Is your pet legal? Assessment of DNA analyses used in wildlife forensics

(1)

Is your pet legal? Assessment of DNA

analyses used in wildlife forensics

Citlalli Limpens

December 2019

Master Forensic Science

Literature Thesis

Student: Citlalli Limpens - 12267988

Number of words: 7963

Supervisor: Dr. Patrick Meirmans

Examiner: Dr. J. A. J. (Hans) Breeuwer

(2)

1 Abstract

Wildlife forensics is the application of science to the law enforcement of crime against wild animals and plants. A big part of wildlife forensics, consists in DNA-based molecular analyses. This review explains three of the most common DNA analyses used in wildlife forensics: species identification, kinship analysis and source location, highlighting the most used markers and techniques, with case examples. The main differences these analyses have with human DNA forensics are the need of identification of more than one species; the lack of standardized protocols and the absence of datasets and genomes for primer design. The biggest difference being the first one mentioned, since out of the analyzed papers, most of them focused on species identification, since most of the times that is enough to define a crime against wildlife. Most of the species the papers were focused on were mammals, since human empathy relates to mammals the most. This review concludes with the crucial remark that collaboration is needed between ecological biologists and statistical geneticists with forensic scientists to stop crime against wildlife.

(4)

2 Introduction

Even though different organizations and police departments have different definitions for it, wildlife forensics can be summarized as the use of law enforcement and investigation to combat crimes against wild flora and fauna (Alacs et al. [2010]; Gupta [2018]; Huffman and Wallace [2012]; CITES [2019]). This investi-gation includes questions regarding the identification of live, deceased and derivatives of animals and plants, in order to solve law enforcement needs. These questions are subdivided by the CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) into five categories: (1) Identification of the species involved, (2) Geographic origin of the items in question, (3) whether the source of the item (or individual) in question is from wild or captive raising, (4) questions regarding individual source of an item, and (5) age, sex and other specific details on the individual (CITES [2019]; Ogden and Linacre [2015]).

According to Interpol, CITES and many other sources (Alacs et al. [2010]; Gupta [2018]; Huffman and Wallace [2012]; Nellemann et al. [2016]; CITES [2019]) billions of dollars are in circulation thanks to international wildlife trade. The products traded range from live animals and plants, to particular sections of an individual to cover a specific marketed need (traditional medicine, ornamental artifacts, musical in-struments, delicatessen, etc.). The amount of money and goods in circulation, make illegal wildlife trade the fourth largest transnational organized crime (Wasser et al. [2015]; Nellemann et al. [2016]). A big contrib-utor to its success is that wild life crime comes with a low risk-high reward ratio, making it profitable and attractive. The low risk part of the ratio comes from the low chance to get caught while committing these crimes, and a milder punishment when compared to other illegal transactions. The high rewards mostly comes from the capture of rare species, which are the most profitable ones. This ends up leading to more exploitation, leading, in the end to extinction. An example of a species that suffered from over exploitation due to its profitability, is the African rhinoceros. A decline in rhinoceros population is depicted in Figure 1, showing how a dramatic loss in the population is correlated with a growth of its poaching in Southern Africa. This figure also presents a world map with the ”hot spots” both for capturing and trading the animal. The exploitation of endangered species, in addition to other human related factors (such as habi-tat loss) give as a result a population depletion that brings species closer and closer to extinction. For this reason, institutions and international conventions, such as CITES, develop agreements between (and within) countries to ensure that international trade in wildlife is not a threat to the survival of a particular species. These conventions apply to both endangered and non-endangered species, in order to ensure sustainable trade for all of them. CITES currently has 37 000 species on their convention. A problem with these regula-tions however, is that the only way wildlife crimes are discovered by the relevant authorities is while crossing international borders, since a lot of countries rich in wildlife do not carry appropriate surveillance (Gupta [2018]; Huffman and Wallace [2012]; CITES [2019]; Bai et al. [2003]).

Once a crime against wildlife is discovered, the relevant forensic questions mentioned above need to be addressed. The first step being identification of the species involved. Previously, identification was based on morphological features by the means of identification keys. This method is still used when possible, however most smuggled and traded remains do not conserve enough morphological features to be identified. For this reason, molecular techniques are the most used ones to solve the five forensically relevant questions. Non-DNA based techniques, such as SEM-EDX, or volatilome analysis can also be used in wildlife forensics, however, this paper focuses only on DNA molecular techniques (Ueland et al. [2020]).

To assess whether the presented techniques have helped with endangered species regulations and safety, this thesis has the following objectives: (1) to review DNA analyses currently used in wildlife foren-sics, with a particular focus on kinship, location and species identification; along with the statistical methods used for their analysis, (2) To assess the differences in validation, techniques and availability between animal and human forensics, and (3) to evaluate the progress of endangered species regulations and trade. For this, I will start by presenting a literature overview in which molecular techniques used for species identifica-tion, individualizaidentifica-tion, parentage analysis and source location are explained with case examples for different species. Followed by the brief explanation of differences between human and animal forensics and validation in wildlife forensic. Finally, discussion, conclusions and recommendations are presented.

(5)

Figure 1: African Rhino smuggling. The figure indicates a map with the traffic route taken by the smugglers, including the “hot spots”, marked in red. The graph at the bottom left indicates the number of Black Rhinos inhabiting Africa, with a dramatic population loss between 1960 and 1980. The graph at the bottom right indicates reported incidents of killed Rhinos. Figure taken from Nellemann et al. [2016].

(6)

3 Literature overview

3.1 Techniques used for DNA analysis in wildlife forensics

3.1.1 Species identification

Species identification is the first question to be answered in Wildlife Forensics. The reason for this being that most of the time, when it comes to illegal trade, the only thing needed for identifying a crime (and the degree of it) is the species involved in it. This is also relevant for food forensics in cases of food fraud. When the origin of a piece of meat is disputed, the mere identification of the species is enough to conclude whether or not a crime is committed.

For species identification, Mitochondrial DNA (mtDNA) markers are often used due to their sta-bility even in the processed samples that are usually encountered in wildlife forensics. It also comes with the advantage that universal primers are readily available for this purpose. The most common markers are mitochondrial cytochrome b, and cytochrome oxidase 1 genes (Cyt b and CO1 respectively). The Cyt b gene has previously been used to identify vertebrate species from illegal trade products from fish (sharks), reptiles (turtles, alligators, snakes), mammals (seals, elephants, tigers) and birds. The method that was first used to differentiate species using this marker is PCR-RFLP (Polymerase Chain Reaction-Restriction Fragment Length Polymorphism). This technique uses restriction enzymes as well as a PCR reaction to view species-specific patterns in the samples. The downfalls this technique has for its application in forensic science, is that it has to be validated for each species it is applied for. For this, phylogenetic trees need to be available for comparison, as well as other information such as hybridization rates between other species in the genus. Gene selection is also an important factor to take into account, since they need to have a high enough mutation rate for its differentiation between species, but stable enough so that all individuals in the same species have it the exact same way (Alacs et al. [2010]; Karlsson and Holmlund [2007]). Even with these downfalls, the technique is still used due to the lack of funding research and innovation has in Wildlife forensics. According to Nishant et al. [2017] the technique was still commonly used in 2017, particularly when referring to casework and actual forensic application.

