A comparative analysis of hybridisation capture and PCR-based eDNA metabarcoding for monitoring bryophytes in riparian ecosystems

449 A comparative analysis of hybridisation capture and PCRbased eDNA metabarcoding for monitoring bryophytes in riparian ecosystems Thomas Reinhart1, Armando Espinosa Prieto1, Thomas Begoc1, Hugues Tinguy2, Francis Bick2, Etienne Chanez1, Jean-Nicolas Beisel1, Laurent Hardion1 1 University of Strasbourg, CNRS, ENGEES, LIVE UMR 7362, 67000 Strasbourg, France 2 Société Botanique d’Alsace, Sélestat, France Corresponding authors: Thomas Reinhart ([email protected]); Armando Espinosa Prieto ([email protected]) Copyright: © Thomas Reinhart et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International – CC BY 4.0). Research Article Abstract Despite their ecological importance, mosses remain under-represented in ecological studies due to their challenging detection in field surveys and morphological identification, exacerbated by the lack of expert botanists. In this study, we optimise an environmental DNA method for the detection of bryophytes from river water samples, with the aim of facilitating their inclusion in biodiversity assessments. We compared three different methods in terms of species detection and community dissimilarity at seven sites along a river. The methods include (i) visual transect surveys conducted by bryologists based on macroand micro-morphology, (ii) multi-marker PCR metabarcoding of the rbcL and the ITS2 markers with newly designed primers targeting bryophytes, and (iii) hybridisation capture (HC) for the same markers. We found that PCR metabarcoding recovered more than 50% (n = 37) of the species observed in the field, while hybridization capture detected only 16% (n = 11). PCR metabarcoding identified the most species, 101 species compared to 68 observed in the field and 27 with HC. Both the PCR and HC metabarcoding approaches identified bryophyte species not recorded in field surveys but expected in the catchment. Molecular methods, particularly PCR metabarcoding, recovered rare and elusive species difficult to observe in the field and occurring outside our transect. The two markers used in the molecular approaches contributed uniquely to species detection, making a multi-marker approach necessary to study this group. Environmental DNA and field surveys represent integrative methods that collectively enhance detection of inconspicuous species and yield the most comprehensive species inventory. Key words: Bryophyte-optimized primers, freshwater ecosystem, ITS2, mosses, rbcL, targeted capture Introduction Mosses, including Bryophyta, Marchantiophyta, and Anthocerophyta, play a crucial role in ecosystems and biodiversity conservation, especially during the first steps of ecological succession as pioneer species, but also for nutrient cycling (Ayres et al. 2006; Lindo and Gonzalez 2010), soil moisture Academic editor: Hugo de Boer Received: 9 April 2025 Accepted: 11 September 2025 Published: 8 October 2025 Citation: Reinhart T, Espinosa Prieto A, Begoc T, Tinguy H, Bick F, Chanez E, Beisel J-N, Hardion L (2025) A comparative analysis of hybridisation capture and PCR-based eDNA metabarcoding for monitoring bryophytes in riparian ecosystems. Metabarcoding and Metagenomics 9: e154548. https://doi.org/10.3897/ mbmg.9.154548 Metabarcoding and Metagenomics 9: 449–467 (2025) DOI: 10.3897/mbmg.9.154548 450 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments (Gornall et al. 2007) and water retention (Sharratt 1997). Mosses contribute significantly to plant biomass across diverse ecosystems from deserts to boreal and arctic regions. They also play a crucial role in forests by maintaining soil moisture and air humidity (Lindo and Gonzalez 2010; Eldridge et al. 2023). They also serve as a source for microorganisms, while providing habitat for a large diversity of microalgae, fungi, and microfauna (Usher and Booth 1986; Suren 1991; Döbbeler 1997; Roger Anderson 2006; Kauserud et al. 2008). With numerous taxa on red lists (Bick et al. 2014), bryophytes pose significant conservation challenges, further complicated by the scarcity of skilled taxonomists able to identify the different bryophyte taxa. In fact, they are equally threatened as vascular flora, and global change negatively impacts bryophyte colonies (Virtanen et al. 2024). However, this group is under-represented in ecological studies (Callaghan 2012) and biodiversity conservation policies, particularly because identification of the genus and species levels relies on micromorphological criteria that requires a high level of expertise (Tinguy 2021). Species identification is challenging insitu, and their small size and low biomass can bias field assessments (Callaghan 2012) and increase reliance on naturalist expertise, risking misidentification (Grace 1995). Molecular identification represents an alternative method that could facilitate the inclusion of these taxa in biodiversity monitoring programs. The emerging field of botanical eDNA presents the next frontier in eDNA research, offering promising avenues for biodiversity studies (Johnson et al. 2023). Currently, bryophytes have only been considered within broader multitaxon surveys, where only a limited number