Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Molecular gut content analysis of different spider body parts

  • Nuria Macías-Hernández ,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Supervision, Validation, Writing – original draft

    Current address: Finnish Museum of Natural History, University of Helsinki, Helsinki, Finland

    Affiliations Department of Entomology, University of Kentucky, Lexington, Kentucky, United States of America, Department of Evolutionary Biology, Ecology and Environmental Sciences, Universitat de Barcelona, Barcelona, Spain, Biodiversity Research Institute (IRBio), Universitat de Barcelona, Barcelona, Spain

  • Kacie Athey,

    Roles Formal analysis, Methodology, Writing – review & editing

    Affiliation Department of Entomology, University of Kentucky, Lexington, Kentucky, United States of America

  • Vanina Tonzo,

    Roles Data curation, Visualization

    Affiliations Department of Evolutionary Biology, Ecology and Environmental Sciences, Universitat de Barcelona, Barcelona, Spain, Biodiversity Research Institute (IRBio), Universitat de Barcelona, Barcelona, Spain

  • Owen S. Wangensteen,

    Roles Data curation, Formal analysis, Methodology, Writing – review & editing

    Affiliation Norwegian College of Fishery Science, UiT The Arctic University of Norway, Tromsø, Norway

  • Miquel Arnedo,

    Roles Conceptualization, Funding acquisition, Investigation, Project administration, Resources, Writing – review & editing

    Affiliations Department of Evolutionary Biology, Ecology and Environmental Sciences, Universitat de Barcelona, Barcelona, Spain, Biodiversity Research Institute (IRBio), Universitat de Barcelona, Barcelona, Spain

  • James D. Harwood

    Roles Conceptualization, Funding acquisition, Project administration, Writing – review & editing

    Affiliation College of Plant Health and Medicine, Qingdao Agricultural University, Qingdao, Shandong, China


Molecular gut-content analysis has revolutionized the study of food webs and feeding interactions, allowing the detection of prey DNA within the gut of many organisms. However, successful prey detection is a challenging procedure in which many factors affect every step, starting from the DNA extraction process. Spiders are liquid feeders with branched gut diverticula extending into their legs and throughout the prosoma, thus digestion takes places in different parts of the body and simple gut dissection is not possible. In this study, we investigated differences in prey detectability in DNA extracts from different parts of the spider´s body: legs, prosoma and opisthosoma, using prey-specific PCR and metabarcoding approaches. We performed feeding trials with the woodlouse hunter spider Dysdera verneaui Simon, 1883 (Dysderidae) to estimate the time at which prey DNA is detectable within the predator after feeding. Although we found that all parts of the spider body are suitable for gut-content analysis when using prey-specific PCR approach, results based on metabarcoding suggested the opisthosoma is optimal for detection of predation in spiders because it contained the highest concentration of prey DNA for longer post feeding periods. Other spiders may show different results compared to D. verneaui, but given similarities in the physiology and digestion in different families, it is reasonable to assume this to be common across species and this approach having broad utility across spiders.


The use of DNA-based methods to study food webs and feeding interactions under natural conditions has revolutionized dietary analysis in a variety of ecosystems, including marine environments [1], agroecosystems, [25], forests [6], and soils [7]. The detection of prey DNA has revealed a broad range of trophic relationships in nature (reviewed in [811], with a multitude of implications for conservation biology and management [12]. DNA-based methods facilitate prey identification in the absence of hard prey remains, as is usual for many invertebrate systems. Compared to other methodologies (e.g. morphological identification of prey remnants in the stomach or faecal samples, enzyme-linked immunosorbent assays (ELISA) techniques, or using monoclonal antibodies, among others), DNA-based prey assays can be developed faster, allow simultaneous screening for multiple prey items and offer a greater taxonomic prey resolution, although DNA prey detectability may span shorter periods of time [10, 13]. For vertebrates [1417] and some invertebrates [18, 19], molecular methods can also be applied using non-invasive methods by analysing regurgitates or faecal samples. Although many predaceous arthropods are liquid feeders or employ extra-oral digestion, usually requiring post-mortem analysis, recent studies have demonstrated the viability of using faecal analysis for prey identification in spiders [20]. However, typically gut-content analysis in invertebrates requires killing the animal and either dissecting their gut (e.g. in beetles), or selecting body parts for DNA extraction, depending on the size and identity of the focal predator [8].

A multitude of inexpensive chemicals (e.g. sodium dodecyl sulfate, SDS; or cetyltrimethyl ammonium bromide, CTAB) [21] have been adopted for gut analysis, although commercial extraction kits (e.g. Qiagen Blood and Tissue kits, Qiagen Inc, Valencia California, USA) are popular for gut content analysis because they are fast and easy to use, and more effective at amplifying gut contents [22]. However, extraction kits limit the amount of predator tissue that can be extracted. For instance, samples when using DNeasy® kits cannot exceed 25 mg in mass. While for small sized predators the whole specimen can be used for DNA extraction, large predatory arthropods may require gut dissection [6, 23] and predators with branching digestive tracts that occupy most of the body represent additional challenges when deciding what section of the body is most useful for detecting prey.

Spiders are liquid feeders utilizing a process of extra-oral digestion for prey consumption. Following ingestion of liquefied material, their midgut branches into highly complex diverticula extending throughout the prosoma and into their legs [24, 25]. Consequently, digestion takes place in many different parts of the body and dissecting the whole gut is near to impossible. When performing molecular gut-content analyses, spiders need to be either small-sized (for total body extraction), split into body parts prior to extraction or be homogenized with a subset used for DNA extraction. Most studies have focused on small-sized spider families such as the Linyphiidae [26, 27], small Lycosidae [20, 28, 29], Theridiidae, Salticidae [30], small Tetragnathidae [31, 32] and Oxyopidae [33], in which DNA extraction was made either by homogenizing the whole spider or by crushing the abdomen. Although many studies have used spiders for molecular gut-content analysis, almost none have used medium to large body size spiders (but see Schmidt et al. [2]).