Recent studies, however opt for a PCR amplification of the mtDNA genes of interest, and subsequent sequencing of the amplicons (Kumar et al. [2019]). Once the ampolicon sequences are obtained, by database search, the researches can obtain an accurate species identification. This method, of course will depend on the presence of the species of interest in the database. Once again, since the markers used are highly conserved, universal primers can be used and the online databases grow every year, making it a reliable and fast-growing technique. This technique has been used to identify mammals, reptiles and birds from poaching in India (Kumar et al. [2019]; Kundu et al. [2019]), as well as for the identification of doubtful meat products (Ghosh et al. [2019a]; Ghosh et al. [2019b]; Jabin et al. [2019]). Most of these studies focus on the amplification and sequencing of the Cytb gene, however one study also used CO1 for one of the species (Kumar et al. [2019]). Another sequencing style that was previously used in Wildlife Forensic, is pyrosequencing, where short DNA fragments (between 10-500 base pairs) are sequenced instead of the full genome, with the purpose of direct species identification. This method has previously been used to design assays for quick species identification for different mammal species in Europe (Karlsson and Holmlund [2007]). Out of the methods explained, PCR-RFLP, and sequencing of PCR products using mtDNA mark-ers are the most commonly used. However, since the markmark-ers are highly conserved, closely related species might be difficult to differentiate. mtDNA markers would also give false results when dealing with hybrid animals, since only the species of the mother would be identified. For this reason, it would be ideal to mix the use of mtDNA markers with nuclear ones. However, nuclear markers are not currently available for most wild species, since their genomes are not widely available (Karlsson and Holmlund [2007]).

(7)

Beginning with issues concerning the fishing industry. According to Palumbi and Cipriano [1998], by the time their paper was written, 69% of the fisheries were overexploited or fully fished. This percentage raised to 90% (61% fully fished and 29% overfished) by 2018 (UNCTAD [2018]). For this reason, wildlife forensics application to marine wildlife has become crucial to protect over exploited and protected areas and species. One important aspect is the relationship between fisheries and whale consumption. Several efforts have been made to stop illegal trade of whale meat in asian markets. All these efforts focus mostly on the identification of the species the particular piece of meat comes from.

In some cases, even the identification of individuals is of use. In Japan, for example, commercial trade of whales is legal when the animals were previously used for research. In these cases, proving that the meat found in the markets actually came from a researched whale becomes useful. Genetic information of researched whales is stored in research centres and universities, thus its comparison with genetic material extracted from markets confirms the legality of the meat. mtDNA markers were used for species identifi-cation purposes in the late 1990s (Palumbi and Cipriano [1998]), and for individualization and comparison with research databases in 2010 (Baker et al. [2010]).

Apart from whales, other species, such as sharks are also affected by international overfishing. A problem when such animals are affected, is that unlike whales, sharks are not flagship species, so conser-vation efforts are not as effective as they are with animals that appeal to the general public. Conserconser-vation efforts should also take into account specific biological characteristics of the animals, since their behaviour and needs have different requirements. Shark finning makes morphological species identification difficult, so molecular methods are needed. Multiplex PCR proved to be a cheap and quick alternative to sequencing techniques for seven shark species identification, with species specific primers from nuclear ribosomal loci, and Cytb mtDNA, and shark universal primers as a control (Shivji et al. [2003]). Unfortunately, techniques such as PCR amplicon sequencing haven’t reached the sea, and have only been used to identify meat samples coming from terrestrial animals (Ghosh et al. [2019a]).

Meat identification proves useful in contexts outside of the ocean as well. For the case of meat origin, not only wildlife forensics is involved, also food forensics concerning costumer’s complaints of doubtful meat samples they purchase. Usual customer complaints happen with processed meat such as jerky, bacon or sausages, since no matter the animal source, the end product will have the same physical appearance. Cytb genes are usually amplified, sequenced and compared in meat-species databases. If database informa-tion availability were as wide for wildlife as it is for farming animals, this technique could be easily used to analyze doubtful meat sources both to clarify customer’s concerns and to confirm cases of bushmeat sales (Linacre et al. [2003]; Ghosh et al. [2019a]; Ghosh et al. [2019b]; Jabin et al. [2019]). Another way to identify meat origin source, is the use of pre-made commercial kits. Since this area of forensics has more funding and prioritization (since it deals with human concerns), kits such as SpInDel have been adapted to identify via multiplex PCR illicit poaching species such as deer and fox. this tehcnique however, is only used as a preliminary measure, and has only been used in casework for human DNA identification (Pereira et al. [2019]). Another available kit for forensic Species Identification, is HyBeacon probe kit. These probes are design for their specific association with species-specific DNA regions, with the advantage of only needing a couple of cells to work. Unfortunately, they have only been designed for some feline species, rhinoceros and pangolin (George et al. [2019]).

Bushmeat can come from any wild animal source. Starting as a survival mechanism, it ended up as profitable market by charging high rates for exotic animal meat. In the Congo area, for example, bushmeat hunting is responsible for local extinction of several species, such as the red Colobus monkey reported extinct in 2000. As stated before, most wildlife crimes are discovered with border crossing. This becomes a problem with bushmeat trade, since in countries from West Africa the goods usually stay within a local environment, making it difficult to discover and prosecute. Identification usually occurs at airports and customs clearance, and bushmeat products always require DNA techniques since the animal is unrecognizable by morphological characteristics. The techniques used for identification in these cases are mostly PCR-RFLP and sequencing from PCR amplicons, using mtDNA Cytb markers. The only downfall is that when the origin of the meat is completely unknown, the lack of reference material could give an inconclusive result. An alternative to

(8)

solve that problem, was proposed by Kelly et al. [2003], in which sample collection from local zoos proved useful for comparison analyses. The same technique has also been used to identify poached species, such as bears, dear and foxes (Bai et al. [2003]; Ghosh et al. [2019a]; Ghosh et al. [2019b]; Jabin et al. [2019]). Ruminant identification in meat is a field well studied and with available databases and guidelines. Future improvements for wildlife forensics could include the application of similar datasets for species other than domestic animals (Lenstra et al. [2003]). An example of this, is the development of algorithms for barcode and sequencing identification, specifically adapted for Wildlife forensic purposes. This algorithm was devel-oped by Conde-Sousa et al. [2019], and it uses online databases such as GenBank for direct identification of sequenced data. This algorithm was only tested with members of the Canidae family, and hasn’t been tested in casework yet.