of moss taxa were identified (Cannon et al. 2016; Brunbjerg et al. 2019; Carvalho-Silva et al. 2021; Banerjee et al. 2022; Ariza et al. 2023). The lower biomass of bryophytes compared to vascular plants likely results in reduced eDNA abundance in the environment, potentially hindering their detection in eDNA samples when analysed alongside vascular plants, a challenge further exacerbated by the use of markers and primers designed primarily for flowering plants, as in these studies. To date, there is no consensus on suitable primers and markers for bryophyte metabarcoding (Liu et al. 2010; Epp et al. 2012; Espinosa Prieto et al. 2023). The few studies targeting bryophyte eDNA used markers within the plastid genome, trnL and rbcL (Liu et al. 2010; Von Cräutlein et al. 2011; Ballin et al. 2019; Yodphaka et al. 2018; Nelson et al. 2021). Epp et al. (2012) designed a pair of bryophytes primers for ancient DNA studies that amplified 50 bp of the trnL P6 loop with a taxonomic resolution at the species level of 30% from a database with 4020 species. However, barcode reference databases exhibit a significant gap in sequences for many bryophyte species compared to flowering plants (Liu et al. 2010). The effectiveness of metabarcoding depends on the taxonomic resolution of markers and the universality of primer pairs, that is, their capacity to amplify a broad range of target taxa (Liu et al. 2010; Hassel et al. 2013; Cheng et al. 2016). PCR metabarcoding overamplifies taxa with the strongest primer affinity and amplicons with the lower GC content, resulting in the amplification bias that compounds over successive reaction cycles (Moinard et al. 2023). Hybridisation capture (HC), also known as targeted capture, target enrichment, or capture enrichment, is a promising alternative to PCR-based plant 451 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments eDNA metabarcoding (Foster et al. 2021). Unlike PCR-based metabarcoding, HC circumvents amplification biases, which can lead to false positives and negatives, and uneven species representation (Krehenwinkel et al. 2017; Kelly et al. 2019). The technique relies on biotinylated RNA or DNA molecules (baits), complementary to target DNA regions to selective capture target sequences while removing non-hybridised DNA (Gnirke et al. 2009; Mamanova et al. 2010; Schuenemann et al. 2011; Carpenter et al. 2013; Marciniak et al. 2015). Hybridisation capture improves species detection accuracy, particularly when targeting multiple loci across genomes (Seeber et al. 2019; Jensen et al. 2021; Li et al. 2023). This method is particularly useful for degraded DNA samples, such as ancient DNA (Murchie et al. 2021; Revéret et al. 2023), stool samples (Aylward et al. 2018), and in tropical environments, where it facilitated the detection of low concentrated DNA and low biomass and rare species (Li et al. 2023). Additionally, it reduces PCR failures due to inhibitors co-extracted with eDNA (Jane et al. 2015; Murchie et al. 2019). This study aims to optimise eDNA methods for bryophyte detection, a taxonomic group underrepresented in ecological research. We compare three methods based on species richness and community dissimilarity. The methods include (i) visual transect surveys conducted by bryologists based on macroand micro-morphology, (ii) PCR metabarcoding of cpDNA rbcL and nrDNA ITS2 markers with newly designed primers, and (iii) a hybridisation capture approach for the same markers. The overarching goal is to facilitate the inclusion of bryophytes in ecological assessments and studies. Methods Field survey Our comparative analysis was carried out at seven sites from a previous eDNA study (Espinosa Prieto et al. 2024b) along the Falkensteinerbach River in the Vosges du Nord Regional Nature Park (Fig. 1). Here, we expand on our previous study in which two moss species were detected as bycatch using eDNA metabarcoding with vascular plant primers (Fontinalis antipyretica Hedw. and Funaria hygrometrica Hedw.). Metabarcoding data (PCR and HC) for this study was generated reusing the eDNA samples of Espinosa Prieto et al. 2024b collected in October 2020. Moss inventories were conducted on February 16 and March 8 2024 along a single 50-meter-long transect parallel to the river’s course, with a 5-meter-wide survey zone covering both riverbanks and the riverbed. The survey proceeded from the downstream to the upstream to minimise sediment disturbance. Bryophyte specimens were collected for species identification using microscopy. To strengthen confidence in our morphological identifications, we used DNA barcoding to verify morphologically challenging taxa. These samples are preserved in the Herbarium of the University of Strasbourg and in the private collections of Francis Bick and Hugues Tinguy. Species rarity for the study region was obtained from the assessment made by Tinguy (2021) and used to identify differences in the recoverability of rare taxa between methods. All taxonomic assignments in this study follow the taxonomic framework of TaxRef v17.0. 