22]. This value is used to modify the calculation of random match probabilities by making allowances for population substructure [27]. However, given the conservative measures already employed in our modified calculations of random match probabilities, use of the coancestry coefficient is not necessary for individuals that are within or between the presented database populations. 2. Results After use of a Bonferroni correction, all loci in all populations were in linkage equilibrium and all loci were in Hardy–Weinberg equilibrium except BM4208 and OheQ (data not shown). The sex-typing marker, sika amelogenin [18], correctly identified all 60 confirmed male and female

individuals used for gender identification. All males displayed a 170.50 base pair (bp) fragment along with either a 214.56-bp fragment or a 223.22-bp fragment (Fig. 4). The 170.50 bp fragment was observed with a frequency of 50%, while the 214.56 and 223.22 bp fragments were observed at frequencies of 6.67 and 43.33%, respectively. Females displayed either a 223.22-bp fragment by itself or in combination with a 214.56bp fragment (Fig. 4). The 214.56 and 223.22 bp fragments were observed with frequencies of 11.67 and 88.33%, respectively. When the data were analyzed with the Structure program it indicated that the best fit to the data was one and not three populations [23]. The highest probability was observed with a simulation of one population with an overall decrease in probability and increase in variance with increasing numbers of populations (Fig. 5). Moreover, when increasing numbers of clusters were used, all three populations partitioned into these clusters equally (data not shown). These results support the hypothesis that these populations can be treated as a single deer population. When data from all populations of mule deer were treated as a single group, it was found that locus Rt30 contained the greatest number of alleles, locus OheN showed the highest observed heterozygosity, locus BM1225 had the fewest alleles and locus BM4208 had the lowest observed heterozygosity (Table 1). Most loci were not dominated by a single allele but had several common alleles. Of the 101 alleles that were detected only 2 had frequencies that were over 50 percent (Table 2). When the deer populations were analyzed separately and compared, the heterozygosity rates and number of alleles were similar for all populations (Table 3). The pooled mule deer data produce a probability of identity of 1 in 1.12  1010. Estimates of genetic differentiation

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

195

Fig. 4. Electropherogram of amelogenin fragments from mule deer. The X-axis measures size in base pairs. The Y-axis measures intensity of fluorescence in relative fluorescent units. The upper panel shows (223.22 bp) fragments from a female mule deer containing the blue (FAM) fluorescent tag. The third panel shows (170.50 and 214.56 bp) fragments from a male mule deer containing the blue (FAM) fluorescent tag. The second and bottom panels show fragments from the GenScan1 400HD [ROX] size standard (Applied Biosystems, Foster City, CA, USA).

between populations (F ST) showed low levels of differentiation (0.0146–0.0280) (Table 4). The overall F ST value for the entire population was 0.0205. The overall coancestry coefficient, theta (u), was 0.0200. 3. Discussion The DNA-typing tests and databases that were developed for mule deer in this study are appropriate for forensic use. The

measures used in calculation of the random match probabilities in mule deer (see Section 1) assures that a conservative estimate will be presented to the courts. Furthermore, given that the estimated population of mule deer in Alberta is 133,000, the degree of discrimination provided by the tests and database is more than adequate for the production of forensically significant results [29]. Results from the amplification of the sika amelogenin primers indicate that fragments from the X and Y chromosomes

Fig. 5. Data from the structure program. The X-axis represents the number of population clusters. The Y-axis represents the likelihood [ln(X/K)] of genetic population clusters based on the STRUCTURE analysis of mule deer populations.

196

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

were amplified. This hypothesis is supported by the fact that the smallest fragment, the 170.50 bp fragment, was always found in males and never detected in females. These results suggest that mule deer, like sika deer, possess amelogenin genes that contain a deletion on the Y chromosome [18]. Larger fragments were found in both males and females, suggesting that they originate from the amelogenin gene on the X chromosome that was not found to have a deletion in sika deer [18]. Interestingly, two size variants were detected, 214.56 and 223.22 bp fragments. The fact that these fragments were found in equal relative proportions to each other in both males and females also supports the hypothesis that these fragments originate from the X chromosome of mule deer. Taken together these results indicate that the sex-typing test presented in this study accurately identifies the gender of mule deer and is appropriate for forensic use. The DNA-typing markers and databases described in this study have already been used to protect mule deer in Alberta. In September of 2006, a landowner in Spirit River, Alberta reported finding a headless mule deer carcass on his property. A fish and wildlife officer attended the scene and found a headless mule deer carcass that had been shot with a rifle. The attending officer collected samples from the carcass. The complainant had taken photographs of a very large male mule deer with distinctive antlers that he had recently observed on his property. As part of the investigation, the Alberta officer contacted her counterpart in nearby Dawson’s Creek, British Columbia (BC) to see if anyone had reported harvesting a large mule deer. The officer from BC reported that a local resident had killed a large male mule deer and that there was a picture of him with his trophy in the window of a taxidermist’s shop. The officers met at the shop compared the picture in the window with the complainant’s picture and found that they appeared to be of the same animal. The antlers and lower jaw of the suspect’s deer were seized and all of the collected evidence was submitted to the forensic lab. Analysis at the forensic lab found that a DNA match existed between the headless carcass and the antlers/jaw. The DNA match consisted of 10 loci and produced a random match probability of 1 in 1.7 billion when using the Fort St. John database. This investigation has led to numerous charges being laid by both BC and Alberta officers but has yet to go to trial. This case illustrates the utility and power of this technique, particularly when an individual is in possession of a large and conspicuous trophy. The development of forensic DNA-typing capability has given our enforcement personnel the ability to prosecute cases that would not have been possible in the past. It is hoped that as news of this new technology spreads in the public that it will act as a deterrent to would be perpetrators of wildlife offences in the future. In addition to serving as an effective tool for enforcement, information obtained from analysis of the mule deer databases may be of interest to individuals who study deer in an academic and/or management capacity. A previous study has also developed a panel of mule deer markers for forensic use in the state of California [9]. The California study used some markers that were different from those used in the current study and reported an average of 10.25 alleles per locus and a probability