Species that are on the verge of extinction thanks to poaching are Rhinoceros and Elephants. Mostly as ornaments (ivory), traditional medicine (rhino horn) and meat. For the case of ivory and pulver-ized horns, species identification can be difficult due to the treatment the animal products receive. Again, Cytb mtDNA PCR products are used for species identification. In cases like this, collaboration between areas such as forensic sciences and evolutionary biology is needed for primer design and database creation (Bollongino et al. [2003]). So far, the techniques presented are used exclusively in the forensic area, however researchers involved in population genetics, also develop tools and approaches towards species identification. These studies focus on finding markers and techniques to identify particular species, and to make pedigree analysis to make sure populations do not encounter bottleneck situation (Magonyi et al. [2019]). Collabora-tion with these researchers and adaptaCollabora-tion of their techniques for their use in forensic species identificaCollabora-tion seems important to expand the number of identifiable endangered animals and plants.

3.1.2 Kinship analysis

Parentage and kinship analysis can be used in wildlife forensics to identify the origin of animals bred in captivity. The techniques used in wildlife forensics are mostly adapted from human kinship analysis, and statistical genetics. The uses of these techniques outside of the forensic area are: (1) Estimating heritability of traits in the wild, (2) minimizing inbreeding in populations bred in captivity, (3) estimating rates of gene flow in populations, (4) adjusting population’s allele frequencies by knowing relatedness in one sample, (5) estimating breeding individuals in a population, and (6) estimating reproductive success of individuals in a population. All these uses are based on relatedness estimation, which is the calculation of how close in a pedigree two individuals may be. It is usually done by calculating the amount of alleles in both individuals that are Identical by Descent (IBD). The success of this method depends on previously calculated allelic frequencies, in order to make a proper estimation of the probability of two individuals being related. Unfor-tunately this is not possible for all wildlife cases, since allelic frequencies are not available for every species. To make these calculations Hardy-Weinberg Equilibrium is often assumed, and sometimes allelic maps1 are also needed in order to choose the alleles used for the comparison (Blouin [2003]). However, kinship analy-sis comes with the additional benefit that breeders use DNA registers, to ensure that their stock is legally acquired. This helps with comparisons, as explained in the case example at the end of this section.

As in the species identification section, collaboration between population geneticists and foren-sic scientists becomes crucial when finding information regarding population dynamics and useful markers. Looking for state of the art literature in ecology and genetics is the first needed step before performing analyses. Population structure information is particularly useful when assessing parentage relationships be-tween individuals. Comparing these relationships in the wild with what happens in captivity should not be done without prior knowledge. Studies such as the one by Jacobs et al. [2018], in which STR markers were used to assess the familial links in a close lemur society, would help forensic investigations for a particular species, even when the studies per se don’t have an obvious forensic relationship. Studies like these have also been made for birds (White et al. [2012]) and reptiles (Mucci et al. [2018]), where forensic applications were performed to identify licit breeding of tortoises and cockatoos. All of these studies analyze STR markers,

1_{An allelic map in this case refers to the geographycal location of certain allelic frequencies found in different populations}

(9)

the same way it is done in human kinship and paternal testing.

A group of animals that is frequently neglected when it comes to forensic cases are reptiles. Most movements involve mammals, since human empathy relates easier to them than with other animals, leading to stricter regulations and sanctions regarding this group. Reptiles are the second most common group involved in illegal trades and seizures worldwide, and they dominate the international black market for pets CITES [2019]. Individualization and paternal analysis is important for this group, since it helps prove whether or not an animal is bred illicitly. This process involves a multiplex STR assay, as well as a database comparison, as an adaptation of the same techniques used in mammals. Although fairly new, and lacking casework validation, this protocol seems promising for both individual identification and identity testing of snakes, and it is a nice example of what can be done for more neglected species. Hopefully efforts like this can involve more species, and more markers in the future (Ciavaglia and Linacre [2018]).

Even though parentage analysis is mostly used for confirming the legality of breeding exotic an-imals, it has also been used in human forensics. One example of this is in a tortoise theft case, in which parental analysis was used to prove who the original owner of the animals was. In this case, 12 STR markers were analyzed and compared between recovered stolen individuals and the alleged parents, belonging to the original breeder (Mucci et al. [2014]). Trade restrictions make breeding and commercializing with exotic animals profitable, which is why theft and capture of them becomes an option with low stakes and high reward. Unfortunately cases like this get more law enforcement attention because they deal with the animals as an individual’s good instead of as the main focus of the investigation.

3.1.3 Source location

Source location is important for finding out whether populations have been moved, as well as to find “hot spots” for poaching and other illegal activities involving wildlife. The problem with source location identification in wild species, is that there needs to be a known genetic structure (known allelic frequencies) of the populations of interest. The reason for this is because separation by source is based on different al-lelic frequencies, particularly hypervariable regions of nDNA; or specific haplotypes from a population. This makes the tests dependent on whether or not disputed populations have been previously sampled, population stratification is well known and defined, and they have been tested for HWE (Klein and Moeschberger [1997]). Two of the main species for which these studies have been performed are African Elephants and Rhinoceros. African Elephant ivory constitutes a major part of wildlife trade. Assigning geographic origin to ivory seizures could help bring new regulations to the places where it is most needed. This is done by using allelic frequency maps from the populations, in this case forest and savanna elephant populations (Wasser et al. [2004]). Most of the samples in this case were assigned to the proper population, within around 400 km accuracy. AS expected, the more isolated the populations were, the more accuracy their identification had. The results from this paper allowed law enforcement to take major changes in the spots were most of the ivory samples were presumably from (Wasser et al. [2015]). Unfortunately, allelic frequency maps are not currently available for most species, thus collaboration is needed with population geneticists to gather this information for forensic cases involving different wildlife. A problem with this method would be, as it is in the whale case, that hybridization may happen in natural populations. Luckily for the elephant case, hybridization appeared to be rare, but that may not be the case for other species. For this reason, testing for hybrids or for population admixture in the populations of interest is an important first step for these analyses. Another thing to be considered, is that elephant populations are widely distributed, giving a far apart set of populations, making the mapping easier and possible. This makes the mapping species-specific and considerations on the ecological dynamics and biology of the species is crucial to create an accurate map. An example of ecological efforts, that ended up having forensic relevance is the paper by Bielikov´a and Turna [2003]. In this case, falcons from Czechia and Slovakia (Falco rusticolus, Falco peregrinus, Falco cherrug) were analyzed in order to create a frequency distribution map for future forensic use. Out of the analyzed papers, this was the only effort for this type of analysis with a forensic purpose that didn’t involve

(10)

big mammals. They however, conclude that the techniques cannot be used for individualization nor for identification of nest robbery cases, and that the mapping was not possible with alleles chosen by the group, since no significant differences were found between the bird populations in different geographic areas. The only significant differences their group could find were between captive and wild individuals. This proves that using techniques that work in a certain way for a species might not be usable for another, making wildlife forensics trickier than human forensics (Bielikov´a and Turna [2003]).