To investigate the detectability of prey DNA within the gut of a medium sized spider (15–20 mm; weight: males 40–60 mg, females 50–100 mg), we use the woodlouse hunter spider Dysdera verneaui Simon, 1883 (Araneae: Dysderidae). Dysdera is a highly speciose genus distributed throughout the Mediterranean basin, with the exception of the cosmopolitan species D. crocata [34]. It has also colonised some oceanic archipelagos, such as the Canary Islands, where it has undergone a major process of local diversification (approximately 50 endemic species occur throughout this archipelago [35]). They are nocturnal wandering hunters that prefer humid and dark ground habitats and during daytime find shelter in silk retreats under rocks, trunks and tree barks. Unlike other spider genera, Dysdera shows a remarkable diversity of body sizes and mouthpart shapes [36, 37]. These differences have been related to both trophic specialisation (some species are generalists while others are woodlice feeder specialists) [38] and the prey capture strategies used to feed on woodlice [39]. The evolution of different levels of prey specialisation and preference has been identified as one of the major drivers of Dysdera diversification in the Canaries [40]. Thus, Dysdera offers an invaluable model for applying molecular gut analysis to decipher their diet in natural habitats.

Here, we combine feeding experiments with two methods of molecular prey detection, namely prey-specific PCR and metabarcoding approaches, to investigate molecular detectability of diet in medium-sized predators that exhibit extra-oral digestion. Specifically, we tested the differences on prey detectability in the spider D. verneaui using DNA extracts from different body parts (i.e., legs, prosoma and opisthosoma) and characterized the time of prey detectability after consumption. Based on these results, we provide suggestions for optimizing detection of prey DNA when using molecular gut content analysis in medium sized spiders.

Material and methods

The Cabildo of Tenerife authorized the collecting permits for the protected natural areas.

Specimen collection

Eighty specimens of D. verneaui were collected at the laurel forest (dominated by Laurus novocanariensis, Ilex canariensis, and Persea indica tree species) of Anaga in Tenerife (28.535600 N, 16.298810 W) (Canary Islands, Spain) during 2013. Spiders were hand-collected by searching under stones and logs, scraping soil and at rocky embankments. Each individual was placed into a separate 1.5 mL vials. Individuals of the potential target prey, the woodlice, Eluma caelata Miers, 1877 (Isopoda: Armadillidiidae), were collected at the same site. Additionally, non-target taxa found in the habitat were collected for construction of a DNA barcode reference library and to test for natural predation. Larvae of Tenebrio molitor Linnaeus, 1758 (Coleoptera: Tenebrionidae) reared in laboratory colonies were used as chaser prey.

Feeding trials

Feeding trials were conducted to determine the detectability half-life (after Greenstone & Hunt [41], reviewed by Greenstone [42]) of prey DNA in the gut of D. verneaui. These approaches were also used to test for differences in detectability between alternative body parts, namely legs, prosoma and opisthosoma. Predators were maintained on a 14:10 h light:dark cycle at room temperature, in individual plastic Petri dishes (55 mm) containing wet filter paper laid at the dish base. All predators were deprived of food for two weeks prior to starting the experiments. After this period, each spider was offered one specimen of the target prey (E. caelata) in the dark because D. verneaui are nocturnal wandering predators.

Spiders were observed to feed and the time of completion of the entire woodlouse was taken as reference time zero (t = 0). At this time, eight specimens were immediately transferred to individual autoclaved 1.5 mL microcentrifuge tubes containing 95% ethanol and subsequently stored at -20°C. All remaining predators were provided with larvae of T. molitor as a ‘chaser prey’ and maintained as above for 2, 4, 8, 18, 24, 48, 72, 96 or 120 h after feeding (n = 8 killed as described above at each time point). All samples were stored at -20°C until DNA extraction.

Spider dissections

To investigate prey DNA detection in different spider body parts, specimens of D. verneaui were dissected into three parts: i) all the legs (excluding metatarsus and tarsus), ii) the prosoma, and iii) the opisthosoma. Dissection was conducted using forceps and scissors, flame sterilized after each dissection to prevent cross contamination. The dissected parts were then subjected to DNA extraction separately. See Fig 1 for more details.

Fig 1. Diagram showing the experimental design and methodology used in the study.

DNA extraction, primer design, PCR amplification and sequencing

To obtain sequences for primer design, total DNA was extracted from legs of target prey, E. caelata, and non-target specimens (S1 Table) using QIAGEN DNeasy Blood and Tissue Kits (QIAGEN Inc., Valencia, California, USA) following the manufacturer’s guidelines. We targeted the animal DNA barcode, the mitochondrial gene coding Cytochrome c oxidase subunit I, hereafter referred as COI, as the molecular marker. Due to difficulties in amplifying some species, DNA was amplified using a combination of several universal COI primers: LCO22me and Hco700dy [43], Fol-degen-F and Fol-degen-R [44], or CI-J-2183 and L2-N-3014 [45] (Table 1). PCRs (50 μL) consisted of 1X Takara buffer (Takara Bio Inc., Shiga, Japan), 0.2 mM of each dNTP, 0.2 mM of each primer, 1.25 U Takara Ex Taq, Bovine serum albumin (BSA) (20mg/ml) (Promega Co., Madison, Wisconsin, USA), and template DNA (4 μL of total DNA, 1/10 diluted). PCRs were carried out in Bio-Rad PTC-200 and C1000 thermal cyclers (Bio-Rad Laboratories, Hercules, California, USA). The PCR cycling protocol was 94°C for 3 min followed by 40 cycles of 94°C for 45 s, 40°C for 45 s, 72°C for 45 s, and a final extension of 72°C for 10 min. Reaction success was determined by electrophoresis of 10 μl of PCR on a 2% SeaKem agarose gel (Lonza, Rockland, Maine, USA), pre-stained with 1X Gel Red nucleic acid gel stain (Biotium, Hayward, California, USA). DNA sequencing was undertaken at Advanced Genomics Technologies Center (University of Kentucky, Lexington, Kentucky, USA).

Table 1. List of primers used in the present study: (1–6): cox1 primers used to amplify target and non-target prey; (7–8): New specific primers designed to amplify Eluma caelata; and (9–10): Primers used to amplify the cox1 region for the metabarcoding analyses.

Resulting COI sequences were edited and assembled using Geneious 7.1.9. (Biomatters Ltd., Auckland, New Zealand) [46], aligned using MUSCLE [47] and visually inspected using BioEdit 7.0.0 (Isis Pharmaceuticals Inc., Carlsbad, California, USA) [48]. Primer conditions were checked using PRIMER 3 [49]. We designed a specific fragment to amplify a 205 bp amplicon in the target prey E. caelata: Elu-F306: GAGGGTTGGTTGAAAGTGGC and Elu2-R510: AAAGGAACTCGATCTATTTTA. The alignment used for the specific primer design contained a mixture of sequences generated in this study and sequences downloaded from GenBank (see S1 Table for accession numbers).