of identity of 1 in 1.3  1010, which are similar to the present results. Rather than testing and comparing localized populations of deer to test for state-wide genetic structure, the California study analyzed mule deer samples taken from the entire state and treated them as a single population. This study did acknowledge that it did not account for population structure but elected to use the fact that the study found that no two deer shared the same profile as a statistic for court purposes. In the Alberta study, it was decided to adopt a strategy that examines and accounts for population structure and uses random match probabilities. This strategy was selected because it is the method that is widely accepted in the forensic community [27]. Using this method also does not exclude representing the collected data as a ‘‘no matching samples’’ statistic if the court requests or requires it. Conversely, if samples have not been collected to represent distinct geographic areas they should not be used to calculate random match probabilities. Although, the current study found that mule deer in Alberta can be treated as a single population, it is not a safe assumption to make for all deer populations. A study of white-tailed deer in Southeastern USA, found a significant amount of population structure in that population [30]. The danger in assuming that one population exists when in fact more than one population is present is that calculated allelic frequencies will actually be averages of several genetically different populations. In most cases, the random match probability of an individual will be lower (more likely) in an individual’s ‘‘home’’ or native population than it is in other populations. This tendency is found because an individual is more likely to possess alleles that are the most common in its native population. The use of averaged allelic frequencies will tend to reduce the frequency of locally common alleles and therefore raise random match probabilities and create a bias against the defendant. Work performed on white-tailed deer populations in the Southeastern USA, show average number of alleles ranging from 5.9 to 9.2, observed heterozygosities ranging from 59 to 68 percent and coancestry coefficients ranging from 0.039 to 0.075 [11,30]. Results from the current study show comparable values for number of alleles and observed heterozygosities but show a lower coancestry coefficient (u = 0.020). From a deer management perspective, the present study shows that mule deer in the Province of Alberta have high levels of genetic variability when compared to other wild deer populations. Additionally, F ST and structure results indicate that there is no major impediment to gene flow in the province and that the mule deer in the Province of Alberta may be managed as a single large population of deer. In summary, the present study represents a set of forensic DNA-typing tests and databases for mule deer that are appropriate for the analysis of evidence for legal purposes in the Province of Alberta. This study also found; levels of genetic variation that were similar to other wild deer populations, no indication of the existence of reproductively isolated or inbred populations of mule deer, and no indication of the existence of any barriers to gene flow. Therefore, no change to management of the studied population is suggested.