Results are always dependant on the statistical model of choice, and choosing the model needs to be done by taking into account the species’ biology, as well as its interaction with humans (knowing what forensic question needs to be solved). Source population for finding killing hotspots as mentioned above is different than when the main focus is location of individuals from areas with different management condi-tions and regulacondi-tions. For the first case, a priori population frequencies are used. As mentioned before, data may not always be available for wildlife, so a Bayesian assignment test might be useful in most cases. This assignment clusters individuals according to their own alleles, giving as a result clusters based on the given evidence Ball et al. [2011]. Genetic divergence is important to take into account when considering the populations in both cases, as well as the markers to be used. The comparison between both techniques has been done for moose populations, proving the need to know previous information on population structure, or at least the sampling of control individuals from every disputed population. The data used for this paper, was obtained from samples obtained by Fish and Wildlife officials in the United States. Since their local regulations enforce them to sample every encounter individual for the creation of a reference database. Ef-forts like this in other countries (or at least in other natural reserves) would help geographical assignment of individuals in the future, by keeping records of most members of the population for future reference (Ball et al. [2011]).

A difficult defining point when dealing with ecology is: when does a species begin and end? How different do two populations need to be to be able to be recognized genetically? Techniques used are both species specific, and specific to the forensic question one wants to answer. Just as it is used with species identification, mtDNA-Cytb is also used for population genetics, so it is used for geographical identification. Unfortunately, populations need to be genetically distant to be able to be distinguished by this method. What makes this method so good and widely used for species identification (inter species variation/intraspecies conservation) is what makes it difficult for population differentiation. For this reason, hypervariable mito-chondrial sections could be used as an alternative, such as the D-loop to distinguish specific lineages. This has previously been used to identify Echidna populations in Australia (Summerell et al. [2019]). By amplify-ing, sequencing it, and comparing haplotypes, the autors were able to succesfully determine source country of their samples. This made this protocol, the first forensically validated protocol to determine source location of disputed individuals.

Nuclear DNA can of course also be used for source population identification. Nuclear markers used include STRs and SNPs. They allow a high distinction between populations by being highly variable, how-ever, they require previous knowledge on the population genetic structure. The more genetically apart two populations are, the less markers may be needed to tell them apart from each other. The choice of analysis once the DNA data is gathered from the questioned individuals, will depend upon available information and number of samples. The most accurate representation would be by creating allele frequencies maps, to assign with certain probabilities an individual to a particular part of the map, the way it was done by Wasser et al. [2004]. This is however unrealistic for a lot of wild species, so clustering information may be the analysis of choice. Clustering software programmes also exist for facilitating the analysis, such as GeneClass (Piry et al. [2004]) and STRUCTURE (Pritchard et al. [2003]). The use of a software, even though it would make the process faster and less tedious for the workers, will also require enough reference material to be able to successfully assign individuals to their origin population (Ogden and Linacre [2015]). Methods for geographic origin assignations, since they are very variable from species to species are not routinely used, and need thorough prior validation. For this reason, this is not as used and accepted as species identification is. However, since the first official protocol was just validated last year (Summerell et al. [2019]), we can expect more advances in this area of forensic science in the future.

(11)

3.2 Differences between human forensic DNA analysis and wildlife forensic

DNA analysis

The main difference between human and wildlife forensic DNA analysis involves the forensic question the analysis is used to solve. While human forensics is focused on identification of the individual who left the sample, wildlife forensics questions rely on many other topics, such as the ones presented above. Table 1 presents a summary of differences regarding this topic. As it can be seen, taxonomic identification and origin determination are the main focus on wildlife cases, while individual identification is the main concern in human related cases. Forensic questions that have more similarities between the two fields are human paternity cases and Disaster Victim Identification (DVI). These are more similar to wildlife forensics pedigree analyses, however reference human databases and material are always available which is not the case for animal forensics.

Forensic Questions DNA in human cases DNA in wildlife cases

Taxonomic identification Rarely needed Commonly needed - mtDNA

Individual identification Commonly needed - nDNA Sometimes needed - mtDNA, nDNA

Number of individuals in a mix Commonly needed - nDNA, mtDNA Sometimes needed - mtDNA, nDNA

Sex identification Commonly needed - Y STR Sometimes needed - species dependent

Origin determination Not needed Commonly needed - captive or wild origin

Geographic assignment Not needed Commonly needed - mtDNA, nDNA

Kinship analysis Commonly needed Commonly needed - nDNA

Table 1: Differences between forensic questions for samples containing human DNA and wildlife DNA. Table adapted from Moore and Frazier [2019]

An aspect to be considered, particularly regarding the most common forensic question in wildlife (what species does the sample come from?) is that wildlife includes a whole array of organisms, while human DNA analyses only need to focus on one. New taxa, and new species require new methods or new databases, thus validation is often needed every time these are presented. Appropriate loci and method determination is required by species, which makes validation individual for each of them, even when the same method is used. New used loci need also to conform with the expected behaviour (HWE) to be used for the analysis, and proof of this is difficult to gather, but necessary for proper validation (Moore and Frazier [2019]). Table 2 presents a comparison summary of available tools for both animal and human DNA forensics.

Humans

Animals

Species involved

One

Any

Kit availability

Available

For some species - rarely

Standardized loci

Available

For some species - rarely

Allelic ladders

Available

For some species - rarely

Databases

Available

For some species - rarely

Reference genome

Available

For some species - sometimes

Mapped STR locations

Available

For some species - rarely

Developmental validation

Rarely

Available

Internal laboratory validation

Available

Table 2: Availability differences between tools used for human and animal forensics. Table adapted from Moore and Frazier [2019].

An unfortunate but important difference is that laws protecting wildlife are not as strict as laws protecting humans. This results in investigations receiving less priority and funding. Law usually views wildlife as an item for human use and/or property. This results in crimes involving human forensics such as property theft (tortoise case) or food fraud receiving higher attention than crimes that are completely against wildlife (such as illicit killing of an endangered species). An example of this being the very first use

(12)

of animal DNA forensics, which was by the use of pet genetics to link suspects to a crime. It should also be acknowledged that in order to solve some wildlife forensic cases, human DNA identification may be needed and of great aid to the investigation (Menotti-Raymond et al. [1997]). For example, when encountering illegally obtained animal material, human DNA could be found on it and used to link a suspect to that particular crime.