Gradient PCRs were performed to adjust optimal PCR conditions for the primer pair Elu-F306 and Elu2-R510. Amplification of the gut-content of D. verneaui from the feeding trials was obtained using the QIAGEN multiplex kit, which has been used successfully in previous gut content studies with spiders [50] and herbivorous insects [51]. The multiplex kit is reported to overcome the effect of PCR inhibitors in the predator [52, 53] that was most likely responsible for the non-successful amplification of the gut-content of D. verneaui using the same PCR conditions described above. Each 10 μL multiplex PCR contained 2 μL of predator DNA, 5 μL of master mix, 0.5 μL of each primer (10 μM), 1.5 μL of Q-solution, 0.2 μL of BSA, and RNase-Free water to adjust the volume. The PCR cycling conditions were as follows: 15 min at 95°C followed by 35 cycles of 94°C for 30 s, 59°C for 90 s, 72°C for 60 s, with a final extension step of 72°C for 10 min. Each PCR set contained two positive controls (E. caelata and D. verneaui fed with E. caelata at t = 0) to assess reaction success, and two negative controls (no template DNA and DNA from D. verneaui tarsus) to check for cross amplification and contamination. All predators showing PCR product of expected size were scored as positive. All samples scoring negative for prey DNA were re-assayed again to check for false negative results and, if necessary, amplified using COI general primers (Table 1) to test for extraction success. To further confirm amplification success, eight PCR products of positive amplifications were sequenced to confirm the identification of prey DNA. Primers were tested against 80 non-targets to determine specificity (S2 Table).


A metabarcoding approach was used to determine the number of prey sequences amplified from the gut of spiders at the longest time after feeding (t = 72 h) in which six specimens tested positive by PCR for E. caelata. Each of the six spiders fixed at 72 h after feeding was separated into three samples, corresponding to the three dissected body parts (legs, prosoma and opisthosoma) and analysed through metabarcoding. All the spiders had eaten both the target prey and the chaser prey (T. molitor). In order to test the ability of the universal metabarcoding primers to amplify the target prey (E. caelata), we included a positive control sample, which consisted in a mixture of 10 spiders (using different body parts) that were fed at shorter periods of time (t = 0–48 h) and tested positive in the PCR with specific primers.

DNA concentration of each extraction was determined using Qubit fluorometric quantitation (dsDNA HS Assay Kit, Thermo Fisher Scientific, Waltham, Massachusetts, USA) to adjust the final concentration of DNA in the sample to 10 μg/μL. DNA was amplified in a single-PCR step for the DNA barcode region using the universal primers mlCOIintF-XT [54] (modified from the mlCOIintF primer of [55]) and jgHCO2198 [56] producing a 313 bp amplicon. Each primer pair included an 8 bp sample tag (the same tag in the forward and reverse primers) and a tail of 2–3 random Ns in the 5' end for increasing sequence variability of the library [57]. Each 20 μL PCR reaction contained 2 μL of predator DNA (mixed DNA from 6 spider individuals in equimolar amounts), 10 μL of AmpliTaq Gold 360 Master Mix (Thermo Fisher Scientific), 1 μL of each tagged primer (5 μM), 0.16 μL of BSA, and DNase-Free water to adjust the volume. The PCR cycling protocol was 95°C for 10 min followed by 35 cycles of 94°C for 1 min, 45°C for 1 min, 72°C for 1 min, and a final extension of 72°C for 5 min. Tagged amplicons were pooled and purified using the MiniElute PCR Purification Kit (Qiagen). Illumina adapters and a library tag were added using the NEXTflex PCR-free DNA sequencing kit (Bioo Scientific, Austin, Texas, USA) and the library was sequenced in an Illumina MiSeq with a V2 2x250 bp paired-end partial run at the University of Salford, UK. The samples for this study were included in a multiplexed MiSeq run with a total of 86 samples, which explains the values for the sequencing depth of our results.

The metabarcoding pipeline was based in the OBITools suite [58]. After checking the quality of the reads with FastQC, paired-end reads were aligned using illuminapairedend and only aligned reads with quality score >40 were kept. The aligned dataset was demultiplexed and the primer sequences were removed using ngsfilter. A length filter (obigrep, 306–320 bp) was applied and reads containing ambiguous bases were removed. The reads were then dereplicated using obiuniq and the uchime-denovo algorithm implemented in vsearch [59] was used to remove chimeric sequences [60]. Molecular Operational Taxonomic Units (MOTUs) were delimited using the Bayesian clustering algorithm implemented in CROP [61] using the parameter values l = 1.5 and u = 2.5 [54]. Taxonomic assignment of the representative sequences for each MOTU was performed using the ecotag algorithm [58]. For the taxonomic assignment, we built a reference database using sequences retrieved by in silico PCR against release R117 of the EMBL-EBI database using ecoPCR [62]. New sequences obtained for our species of interest were added to this reference database. This combined reference database is publicly available from Mendeley Data [63]. The final refining of the dataset included taxonomic clustering of MOTUs assigned to the same species and abundance renormalization for removing false positive results [57]. This abundance renormalization procedure is based on calculating the cumulative frequencies of relative abundances of each MOTUs in every multiplexed sample, and equalling to zero the values of those samples whose cumulative frequency is < 1%. This procedure was applied to the whole multiplexed library of samples that were analysed together in the MiSeq run, and it is necessary for removing false positives resulting from random tag switching [64]. After this step, we checked that this correction did not affect the results of abundances for both studied prey items (in the sense of removing any true positive result of both prey items from the body part samples).

Statistical analyses

The rate of decay of prey DNA (E. caelata) within the guts of spiders was calculated using the positive scores of PCR amplification with specific target-prey primers of each spider´s body part separately. Detectability half-lives of each body part tested (legs, prosoma and opisthosoma) was calculated using probit analysis, and Chi-square (X2) test were used to determine how well a probit model fit the data. The results of the three body parts were compared using the recommended 83% fiducial confidence limits [65]. All analyses were performed in SAS 9.4, (SAS Institute, Cary, North Carolina, USA).


Spider´s body part detection and DNA decay rates

Sequences of E. caelata matched those on GenBank with 100% identity. The primers designed in the present study were specific to the target prey with no cross reactivity with 80 non-target taxa (S2 Table).