R.M. Jobin et al. / Forensic Science International: Genetics 2 (2008) 190–197

Acknowledgments This work was supported by funds from the Government of Alberta, the Alberta Professional Outfitters Society and the TD Friends of the Environment Foundation. In-kind support was provided by the Alberta Conservation Association. References [1] A.E. Anderson, O.C. Wallmo, Odocoileus hemionus, Mamm. Species 219 (1984) 1–9. [2] M.E. Nelson, L.D. Mech, Mortality of white-tailed deer in Northeastern Minnesota, J. Wildl. Manage. 50 (4) (1986) 691–698. [3] S.M. McCorquodale, Movements, survival and mortality of black-tailed deer in Klickitat Basin of Washington, J. Wildl. Manage. 63 (1999) 861–871. [4] T.K. Fuller, Dynamics of a declining white-tailed deer population in northcentral Minnesota, Wildl. Monogr. 110 (1990) 1–37. [5] P. Gill, A.J. Jeffreys, D.J. Werrett, Forensic application of DNA ‘fingerprints’, Nature 318 (1985) 577–579. [6] C.P. Kimpton, P. Gill, A. Walton, A. Urquart, E.S. Millican, M. Adams, Automated DNA profiling employing multiplex amplification of short tandem repeat loci, PCR Methods Appl. 3 (1993) 13–22. [7] J.E. Lygo, P.E. Johnson, D.J. Holdaway, S. Woodroffe, J.P. Whitaker, T.M. Clayton, C.P. Kimpton, P. Gill, The validation of short tandem repeat (STR) loci for use in forensic casework, Int. J. Legal Med. 107 (1994) 77–89. [8] M. Poetsch, S. Seefeldt, M. Maschke, E. Lignitz, Analysis of microsatellite polymorphism in red deer, roe deer and fallow deer—possible employment in forensic applications, Forensic Sci. Int. 116 (2001) 1–8. [9] K.C. Jones, K.F. Levine, J.D. Banks, DNA-based genetic markers in black-tailed and mule deer for forensic applications, Calif. Fish Game 86 (2000) 115–126. [10] E.A. Guglich, P.J. Wilson, B.N. White, Application of DNA fingerprinting to enforcement of hunting regulations in Ontario, J. Forensic Sci. 38 (1993) 48–59. [11] R.W. DeYoung, S. Demarais, R.L. Honeycutt, R.A. Gonzales, K.L. Gee, J.D. Anderson, Evaluation of a DNA microsatellite panel useful for genetic exclusion studies in white-tailed deer, Wildl. Soc. B 31 (2003) 220–232. [12] R.A. Moscarella, M. Aguilera, A.A. Escalante, Pylogeography, population structure, and implications fro conservation of white-tailed deer (Odecoileus virginianus) in Venezuela, J. Mammal. 84 (2003) 1300–1315. [13] H.G. Broders, S.P. Mahoney, W.A. Montevecchi, W.S. Davidson, Population genetic structure and effect of founder events on the genetic variability of moose, Alces alces, in Canada, Mol. Ecol. 8 (1999) 1309–1315.

197

[14] D.K. Patterson, R.M. Jobin, Assessment of DNA extraction methods for forensic wildlife applications, in: 12th International Symposium on Human Identification, 9–12 October 2001, Biloxi, Mississippi, USA, 2001. [15] M.D. Bishop, S.M. Kappes, J.W. Keele, R.T. Stone, S.L.F. Sunden, G.A. Hawkins, A genetic linkage map for cattle, Genetics 136 (1994) 619–639. [16] J.A. DeWoody, R.L. Honeycutt, L.C. Skow, Microsatellite markers in white-tailed deer, J. Hered. 86 (1985) 317–319. [17] G.A. Wilson, C. Strobeck, C.L. Wu, J.W. Coffin, Characterization of microsatellite loci in caribou Rangifer tarandus, and their use in other artiodactyls, Mol. Ecol. 6 (1997) 697–699. [18] K. Yamauchi, S. Hamasaki, K. Miyazaki, T. Kikusui, Y. Takeuchi, Y. Mori, Sex determination based on fecal DNA analysis of the amelogenin gene in sika deer (Cervus nippon), J. Vet. Med. Sci. 62 (2000) 669–671. [19] S.D.E. Park, Trypanotolerance in West African Cattle and the Population Genetic Effects of Selection, Thesis, University of Dublin, Dublin, Ireland, 2001. [20] M. Raymond, F. Rousset, Genepop (version 1.2), population genetics software for exact tests and ecumenicism, J. Hered. 86 (1995) 248– 249. [21] W.R. Rice, Analyzing tables of statistical tests, Evolution 43 (1989) 223– 225. [22] J. Goudet, FSTAT (version. 1.2): a computer program to calculate Fstatistics, J. Hered. 86 (1995) 485–486. [23] J.K. Pritchard, M. Stephens, P. Donnelly, Inference of population structure using multilocus genotype data, Genetics 155 (2000) 945–959. [24] N. Valiere, Gimlet: a computer program for analyzing genetic individual identification data, Mol. Ecol. Notes 2 (2002) 377–379. [25] R. Chakraborty, Sample size requirements for addressing the population genetic issues of forensic use of DNA typing, Hum. Biol. 64 (1992) 141– 159. [26] I.W. Evett, P. Gill, A discussion of the robustness of methods for assessing the evidential value of DNA single locus profiles in crime investigations, Electrophoresis 12 (1991) 226–230. [27] National Research Council, Evaluation of Forensic DNA Evidence, National Academy Press, Washington, DC, 1996. [28] R.M. Jobin, Use of forensic DNA typing in wildlife investigations, in: H.M. Coyle (Ed.), Nonhuman DNA Typing Theory and Casework Applications, CRC Press, Boca Raton, 2007, pp. 99–116. [29] Alberta Sustainable Resource Development. Information on species that may be hunted in Alberta. Accessed 04 October 2007. http://www.srd. gov.ab.ca/fishwildlife/livingwith/huntingalberta/muledeer.aspx, 2002 [30] R.W. DeYoung, S. Demarais, R.L. Honeycutt, A.P. Rooney, R.A. Gonzaes, K.L. Gee, Genetic consequences of white-tailed deer (Odecoileus virginianus) restoration in Mississippi, Mol. Ecol. 12 (2003) 3237–3252.