3.3 Future developments

As sequencing techniques become cheaper and readily available, non-model organisms are being se-quenced more and more. This will make the primer design easier for a wide variety of species, allowing to use more markers, more accurate identification methods, and more species-specific adaptation of techniques. Lowering sequencing pricing is also slowly allowing its use to directly identify disputed samples (Summerell et al. [2019]; Kumar et al. [2019]; Kundu et al. [2019]; Ghosh et al. [2019a]; Ghosh et al. [2019b]).

Validation in wildlife forensics will also become more widely spread. As mentioned before, vali-dation needs to be made per species and per technique. As more species are being associated with wildlife crime, and more research is being done by geneticists and biologists, more techniques will be validated for their use in forensic science, and subsequently accepted in court. Research papers need to be published before using the data for forensics. This disconnects forensic science with ecology and population genetics, since one does not know beforehand what kind of data would be needed to solve a particular crime. This also should be seen as an encouragement for forensic scientists to be aware of those publications and availability of data. Population genetics research in aspects such as allelic frequencies, needs to be constantly gathered by forensic scientists at all time, making part of their job to be aware of state of the art research papers in the area. This will also allow to have at hand more realistic models to apply to the populations, considering as well the constant dynamic change they have over the years, particularly due to over exploitation and interaction with human populations (Cassidy and Gonzales [2005]; Ogden and Linacre [2015]; Summerell et al. [2019]).

Apart from scientific advances and validation, in the future more police forces and raids could be expected to find illicit wildlife transportation. An example of a recent raid from the Interpol in 2018 is Oper-ation Thunderstorm. This OperOper-ation consisted in the follow up and seizure of the main wildlife internOper-ational trade networks. This makes the Operation a good example of cooperation between countries (with 93 par-ticipants), and associations (police, border control and environmental agencies, proving that collaboration and communication between fields and countries is crucial for the finding (and fighting) wildlife crime. A summary of the results of Operation Thunderstorm can be seen in Figure 2 (INTERPOL [2018]).

An increase in the rates of wildlife crimes seems obvious when comparing the rates of the last 40 years, as depicted in Figure 3. However, this increase may be due to more discovery of the crime rather than to more crimes per se. This could mean that the growth of development and application of techniques have been more widely applied in the last years. There is no foreseeable decrease in the rates of wildlife crime, however, with more techniques and more resources, the discovery of more crimes will be an obvious result. According to the Interpol and CITES reports, wildlife crime is not expected to decrease anytime soon, however the application and enhancement of molecular techniques will help with detecting it more, and eventually reducing it (Johnson et al. [2014]; Nellemann et al. [2016]; CITES [2019]).

(13)

Figure 2: Summary of the results of Operation Thunderstorm by Interpol. Figure taken from: INTERPOL [2018].

4 Discussion

The main problem wildlife forensics encounters is the low priority it receives, when compared to hu-man forensic cases. This low priority results in low legislation, low law enforcement, and low budget. The only way to tackle this issue would be by, however unrealistic it may sound, giving higher punishments, comparable to other illicit trades as well as enforcing the law at the source of the individuals, and not only at border control. As mentioned in the introduction of this paper, wildlife crime comes with a low-risk, high-reward ratio. Heightening the risk and the punishment would change this ratio, making it less profitable. Protecting areas in situ also requires new laws, and new law enforcement officers dedicated to the protection and data collection of animals and plants. Evidence collection and casefile writing already follows the same standards and regulations as human forensics, and enforcing this equality between areas would be helpful in the investigation of crime.

Another problem in this area of Forensics is the lack of previous research on most of the species that need to be analyzed. As explained in other sections, this problem could be tackled by close collabora-tion between researchers in ecology, populacollabora-tion dynamics and statistical genetics. Statistical analysis and validation should be species-specific, as should be the research. A big limit this specificity has is that the definition of species, subspecies and populations is still something widely debated in ecological biology, and statistical genetics, enhancing the point that collaboration between the areas is crucial.

Most of the research analyzed for this paper focused on species identification, which was expected given that the mere identification of the species of origin of a sample might make a person liable of a crime. This makes sense when comparing it to the most common trade categories that were investigated: poaching and meat. For both of those trades, species identification is the only requirement for defining the crime as well as the gravity of it. Most of the current research is made in mammals, particularly big species, such as whales and elephants. Again, this result is expected, since mammals are the most common group used for poaching and meat. This result also follows the expectation of having stricter protection rules towards flag-ship species, species that also the general public might be more inclined to protect, and to donate resources and time into. A summary of these results is shown in Figure 4.

(14)

Year Total 0 50000 100000 150000 1980 1990 2000 2010

Mammal trade per year

(a) Year Reptiles 0 1000000 2000000 3000000 4000000 1980 1990 2000 2010

Reptile trade per year

(b)

Figure 3: Figure 4: Mammal and reptile trade per year. Graphs made with information from the CITES trade database (https://trade.cites.org/)

because the general public is more aware nowadays of environmental aspects, and more and more species are listed as protected each year, giving more attention to their protection. This however was not the case after reviewing the literature and the data available. Analyzing the data from CITES [2016] proved that since the 2000s animal trade had a peak in number of individuals traded. This peak continued for reptiles until 2018, but stopped growing for mammals around 2010. This can be explained by the awareness and protection mammals have, compared to reptiles, even though reptiles are the most common species traded as exotic pets. Figure 3 shows the total amounts of traded live mammals and reptiles from the beginning of the CITES (1975) until mid 2018. I believe that the increase in wildlife trade is mostly due to an increase of the discovery of the traded individuals, rather than an increase in the crime rate per se. The combination of heightened airport security, and the increase of available identification tools could the peak in wildlife trade starting in the 2000s.

Another expected result from this work, was the increase of law enforcement and regulations re-garding wildlife crime. Most of the regulations, however, are made after the damage to the wild is done. Species will not be listed as protected until the population suffers an extreme decrease. This ends up being counterproductive, since preventing this population decrease should be the main goal of the regulations, in stead of maintaining an already poor population. Another problem encountered with law making and regu-lations, is the removal of species from protection lists after their population being ”recovered”. Once again, this removal of the protection over a species, can only lead to more trade on an exotic species, leading it to extinction (as it happened with several Rhinoceros). Species should be protective as preventive measure, rather than as a way to compound the problem, and once granted, the protection towards a species should be permanent. A timeline of important events and regulations can be seen in Table 3.