Results of target prey amplification by PCR for each body part (legs, prosoma and opisthosoma) is shown in Table 2. Probit analyses of the feeding trials showed that median detection time (MDT) of the three body parts tested overlap in 83 h (see Table 3). Decay rate curves of E. caelata DNA within the guts of predators of the three body parts is shown in S1 Fig.

Table 2. Detection of target prey when using prey-specific PCR approach for each body part tested (legs, prosoma and opisthosoma).

Table 3. DNA detectability half-life of the different body parts tested calculated with probit models, and 83% fiducial confidence limits calculated to compared half-lives.

Metabarcoding results

The Illumina sequencing generated 339,385 reads after quality control for the three samples analysed (legs, prosoma and opisthosoma of D. verneaui). The number of unique sequences in these samples was 75,265, which were clustered into 443 different MOTUs by Bayesian clustering and collapsed into a final dataset of 144 MOTUs by taxonomic clustering. After abundance renormalization, this final dataset included 338,559 reads. On the other hand, the positive control sample generated 285,729 final reads. A graphical summary of read abundance of items detected in the samples is shown in Fig 2.

Fig 2. Abundance of reads of detected MOTUs in different body parts of D. verneaui 72 hours after feeding and in a positive control (a mix of 10 spiders recently fed).

The total number of reads obtained for each sample (after quality control) is shown above the bar plots. Relative abundance of reads (above) and relative abundance of log-transformed reads (below) are shown, for a better visualization of MOTUs detected in low abundances.

In all samples, most reads belonged to D. verneaui DNA (90.85% of the three samples and 90.3% of the positive control), followed by 8.97% of the reads (8.85% of the control) that were assigned (with 94.9% sequence identity) to a Wolbachia endosymbiont specifically localized in the opisthosoma and positive control samples (after abundance renormalization for removing false positives). In the control sample a total of 387 reads (0.13%) were assigned to the target prey E. caelata. In the three analysed body parts (legs, prosoma and opisthosoma) a total of 411 reads (0.12%) were assigned to T. molitor (5 reads in the prosoma sample and 406 reads in the opisthosoma), whereas E. caelata was not detected in any of these samples.


This study compared the results of a traditional PCR approach with the metabarcoding technology to identify optimal body part locations of a spider that were most suitable for amplifying partially digested prey DNA. Using a PCR approach with specific primers for a target prey, we demonstrated that all parts of the spider body (legs, prosoma and opisthosoma) are equally suitable to detect consumed prey, regardless of the different after-feeding periods tested. Traditional PCR is not quantitative and hence is unable to measure the amount of DNA contained on each sample. Alternatively, more refined techniques such as qPCR [6669] or post-PCR visualization using a capillary electrophoresis system (e.g. [29, 67, 70]) could be utilized to achieve a more precise quantification of which body part contains more target prey DNA.

Instead, here we used a metabarcoding approach as a quantitative method. Generally, in biodiversity assessment or dietary studies, metabarcoding is not considered a quantitative method because prey amplification with universal primers might be biased toward certain taxonomic groups and many factors such as annealing temperature and PCR cycles may affect the amplification efficiency and final success [7173]. Although with some caution because nothing is known about the digestion rate in the different spider body compartments, we propose that in this study, the number of reads obtained in the metabarcoding analyses can be used to quantify the relative abundance of prey remains in the different body parts of the same specimen, because the putative biases in the PCR can apply similarly to the three samples analysed.

Comparison of the metabarcoding analyses revealed differences in the number of prey reads (of T. molitor) detected in the different body parts tested, with the opisthosoma being the tissue containing the highest number of reads (406) and hence the best alternative for amplifying gut content after long post feeding periods. These results are in contrast to the results from PCR using specific primers, which suggested no differences among the spider body parts, but in agreement with other studies also using metabarcoding [74].

Prey DNA in the gut of a predator is highly degraded and at lower concentration than predator DNA [75] making it more difficult to amplify. Another complication is that the high concentration of predator DNA in the whole-body extracts may inhibit prey detectability [53, 76]. Although there are several methods to enhance PCR amplification and reduce the action of PCR inhibitors, such as adding an amplification facilitator (e.g. bovine serum albumin) [52] or purifying DNA extracts before amplification [29, 53], it is always advisable to reduce the amount of predator tissue in the DNA extraction process [9]. Extraction of large spiders using the whole specimen with later homogenization and then using a small portion of the DNA extract may not be advisable because of the overabundance of predator DNA in the sample. The amount of spider tissue can be reduced by selecting one portion of the spider body (legs, prosoma or opisthosoma) and removing body parts that do not include digestive tissue (chelicerae or final segments of legs). A method allowing the enrichment of prey DNA from extractions of predators has been recently described by Krehenwinkel et al. [74], in which prey DNA is separated from predator DNA by size selection (high molecular weight correspond to predator DNA while shorter fragments correspond to degraded prey DNA) using a purification step with AMPure XP beads. This latter method is a promising technique to enrich prey DNA by removing predator DNA in gut content analyses and it might have broad utility for use in spiders where it is not feasible to dissect the gut.

Although suboptimal, in some situations the use of legs of medium to large sized spiders, instead of the opisthosoma, may be advisable. Recent studies have revealed that it is possible to detect prey DNA in spiders using non invasive methods such as analysing faecal samples [20] or extracting prey DNA from spiders web [77]. It has also been shown that leg autotomy does not compromise the survival of adult spiders (at least in Mygalomorphae) [78] and there are few apparent fitness costs [79]. In juvenile spiders, lost appendages are even replaced after moulting [25, 80]. Therefore, the utilization of legs (better used for detecting predation with specific primers, or at short time periods after feeding when using metabarcoding) instead of opisthosoma or carapaces, provide an additional non-lethal prey DNA source that may complement the use of other less invasive techniques. These could be most relevant for studying rare or protected spiders. Additionally, the use of legs may be preferred in cases where it is important to minimise damage to the voucher (e.g. unique specimens) or preserve relevant taxonomic characters such as female genitalia or spinnerets.