(15)

Date Regulations Important events 1700-1880s Poaching is allowed in

Europe as a way of gathering food

1903 First national wildlife refuge in the US 1900s Lacey Act in the US - illegal to sell poached

animals to people in other states

1914 Extinction of the passenger pigeon 1918 Migratory Bird Act

1940 Siberian tiger population reduced 40% due to hunting in Russia

1944 Whooping bird populations drops to 21 individuals 1948 International Whaling commision

1960s CITES

1969 Crustaceans and molluscs added for protection

1970 Peregrin falcon enlisted as endangered 1972 DDT banned as a pesticide

1973 Endangered Species Act

1973 CITES signed by 80 nations Operation rhino for breeding 1976 Magnuson–Stevens Fishery Conservation and Management Act

1977 First plant species listed as endangered

1981 Redescovery of black footed ferrets (thought to be extinct) 1985 Commertial whaling halted Last 9 wild condors got captured

1987 American alligator delisted reduced elephant population from 1.3M to 600K 1987 Red wolf reintroduced to the wild

1988 Japanese scientific whale killing permit 1989 Ban on ivory trade - CITES

1989 US State laws The Interstate Wildlife Violator Compact 1990 Northern spotted own listed as threatened

1991 Black footed ferrets and condors reintroduced into the wild 1994 United Nations Convention on the LAw of the Sea Gray whales and peregrine falcon delisted

1995 US supreme court includes habitat destruction as

”harm” to the species Gray wolves are reintroduced into the wild 1998 Whales used for research can be commercially traded

2000 Shark Finning Prohibition Act

2000 Miss Waldeon’s red colobus monkey reported extinct 2003 UK increase in wild meat imports - 427kg of

animal products each week

2006 Greenpeace anti-whaling campaign Spike in elephant poaching - 120 carcasses 2006 shark fishing represents 1.8% of global fishing,

with 82M tons

2008 Tokyo district prosecution investigation whale meat valued at 550USD - from 18 months to 10 years in jail

2009 Society of Wildlife Forensic Science

2009 Agreement on Port State Measures to Prevent, Deter and Eliminate Illegal, Unreported and Unregulated Fishing 2010 IWC whaling meeting - Japan, Norway and Iceland whale

hunting ban Polar bear listed as threatened 2010 Shark Conservation Act

2011 exctinction of Java rhino in Vietnam 2013 Rhino Conservation Botswana 2000 dead pangolins seized 2015 Sumatran rhinos declared extinct 2016 Recognition of 2 species of African elephants

- both endangered

2016 Shut down of ivory market in the US

2017 Lawsuit against Trump for allowing import of elephant and lion trophies

Table 3: Timeline of legislation, and some important events in conservation. Table made with information from: Moore and Frazier [2019]; Palumbi and Cipriano [1998]; Wasser et al. [2004]; Baker et al. [2010]; UNCTAD [2018]; wwf; Kelly et al. [2003]; INTERPOL [2018]; CITES [2019] Foundation [2018]

(16)

Habitat loss 11.8% Pets 11.8% Theft 11.8% Meat 29.4% Poaching 35.3% Trade category (a) Fish 5.3% Domestic animals 5.3% Reptiles 15.8% Birds 26.3% Whales 10.5% Mammals 36.8% Groups involved (b) Kinship analysis 27.3% Source population 18.2% Individualization 18.2% Species identification 36.4% Goal (c) Universal variable 4.8% bio-banking 4.8% STR 57.1% mtDNA - Cytb 33.3% Markers (d)

Figure 4: Distribution of the analyzed research articles. (a) Illegal activities encountered. Most of the research regarding wildlife forensics is in the area of poaching. Most of the poaching is done with the purpose of selling the meat of the animals, but since meat is such a significant proportion of all the papers read, it has its own category. (b) The group with the most attention is Mammals. Since whales also receive a lot of attention of their own, again, they receive their own category. (c) Species identification, as expected, is the most common goal for forensic questions. (d) The most used markers, are STRs and mtDNA-Cytb.

5 Conclusions and recommendations

Among the problems in the field previously discussed, there are no guidelines for animal forensic test-ing, the only available ones are designated for human DNA typing and parentage analysis. Some wildlife forensic reports even conclude that their methods are completely reliable, with no reports on error rate (Bai et al. [2003]; Bollongino et al. [2003]), which only shows the lack of proper validation and error rate calcula-tions done in the field. However there is no consensus on what the appropriate quality control guidelines are, or how to follow and adapt human DNA forensic analysis. Workers that use animal samples are often less careful with them, since human DNA contamination is not usually a problem when dealing with different species, which could lead to a problem of lower quality in the final sample, and maybe hybridisation of the DNA strands with each other, complicating the PCR process. Standard Operating Procedures should be created and followed exclusively for the use in animal forensics. Extrapolating protocols from human DNA testing is not always useful, since allele selection, designation, and population dynamics are different. Collaboration between population geneticists and ecological biologists is crucial for the selection of methods and statistical calculations. Validation is also an important point, since techniques should be fit for purpose not only for the forensic presentation of the result, but also for the adaptation towards the species involved (Budowle et al. [2005]).

(17)

Even though more and more species are being regulated, there are issues with the regulation of commercialization of some of them. Hybrids, for instance, are not regulated. Hybridization between do-mestic and wild animals could get a conclusion of no illicit trade. For example, hybridization seems to occur between domesticated pigs and wild boars2_{, this could lead to misidentifying illicit boar killing, as the}

killing of a domestic pig (Iacolina et al. [2018]). This could be solved by making regulations stricter and wider, protecting wild species in general, and not only focusing on particular species. This solution, albeit unrealistic, would also solve the bias towards mammals current wildlife forensic science has. Mammals seem to receive more attention both in regulations and research, even though the most commonly traded groups are reptiles and birds (CITES [2019]).

Finally, a specialized group of scientists is needed to deal with these situations. People with the forensic thinking and reasoning for case solving, scenario creating and analysis and care for evidence, but summed up with a background on evolutionary biology and statistical genetics. This would lead to a fairer law enforcement for all species, and improvement in case management for wildlife forensics and regulations.

2_{Wild boar hunting regulations vary across Europe. In the UK, for example, it is only legal to hunt them in specific locations.}

(18)

References

Erika A Alacs, Arthur Georges, Nancy N FitzSimmons, and James Robertson. Dna detective: a review of molecular approaches to wildlife forensics. Forensic science, medicine, and pathology, 6(3):180–194, 2010. Sandeep Kumar Gupta. Application of dna fingerprinting and wildlife forensics. In DNA Fingerprinting:

Advancements and Future Endeavors, pages 77–87. Springer, 2018.

Jane E Huffman and John R Wallace. Wildlife forensics: methods and applications, volume 6. John Wiley & Sons, 2012.

CITES. Cites, 2019. URL http://www.cites.org/.

Rob Ogden and Adrian Linacre. Wildlife forensic science: a review of genetic geographic origin assignment. Forensic Science International: Genetics, 18:152–159, 2015.

Christian Nellemann, Rune Henriksen, Arnold Kreilhuber, Davyth Stewart, Maria Kotsovou, Patricia Raxter, Elizabeth Mrema, Sam Barrat, et al. The rise of environmental crime: a growing threat to natural resources, peace, development and security. United Nations Environment Programme (UNEP), 2016. Samuel K Wasser, Lola Brown, C Mailand, S Mondol, W Clark, C Laurie, and BS Weir. Genetic assignment

of large seizures of elephant ivory reveals africa’s major poaching hotspots. Science, 349(6243):84–87, 2015.

SY Bai, Y Jin, YC Xu, YK Ma, SY Jing, and XL Zhang. Species identification of the brown bear and the asian black bear. In FORENSIC SCIENCE INTERNATIONAL, volume 136, pages 375–375. ELSEVIER, 2003.

Maiken Ueland, Amber Brown, Cecilia Bartos, Greta J Frankham, Rebecca N Johnson, and Shari L Forbes. Profiling volatilomes: A novel forensic method for identification of confiscated illegal wildlife items. Sep-arations, 7(1):5, 2020.

Andreas O Karlsson and Gunilla Holmlund. Identification of mammal species using species-specific dna pyrosequencing. Forensic science international, 173(1):16–20, 2007.

K Nishant, KY Vrijesh, and KR Ajay. Wildlife forensic: current techniques and their limitations. J Forensic Sci Criminol, 5(4):402, 2017.

Vikas Kumar, Kailash Chandra, Shantanu Kundu, Kaomud Tyagi, Boni Amin Laskar, Devkant Singha, Rajasree Chakraborty, and Avas Pakrashi. Utility of mitochondrial dna in wildlife forensic science: reliable identification of confiscated materials from eastern india. Mitochondrial DNA Part B, 4(1):583–588, 2019. Shantanu Kundu, Vikas Kumar, Kaomud Tyagi, Shibananda Rath, Avas Pakrashi, Phakir Chandra Saren, Laishram Kosygin, and Kailash Chandra. Mitochondrial dna detects arctonyx collaris from burnt body parts: a wildlife forensic case study in indo-burma biodiversity hotspot. Mitochondrial DNA Part B, 4(1): 1172–1176, 2019.

Avijit Ghosh, Sambadeb Basu, Hiren Khatri, Kailash Chandra, and Mukesh Thakur. Ascertaining species of origin from confiscated meat using dna forensics. Mitochondrial DNA Part B, 4(1):329–331, 2019a. Avijit Ghosh, Shambadeb Basu, Gul Jabin, Hiren Khatri, Sujeet K Singh, Gopinathan Maheswaran, Kailash

Chandra, and Mukesh Thakur. Wildlife forensics in voiding false offences: A case study to deal with unidentified cooked meat. Forensic Science International: Reports, 1:100011, 2019b.

Gul Jabin, Avijit Ghosh, Shambadeb Basu, Hiren Khatri, Sujeet K Singh, Kailash Chandra, Lalit K Sharma, and Mukesh Thakur. Wildlife forensics in nullifying the false accusation: a case to deal with raw meat. Mitochondrial DNA Part B, 4(1):736–739, 2019.

AR Palumbi and F Cipriano. Species identification using genetic tools: the value of nuclear and mitochondrial gene sequences in whale conservation. Journal of Heredity, 89(5):459–464, 1998.

(19)

UNCTAD. 90% of fish stocks are used up – fisheries subsidies must stop, 2018. URL https://unctad.org/ en/pages/newsdetails.aspx?OriginalVersionID=1812.

C Scott Baker, Debbie Steel, Yeyong Choi, Hang Lee, Kyung Seok Kim, Sung Kyoung Choi, Yong-Un Ma, Charles Hambleton, Louie Psihoyos, RL Brownell, et al. Genetic evidence of illegal trade in protected whales links japan with the us and south korea. Biology letters, 6(5):647–650, 2010.

MS Shivji, JE Magnussen, and E Pikitch. A high-density multiplex pcr design for efficient identification of shark body parts: Application to cites/iucn species of concern. In FORENSIC SCIENCE INTERNA-TIONAL, volume 136, pages 378–379. ELSEVIER, 2003.

A Linacre, HM Hsieh, CC Tsai, LC Tsai, NE Huang, RTP Shih, and JCI Lee. Identification of the origin of meat samples sold in markets. In FORENSIC SCIENCE INTERNATIONAL, volume 136, pages 376–376. ELSEVIER, 2003.

Filipe Pereira, C´ıntia Alves, Cátia Couto, Lourdes López D´ıaz, David Parra, Sandra Furfuro, Mercedes Aler, Luis Burillo Borrego, Tereza Olekˇsáková, Filipa Balsa, et al. Species identification in routine casework samples using the spindel kit. Forensic Science International: Genetics Supplement Series, 2019.

Kimberley George, Alice Masters, and Nick Dawnay. Development of hybeacon probes for the forensicR

detection of panthera, rhinoceros, and pangolin species. Molecular and Cellular Probes, 48:101450, 2019. J Kelly, T Carter, and MD Cole. Forensic identification of non-human mammals involved in the bushmeat

trade using restriction digests of pcr amplified mitochondrial dna. In Forensic Science International, volume 136, pages 379–380. ELSEVIER, 2003.

JA Lenstra, IJ Nijman, L Van Cann, and ELC Verkaar. Identification of animal species by dna methods. In Forensic Science International, volume 136, pages 380–381. ELSEVIER, 2003.

E Conde-Sousa, N Pinto, and A Amorim. Reference dna databases for forensic species identification: Auditing algorithms. Forensic Science International: Genetics Supplement Series, 2019.

R Bollongino, J Burger, and S Hummel. Dna typing from elephant tusks and rhinoceros horn-strategies and applications in wildlife conservation and animal forensics. In Forensic Science International, volume 136, pages 374–375. ELSEVIER, 2003.

Nóra M Magonyi, Róbert Mátics, Krisztián Szabó, Péter Fehérvári, Szabolcs Solt, Péter Palatitz, and Nóra Vili. Characterization of novel microsatellite markers for the red-footed falcon (falco vespertinus) and cross-species amplification in other falco species. European Journal of Wildlife Research, 65(4):62, 2019. Michael S Blouin. Dna-based methods for pedigree reconstruction and kinship analysis in natural populations.

Trends in Ecology & Evolution, 18(10):503–511, 2003.

Samuel K Wasser, Andrew M Shedlock, Kenine Comstock, Elaine A Ostrander, Benezeth Mutayoba, and Matthew Stephens. Assigning african elephant dna to geographic region of origin: applications to the ivory trade. Proceedings of the National Academy of Sciences, 101(41):14847–14852, 2004.