Considering the metabarcoding analyses, prey DNA recovery in the three samples tested varied between E. caelata, with zero reads generated, and T. molitor, with 411 reads generated. The difference in time from the ingestion of prey (E. caelata 72 h, T. molitor 65 h, chaser prey consumed after target prey) could be one explanation for the unsuccessful prey recovery of E. caelata, since longer periods after feeding reduce the detectability of prey DNA [52]. An alternative explanation would be a significant primer bias when amplifying different prey, which would severely affect detectability, favouring the amplification of Coleoptera over Isopoda, as has been reported in previous metabarcoding studies with similar cox1 primers [74]. On the other hand, in the control sample that combined spiders at shorter times after feeding (t = 0–48 h) (higher amount of prey DNA in the gut), E. caelata was readily amplified with the universal primers used in this study. Hence a combination of factors (time after feeding and possible primer bias) might be affecting prey detectability.

Another complication inherent to metabarcoding analyses is the probability of having false positives, in which prey that were not consumed by the predator are detected due to contaminations, amplification or sequencing errors, including tag switching [81, 82]. In our study the presence of false positives seems unlikely, because the spiders used were feed with known taxa, although some precautions should be taken in standard environmental DNA metabarcoding (eDNA) or molecular prey detection studies to avoid contaminations during the different steps of sampling, lab protocols and bioinformatics analyses, especially if samples from other studies are multiplexed within the same sequencing run (see Ficetola et al. [83] for more details).

In our study, the time after feeding at which half predators tested positive for prey DNA using PCR with specific primers approach was 83 h, which is within the typical range observed for spiders (see Greenstone [42]). Studies conducted on spiders have revealed a broad range of detectability half-lives, from less than 10 h (e.g. Lycosidae fed with one aphid [84] or one springtail [85], to more than 100 h (Tetragnathidae fed with one mirid [86]) (see Greenstone [42] for further details).

Although focused in one particular species, our findings might be applicable cautiously, to species other than Dysdera since it has been shown that prey DNA detection success is similar within taxonomically related species [87], but also to similar body sized spiders. It has to be considered that prey detectability half-lives in molecular gut content studies might be affected by the predator/prey combination, thus different prey items for the same predator could have different detection half-lives [88, 89]. It is important to mention that considering all the many factors influencing detection half-lives, each study should adapt conditions to the studied organism and system.

Although there are an increasing number of studies that use metabarcoding and metagenomics to study diet [74, 9093], only two, Paula et al. [91] and Srivathsan et al. [90], have estimated the range of prey decay, and thus there is a lack of studies that compare the relative efficiency of specific primer approaches versus high throughput sequencing approaches, so more studies in this topic are needed. Paula et al. [91], using PCR-free direct shotgun sequencing, found similar decay rates to previous studies using PCR based methods and in some cases, metagenomic sequencing appears to enable prey detection for longer.

Although in our study we have demonstrated that all parts (legs, prosoma and opisthosoma) of the medium-sized spider, D. verneaui, are suitable for prey DNA detection when using PCR with specific primers, we further recommend using the opisthosoma to amplify gut content after long periods after feeding, in agreement with previous studies [74]. We also suggest using metabarcoding with caution in these studies due to differential amplification of DNA from different groups. Employing metabarcoding approaches with verification studies using specific primers may clarify these feeding interactions.

Supporting information

S1 Table. List of target and non-target prey used for specific primer design, indicating the cox1 primers used to amplify them (see also Table 1).

GenBank accession numbers of sequences downloaded (-), plus the new sequences generated in this study.


S2 Table. List of non-target prey tested against the specific primer designed for E. caelata.

Primers used to amplify cox1 to test for extraction success.


S1 Fig. DNA decay rate curves in the three body part tested in D. verneaui.

Lines are fitted probit models with 83% fiducial confidence limits (dashed lines).



We are indebted to Carla Hernández and Salvador de la Cruz for helping during fieldwork, to A.J. Pérez and D. Suárez for helping identifying species, and to Sabrina Clemente for suggestions in some statistical analyses.