Rachel L Jacobs, David C Frankel, Riley J Rice, Vera J Kiefer, and Brenda J Bradley. Parentage complexity in socially monogamous lemurs (eulemur rubriventer): Integrating genetic and observational data. American journal of primatology, 80(2):e22738, 2018.

Nicole E White, Rick Dawson, Megan L Coghlan, Silvana R Tridico, Peter R Mawson, James Haile, and Michael Bunce. Application of str markers in wildlife forensic casework involving australian black-cockatoos (calyptorhynchus spp.). Forensic Science International: Genetics, 6(5):664–670, 2012.

Nadia Mucci, Patrizia Giangregorio, Letizia Cirasella, Gloria Isani, and Chiara Mengoni. A new str panel for parentage analysis in endangered tortoises. Conservation Genetics Resources, pages 1–9, 2018. Sherryn Ciavaglia and Adrian Linacre. Ozpythonplex: An optimised forensic str multiplex assay set for the

(20)

Nadia Mucci, Chiara Mengoni, and Ettore Randi. Wildlife dna forensics against crime: Resolution of a case of tortoise theft. Forensic Science International: Genetics, 8(1):200–202, 2014.

John P Klein and Melvin L Moeschberger. Statistics for biology and health. Stat. Biol. Health, New York, 27238, 1997.

M Bielikov´a and J Turna. Genetic variability of raptors in slovakia. In Forensic Science International, volume 136, pages 374–374. ELSEVIER SCI IRELAND LTD CUSTOMER RELATIONS MANAGER, BAY 15, SHANNON . . . , 2003.

MC Ball, LA Finnegan, T Nette, HG Broders, and PJ Wilson. Wildlife forensics:“supervised” assignment testing can complicate the association of suspect cases to source populations. Forensic Science Interna-tional: Genetics, 5(1):50–56, 2011.

AE Summerell, GJ Frankham, P Gunn, and RN Johnson. Dna based method for determining source country of the short beaked echidna (tachyglossus aculeatus) in the illegal wildlife trade. Forensic science international, 295:46–53, 2019.

Sylvain Piry, Alexandre Alapetite, J-M Cornuet, David Paetkau, Luc Baudouin, and Arnaud Estoup. Geneclass2: a software for genetic assignment and first-generation migrant detection. Journal of heredity, 95(6):536–539, 2004.

Jonathan K Pritchard, William Wen, and Daniel Falush. Documentation for structure software: Version 2. 2003.

M Katherine Moore and Kim Frazier. Humans are animals, too: Critical commonalities and differences between human and wildlife forensic genetics. Journal of forensic sciences, 2019.

Marilyn A Menotti-Raymond, Victor A David, and Stephen J O’Brien. Pet cat hair implicates murder suspect. Nature, 386(6627):774, 1997.

Brandt G Cassidy and Robert A Gonzales. Dna testing in animal forensics. The Journal of wildlife man-agement, 69(4):1454–1462, 2005.

INTERPOL. Wildlife crime, 2018. URL https://www.interpol.int/Crimes/Environmental-crime/ Wildlife-crime.

Rebecca N Johnson, Linzi Wilson-Wilde, and Adrian Linacre. Current and future directions of dna in wildlife forensic science. Forensic Science International: Genetics, 10:1–11, 2014.

CITES. Cites trade dashboard, 2016. URL http://cites-dashboards.unep-wcmc.org/global. History of the wwf. URL https://www.worldwildlife.org/about/history.

International Rhino Foundation. International rhino foundation, Aug 2018. URL https://rhinos.org/. Bruce Budowle, Paolo Garofano, Andreas Hellman, Melba Ketchum, Sree Kanthaswamy, Walther Parson,

Wim Van Haeringen, Steve Fain, and Tom Broad. Recommendations for animal dna forensic and identity testing. International Journal of Legal Medicine, 119(5):295–302, 2005.

Jurgen Tack. wild boar (sus scrofa) populations in europe. A scientific review of population trends and implications for management, pages 29–30, 2018.

Laura Iacolina, Cino Pertoldi, Marcel Amills, Szilvia Kusza, Hendrik-Jan Megens, Valentin Adrian Bˆalteanu, Jana Bakan, Vlatka Cubric-Curic, Ragne Oja, Urmas Saarma, et al. Hotspots of recent hybridization between pigs and wild boars in europe. Scientific reports, 8(1):17372, 2018.

Erin P Meredith, Jillian K Adkins, and Jeff A Rodzen. Ursaplex: An str multiplex for forensic identification of north american black bear (ursus americanus). Forensic Science International: Genetics, 44:102161, 2020.

(21)

Appendix: Literature search strategy

All literature was first searched for using Google Scholar and the search query: ”wildlife forensics”. From the recovered articles, I looked into the ones that used DNA in wildlife forensics. I proceeded with more specific search queries, in order to gather articles in other areas of the explained analyses, such as ”DNA in wildlife forensics” and ”kinship analysis in wildlife forensics”.

To gather the newest, state of the art literature, I used the date filters from Google Scholar on the same search queries, to gather articles from 2019 on. After looking into those, I searched for articles from 2018 on, and from 2017 on.Once some papers were gathered, particularly reviews (such as Alacs et al. [2010]), while reading important concepts in them such as definitions or data on economy and popularity of black market wildlife trade, I would look into the papers they cited in that information. Those papers were particularly useful when looking for information on popularity and numerical data regarding the importance (and impact) of wildlife forensics. Some references gathered that way are: Nellemann et al. [2016]; CITES [2016]; Foundation [2018].

Four papers were given by my supervisor, after the discussion of the initially proposed literature list. These papers are: Wasser et al. [2004]; Wasser et al. [2015]; Baker et al. [2010] and Palumbi and Cipriano [1998].

Finally, after receiving the initial feedback on my report that I neglected to include a lot of modern molec-ular techniques, I performed a new Google Scholar search, this time with the queries ”most used molecmolec-ular techniques in wildlife forensics”, ”wildlife forensics DNA” and ”Wildlife forensics sequencing”, again with the filter from 2019 on. These papers are: Kumar et al. [2019], Nishant et al. [2017], Pereira et al. [2019], Conde-Sousa et al. [2019], Ueland et al. [2020], Ghosh et al. [2019a], Jabin et al. [2019], Kundu et al. [2019], George et al. [2019], Meredith et al. [2020], Summerell et al. [2019], Ghosh et al. [2019b].

All the papers and books gathered by me were downloaded into my computer thanks to PulseSe-cure, using my University of Amsterdam login information.

Is your pet legal? Assessment of DNA analyses used in wildlife forensics