  1. 1. Leal MC, Ferrier-Pagès C, Calado R, Thompson ME, Frischer ME, Nejstgaard JC. Coral feeding on microalgae assessed with molecular trophic markers. Mol Ecol. 2014; 23(15): 3870–6. pmid:24112432
  2. 2. Schmidt JM, Barney SK, Williams MA, Bessin RT, Coolong TW, Harwood JD. Predator-prey trophic relationships in response to organic management practices. Mol Ecol. 2014; 23(15): 3777–89. pmid:24673741
  3. 3. Welch KD, Schofield MR, Chapman EG, Harwood JD. Comparing rates of springtail predation by web-building spiders using Bayesian inference. Mol Ecol. 2014; 23(15): 3814–25. pmid:24635414
  4. 4. Traugott M, Bell J, Raso L, Sint D, Symondson WOC. Generalist predators disrupt parasitoid aphid control by direct and coincidental intraguild predation. Bull Entomol Res. 2012; 102(02): 239–47. pmid:22008233
  5. 5. King R, Symondson W, Thomas R. Molecular analysis of faecal samples from birds to identify potential crop pests and useful biocontrol agents in natural areas. Bull Entomol Res. 2015; 105(3): 261. pmid:25572526
  6. 6. Jelaska ŠL, Jurasović J, Brown DS, Vaughan IP, Symondson WO. Molecular field analysis of trophic relationships in soil-dwelling invertebrates to identify mercury, lead and cadmium transmission through forest ecosystems. Mol Ecol. 2014; 23(15): 3755–66. pmid:24138157
  7. 7. Lundgren JG, Fergen JK. Predator community structure and trophic linkage strength to a focal prey. Mol Ecol. 2014; 23(15): 3790–8. pmid:24612337
  8. 8. Symondson WOC. Molecular identification of prey in predator diets. Mol Ecol. 2002; 11(4): 627–41. pmid:11972753
  9. 9. King RA, Read DS, Traugott M, Symondson WOC. Molecular analysis of predation: a review of best practice for DNA-based approaches. Mol Ecol. 2008; 17(4): 947–63. pmid:18208490
  10. 10. Pompanon F, Deagle BE, Symondson WOC, Brown DS, Jarman SN, Taberlet P. Who is eating what: diet assessment using next generation sequencing. Mol Ecol. 2012; 21(8): 1931–50. pmid:22171763
  11. 11. Sheppard SK, Harwood JD. Advances in molecular ecology: tracking trophic links through predator–prey food-webs. Funct Ecol. 2005; 19(5): 751–62.
  12. 12. Clare EL. Molecular detection of trophic interactions: emerging trends, distinct advantages, significant considerations and conservation applications. Evol Appl. 2014; 7(9): 1144–57. pmid:25553074
  13. 13. Dunshea G. DNA-based diet analysis for any predator. PLoS ONE. 2009; 4(4): e5252. pmid:19390570
  14. 14. Beja-Pereira A, Oliveira R, Alves PC, Schwartz MK, Luikart G. Advancing ecological understandings through technological transformations in noninvasive genetics. Mol Ecol Resour. 2009; 9(5): 1279–301. pmid:21564900
  15. 15. Zeale MR, Butlin RK, Barker GL, Lees DC, Jones G. Taxon-specific PCR for DNA barcoding arthropod prey in bat faeces. Mol Ecol Resour. 2011; 11(2): 236–44. pmid:21429129
  16. 16. Brown DS, Jarman SN, Symondson WOC. Pyrosequencing of prey DNA in reptile faeces: analysis of earthworm consumption by slow worms. Mol Ecol Resour. 2012; 12(2): 259–66. pmid:22176947
  17. 17. Boyer S, Cruickshank RH, Wratten SD. Faeces of generalist predators as "biodiversity capsules": a new tool for biodiversity assessment in remote and inaccessible habitats. Mitochondrial DNA. 2015; 22(1): 3–9.
  18. 18. Seeber J, Rief A, Seeber GUH, Meyer E, Traugott M. Molecular identification of detritivorous soil invertebrates from their faecal pellets. Soil Biol Biochem. 2010; 42(8): 1263–7.
  19. 19. Waldner T, Traugott M. DNA-based analysis of regurgitates: a noninvasive approach to examine the diet of invertebrate consumers. Mol Ecol Resour. 2012; 12(4): 669–75. pmid:22443278
  20. 20. Sint D, Thurner I, Kaufmann R, Traugott M. Sparing spiders: faeces as a non-invasive source of DNA. Front. Zool. 2015; 12(1): 3.
  21. 21. Chen H, Rangasamy M, Tan SY, Wang H, Siegfried BD. Evaluation of five methods for total DNA extraction from western corn rootworm beetles. PLoS ONE. 2010; 5(8): e11963. pmid:20730102
  22. 22. Simonelli P, Troedsson C, Nejstgaard JC, Zech K, Larsen JB, Frischer ME. Evaluation of DNA extraction and handling procedures for PCR-based copepod feeding studies. J Plankton Res. 2009; 31(12): 1465–74.
  23. 23. Hatteland BA, Symondson WOC, King RA, Skage M, Schander C, Solhøy T. Molecular analysis of predation by carabid beetles (Carabidae) on the invasive Iberian slug Arion lusitanicus. Bull Entomol Res. 2011; 101(06): 675–86. pmid:21342604
  24. 24. Cohen AC. Extra-oral digestion in predaceous terrestrial Arthropoda. Annu Rev Entomol. 1995; 40(1): 85–103.
  25. 25. Foelix RF. Biology of Spiders: New York, Oxford University Press; 1996.
  26. 26. Agustí N, Shayler SP, Harwood JD, Vaughan IP, Sunderland KD, Symondson WOC. Collembola as alternative prey sustaining spiders in arable ecosystems: prey detection within predators using molecular markers. Mol Ecol. 2003; 12: 3467–75. pmid:14629361
  27. 27. Welch KD, Whitney TD, Harwood JD. Non-pest prey do not disrupt aphid predation by a web-building spider. Bull Entomol Res. 2016; 106(1): 91–8. pmid:26584533
  28. 28. Monzó C, Sabater-Muñoz B, Urbaneja A, Castañera P. Tracking medfly predation by the wolf spider, Pardosa cribata Simon, in citrus orchards using PCR-based gut-content analysis. Bull Entomol Res. 2010; 100(02): 145–52. pmid:19500431
  29. 29. Sint D, Raso L, Kaufmann R, Traugott M. Optimizing methods for PCR-based analysis of predation. Mol Ecol Resour. 2011; 11(5): 795–801. pmid:21507208
  30. 30. Furlong MJ, Rowley DL, Murtiningsih R, Greenstone MH. Combining ecological methods and molecular gut-content analysis to investigate predation of a lepidopteran pest complex of Brassica crops. Entomol Exp Appl. 2014; 153(2): 128–41.
  31. 31. Chapman EG, Romero SA, Harwood JD. Maximizing collection and minimizing risk: does vacuum suction sampling increase the likelihood for misinterpretation of food web connections? Mol Ecol Resour. 2010; 10(6): 1023–33. pmid:21565112
  32. 32. Kerzicnik LM, Chapman EG, Harwood JD, Peairs FB, Cushing PE. Molecular characterization of Russian wheat aphid consumption by spiders in winter wheat. J Arachnol. 2012; 40(1): 71–7.
  33. 33. Greenstone MH, Tillman PG, Hu JS. Predation of the newly invasive pest Megacopta cribraria (Hemiptera: Plataspidae) in soybean habitats adjacent to cotton by a complex of predators. J Econ Entomol. 2014; 107(3): 947–54. pmid:25026652
  34. 34. World Spider Catalog. Natural History Museum Bern, online at, version 19.0, accessed on 06.03.2018
  35. 35. Macías-Hernández N, de la Cruz López S, Roca-Cusachs M, Oromí P, Arnedo MA. A geographical distribution database of the genus Dysdera in the Canary Islands (Araneae, Dysderidae). ZooKeys. 2016; 625: 11–23. pmid:27833424
  36. 36. Deeleman-Reinhold CL, Deeleman PR. Révision des Dysderinae. Tijdschr Entomol. 1988; 131: 141–269.
  37. 37. Arnedo MA, Oromí P, Ribera C. Radiation of the spider genus Dysdera (Araneae, Dysderidae) in the Canary Islands: Cladistic assessment based on multiple data sets. Cladistics. 2001; 17: 313–53.
  38. 38. Toft S, Macías-Hernández N. Metabolic adaptations for isopod specialization in three species of Dysdera spiders from the Canary Islands. Physiol Entomol. 2017; 42(2):191–198.
  39. 39. Řezáč M, Pekár S, Lubin Y. How oniscophagous spiders overcome woodlouse armour. J Zool. 2008; 275(1): 64–71.
  40. 40. Arnedo MA, Oromí P, Múrria C, Macías-Hernández N, Ribera C. The dark side of an island radiation: systematics and evolution of troglobitic spiders of the genus Dysdera Latreille (Araneae: Dysderidae) in the Canary Islands. Invertebr Syst. 2007; 21: 623–60.
  41. 41. Greenstone M, Hunt J. Determination of prey antigen half-life in Polistes metricus using a monoclonal antibody-based immunodot assay. Entomol Exp Appl. 1993; 68(1): 1–7.
  42. 42. Greenstone MH, Payton ME, Weber DC, Simmons AM. The detectability half-life in arthropod predator–prey research: what it is, why we need it, how to measure it, and how to use it. Mol Ecol. 2014; 23(15): 3799–813. pmid:24303920
  43. 43. Walker JM, Curole JP, Wade DE, Chapman EG, Bogan AE, Watters GT, et al. Taxonomic distribution and phylogenetic utility of gender-associated mitochondrial genomes in the Unionoida (Bivalvia). Malacologia-Philadelphia. 2006; 48(1/2): 265.
  44. 44. Yu DW, Ji Y, Emerson BC, Wang X, Ye C, Yang C, et al. Biodiversity soup: metabarcoding of arthropods for rapid biodiversity assessment and biomonitoring. Methods Ecol Evol. 2012; 3(4): 613–23.
  45. 45. Simon C, Frati F, Beckenbach A, Crespi B, Liu H, Flook P. Evolution, weighting, and phylogenetic utility of mitochondrial gene sequences and a compilation of conserved polymerase chain reaction primers. Ann Entomol Soc Am. 1994; 87(6): 651–701.
  46. 46. Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, et al. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 2012; 28(12): 1647–9. pmid:22543367
  47. 47. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucl Acids Res. 2004; 32(5): 1792–7. pmid:15034147
  48. 48. Hall T, Biosciences I, Carlsbad C. BioEdit: an important software for molecular biology. GERF Bull Biosci. 2011; 2(1): 60–1.
  49. 49. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC, Remm M, et al. Primer3-new capabilities and interfaces. Nucl Acids Res. 2012; 40(15): e115–e. pmid:22730293
  50. 50. Virant-Doberlet M, King RA, Polajnar J, Symondson WOC. Molecular diagnostics reveal spiders that exploit prey vibrational signals used in sexual communication. Mol Ecol. 2011; 20(10): 2204–16. pmid:21352388
  51. 51. Pumarino L, Alomar O, Agustí N. Development of specific ITS markers for plant DNA identification within herbivorous insects. Bull Entomol Res. 2011; 101(03): 271–6.
  52. 52. Juen A, Traugott M. Amplification facilitators and multiplex PCR: Tools to overcome PCR-inhibition in DNA-gut-content analysis of soil-living invertebrates. Soil Biol Biochem. 2006; 38(7): 1872–9.
  53. 53. Raso L, Sint D, Mayer R, Plangg S, Recheis T, Brunner S, et al. Intraguild predation in pioneer predator communities of alpine glacier forelands. Mol Ecol. 2014; 23(15): 3744–54. pmid:24383765
  54. 54. Wangensteen OS, Palacín C, Guardiola M, Turon X. DNA metabarcoding of littoral hard-bottom communities: high diversity and database gaps revealed by two molecular markers. PeerJ. 2018; 6, e4705.
  55. 55. Leray M, Yang JY, Meyer CP, Mills SC, Agudelo N, Ranwez V, et al. A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: application for characterizing coral reef fish gut contents. Front. Zool. 2013; 10(1): 34.
  56. 56. Geller J, Meyer C, Parker M, Hawk H. Redesign of PCR primers for mitochondrial cytochrome c oxidase subunit I for marine invertebrates and application in all-taxa biotic surveys. Mol Ecol Resour. 2013; 13(5): 851–61. pmid:23848937
  57. 57. Wangensteen OS, Turon X. Metabarcoding techniques for assessing biodiversity of marine animal forests. In: Rossi S. B L, Gori A., Orejas C. (Eds.), editor. Marine Animal Forests The ecology of benthic biodiversity hotspots. Switzerland: Springer International Publishing; 2017. pp. 445–473. (Print ISBN: 978-3-319-21011-7).
  58. 58. Boyer F, Mercier C, Bonin A, Le Bras Y, Taberlet P, Coissac E. Obitools: a unix-inspired software package for DNA metabarcoding. Mol Ecol Resour. 2016; 16(1): 176–82. pmid:25959493
  59. 59. Rognes T, Flouri T, Nichols B, Quince C, Mahé F. VSEARCH: a versatile open source tool for metagenomics. PeerJ. 2016; 4: e2584. pmid:27781170
  60. 60. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011; 27(16): 2194–200. pmid:21700674
  61. 61. Hao X, Jiang R, Chen T. Clustering 16S rRNA for OTU prediction: a method of unsupervised Bayesian clustering. Bioinformatics. 2011; 27(5): 611–8. pmid:21233169
  62. 62. Ficetola GF, Coissac E, Zundel S, Riaz T, Shehzad W, Bessière J, et al. An in silico approach for the evaluation of DNA barcodes. BMC Genomics. 2010; 11(1): 434.
  63. 63. Reference database for COI metabarcoding of gut contents of Dysdera verneaui. Mendeley Data, v2 2017.
  64. 64. Carlsen T, Aas AB, Lindner D, Vrålstad T, Schumacher T, Kauserud H. Don't make a mista (g) ke: is tag switching an overlooked source of error in amplicon pyrosequencing studies? Fungal Ecol. 2012; 5(6): 747–9.
  65. 65. Payton ME, Greenstone MH, Schenker N. Overlapping confidence intervals or standard error intervals: what do they mean in terms of statistical significance? J. Insect Sci. 2003; 3(1): 34.
  66. 66. Weber DC, Lundgren JG. Detection of predation using qPCR: effect of prey quantity, elapsed time, chaser diet, and sample preservation on detectable quantity of prey DNA. J. Insect Sci. 2009; 9(41): 1–12. pmid:19619033
  67. 67. Eitzinger B, Unger EM, Traugott M, Scheu S. Effects of prey quality and predator body size on prey DNA detection success in a centipede predator. Mol Ecol. 2014; 23(15): 3767–76. pmid:24383982
  68. 68. Leal MC, Nejstgaard JC, Calado R, Thompson ME, Frischer ME. Molecular assessment of heterotrophy and prey digestion in zooxanthellate cnidarians. Mol Ecol. 2014; 23(15): 3838–48. pmid:24118448
  69. 69. Redd KS, Ling SD, Frusher SD, Jarman S, Johnson CR. Using molecular prey detection to quantify rock lobster predation on barrens-forming sea urchins. Mol Ecol. 2014; 23(15): 3849–69. pmid:24844936
  70. 70. Wallinger C, Staudacher K, Schallhart N, Peter E, Dresch P, Juen A, et al. The effect of plant identity and the level of plant decay on molecular gut content analysis in a herbivorous soil insect. Mol Ecol Resour. 2013; 13(1): 75–83. pmid:23167731
  71. 71. Piñol J, Mir G, Gomez-Polo P, Agustí N. Universal and blocking primer mismatches limit the use of high-throughput DNA sequencing for the quantitative metabarcoding of arthropods. Mol Ecol Resour. 2014; 15(4): 819–30. pmid:25454249
  72. 72. Shehzad W, Riaz T, Nawaz MA, Miquel C, Poillot C, Shah SA, et al. Carnivore diet analysis based on next-generation sequencing: application to the leopard cat (Prionailurus bengalensis) in Pakistan. Mol Ecol. 2012; 21(8): 1951–65. pmid:22250784
  73. 73. Clarke LJ, Soubrier J, Weyrich LS, Cooper A. Environmental metabarcodes for insects: in silico PCR reveals potential for taxonomic bias. Mol Ecol Resour. 2014; 14(6): 1160–70. pmid:24751203
  74. 74. Krehenwinkel H, Kennedy S, Pekár S, Gillespie RG. A cost-efficient and simple protocol to enrich prey DNA from extractions of predatory arthropods for large-scale gut content analysis by Illumina sequencing. Methods Ecol Evol. 2016; 8: 126–34.
  75. 75. Deagle B, Eveson JP, Jarman S. Quantification of damage in DNA recovered from highly degraded samples—a case study on DNA in faeces. Front. Zool. 2006; 3(1): 11.
  76. 76. Piñol J, San Andrés V, Clare EL, Mir G, Symondson WOC. A pragmatic approach to the analysis of diets of generalist predators: the use of next-generation sequencing with no blocking probes. Mol Ecol Resour. 2014; 14(1): 18–26. pmid:23957910
  77. 77. Xu CCY, Yen IJ, Bowman D, Turner CR. Spider web DNA: a new spin on noninvasive genetics of predator and prey. PLoS ONE. 2015; 10(11): e0142503. pmid:26606730
  78. 78. Longhorn SJ, Nicholas M, Chuter J, Vogler AP. The utility of molecular markers from non-lethal DNA samples of the CITES II protected "tarantula" Brachypelma vagans (Araneae, Theraphosidae). J Arachnol. 2007; 35(2): 278–92.
  79. 79. Brueseke MA, Rypstra AL, Walker SE, Persons MH. Leg autotomy in the wolf spider Pardosa milvina: a common phenomenon with few apparent costs. Am. Midl. Nat. 2001; 146(1): 153–60.
  80. 80. Wrinn KM, Uetz GW. Impacts of leg loss and regeneration on body condition, growth, and development time in the wolf spider Schizocosa ocreata. Can J Zool-Rev Can Zool. 2007; 85(7): 823–31.
  81. 81. Carpenter JM. Random cladistics. Cladistics. 1992; 8: 147–53.
  82. 82. Ficetola GF, Pansu J, Bonin A, Coissac E, Giguet-Covex C, De Barba M, et al. Replication levels, false presences and the estimation of the presence/absence from eDNA metabarcoding data. Mol Ecol Resour. 2015; 15(3): 543–56. pmid:25327646
  83. 83. Ficetola GF, Taberlet P, Coissac E. How to limit false positives in environmental DNA and metabarcoding? Mol Ecol Resour. 2016; 16(3): 604–7. pmid:27062589
  84. 84. Kuusk A-K, Cassel-Lundhagen A, Kvarnheden A, Ekbom B. Tracking aphid predation by lycosid spiders in spring-sown cereals using PCR-based gut-content analysis. Basic Appl Ecol. 2008; 9(6): 718–25.
  85. 85. Kuusk A-K, Ekbom B. Lycosid spiders and alternative food: feeding behavior and implications for biological control. Biol Control. 2010; 55(1): 20–6.
  86. 86. Kobayashi T, Takada M, Takagi S, Yoshioka A, Washitani I. Spider predation on a mirid pest in Japanese rice fields. Basic Appl Ecol. 2011; 12(6): 532–9.
  87. 87. Waldner T, Sint D, Juen A, Traugott M. The effect of predator identity on post-feeding prey DNA detection success in soil-dwelling macro-invertebrates. Soil Biol Biochem. 2013; 63: 116–23.
  88. 88. Greenstone MH, Szendrei Z, Payton ME, Rowley DL, Coudron TC, Weber DC. Choosing natural enemies for conservation biological control: use of the prey detectability half-life to rank key predators of Colorado potato beetle. Entomol Exp Appl. 2010; 136(1): 97–107.
  89. 89. Gagnon AÈ, Doyon J, Heimpel GE, Brodeur J. Prey DNA detection success following digestion by intraguild predators: influence of prey and predator species. Mol Ecol Resour. 2011; 11(6): 1022–32. pmid:21749673
  90. 90. Paula DP, Linard B, Crampton-Platt A, Srivathsan A, Timmermans MJ, Sujii ER, et al. Uncovering trophic interactions in arthropod predators through DNA shotgun-sequencing of gut contents. PloS ONE. 2016; 11(9): e0161841. pmid:27622637
  91. 91. Srivathsan A, Ang A, Vogler AP, Meier R. Fecal metagenomics for the simultaneous assessment of diet, parasites, and population genetics of an understudied primate. Front. Zool. 2016; 13(1): 1.
  92. 92. Srivathsan A, Sha JCM, Vogler AP, Meier R. Comparing the effectiveness of metagenomics and metabarcoding for diet analysis of a leaf-feeding monkey (Pygathrix nemaeus). Mol Ecol Resour. 2015; 15(2): 250–61. pmid:25042073
  93. 93. Paula DP, Linard B, Andow DA, Sujii ER, Pires CSS, Vogler AP. Detection and decay rates of prey and prey symbionts in the gut of a predator through metagenomics. Mol Ecol Resour. 2015; 15(4): 880–92. pmid:25545417