452 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Designing and adjusting primer pairs for bryophytes We first reviewed the literature on plant barcoding and metabarcoding to identify suitable primers to amplify the cpDNA rbcL gene and the nrDNA ITS2 in Bryophyta, Marchantiophyta, and Anthocerophyta taxa. We queried the Web of Science and PubMed databases in February 2024 (see Suppl. material 1: table S1 for a detailed list of primers). We performed in silico PCRs using ecoPCR from the OBITools package (Boyer et al. 2016) to evaluate 17 primer pairs for rbcL and 21 for ITS2, allowing up to three mismatches between primers and binding sites (Fig. 2; Suppl. material 1: table S2). Amplicon length was set between 100 bp and a marker-specific maximum: 600 bp for ITS2 and 700 bp for rbcL. In the absence of a satisfactory primer pair for ITS2 in the literature, we designed a primer pair that preferentially amplifies bryophytes: BraF 5’-CGCAAGTTGCGCCCGAGGCT-3’ (forward) and BraR 5’-GTGATATGCTTAAACTCAGCGGG-3’ (reverse). We manually explored the alignment and consensus sequences of all available bryophyte ITS2 sequences from the NCBI database using Geneious Prime (Dotmatics, Boston, USA). By comparing the consensus sequence with that of vascular plants, we identified a section within the flanking regions common to all mosses that differed by at least two nucleotides from vascular plants (Fig. 3). The cumulative number of mismatches between the primers and non-target taxa is known from previous studies (Espinosa Prieto et al. 2024a; Kolter and Gemeinholzer 2021) to result in preferential amplification of the target group. Moreover, the ITS2 flanking regions are highly conserved among Streptophyta supporting our design. Conservation of the priming sites was assessed using the Figure 1. Map of the study area within the Vosges du Nord Regional Nature Park, showing the seven sites and the seven eDNA sampling locations along the Falkensteinerbach, 1:150000 scale. The map was created in QGIS V3.32.2 with data from the DataGrandEst Qgis plug-in. 453 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments OBITools commands ‘ecopcr.forward.shanon’ and ‘ecopcr.reverse.shanon’, and the number of mismatches against mosses and vascular plants was visualised with the command ‘mismatchplot’. We applied standard primer design principles, ensuring a melting temperature difference within 5 °C between primers, primer length between 18 and 25 nucleotides, and we verified that the melting temperature of primer dimers and hairpin structures exceeded the annealing temperature. Furthermore, we prioritised GC clamps on the 3’ end while minimising 3’ end mismatches against bryophytes and promoting such mismatches against vascular plants (Suppl. material 2: fig. S1). The designed primer pair was then subjected to the same in silico analyses as the other primers (Suppl. material 2: fig. S2). Initial in silico analysis of rbcL primers from the literature also yielded unsatisfactory results. Unlike for ITS2, we did not find a better priming site that allowed for the design of metabarcoding primers specific to Bryophytes following the standards above. Instead, we customised primers from the literature (marked *) and rerun the in silico analysis. Ten primers were identified, generating amplicons ranging from 107 to 695 bp. More than 95% of the species in the reference database were amplified by most primer pairs (barcode coverage, Bc) except for those pairs with the forward 640F or the reverse 804 hR (Fig. 4). The three primer pairs containing the forward rbcL_aF primer, which generated Figure 3. Pie chart depicting the number of mismatches (ranging from zero to three) on the forward primer BRaF (x-axis) and the reverse primer BRaR (y-axis) for the ITS2 marker in mosses and vascular plants. The size of each pie represents the number of sequences successfully amplified in silico. Greater mismatch numbers on either primer indicate reduced PCR efficiency. Figure 2. Map of the best primers tested in silico. Primers linked with a bar were used for the amplification of eDNA samples. 454 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments amplicons of approximately 500 bp, exhibited the highest species-level barcode specificity (Bs) at around 75% (Fig. 4; Suppl. material 1: table S3). In contrast, primer pairs producing c. 200 bp amplicons, despite achieving 95% barcode coverage (Bc), demonstrated lower Bs values of approximately 30%, with the highest specificity attributed to primer pairs with the forward rbcL265 (Fig. 4). Environmental DNA methods PCR metabarcoding The PCR mixtures were prepared under a UV-sterilised hood, cleaned with DNA/ RNA-ExitusPlus™ IF (PanReac AppliChem, Germany) and with a Bunsen burner to prevent contamination of DNA samples from the laboratory environment. We used our new primer pair BraF + BraR for ITS2 and for the rbcL the primer pair rbcL265* + rbcL556* with our suggested modifications shown in bold 5‘-ATYGCTTAYGTTGCTTAYCC-3’ (forward)+ 5‘-CAYTCRTAWACWGCTCTACC-3’ (reverse) (modified from Tsubota et al. 1997; Aziz et al. 2017). PCRs were prepared in triplicate using fusion primers with inline dual indexes to tag each sample. PCR reactions were prepared in a final volume of 25 µL following the manufacturer’s instructions for the GoTaq® G2 Hot Start Master Mix (Promega, Madison, USA), comprising 12 µL of GoTaq Mmix, 2.5 µL of forward primer, 2.5 µL of reverse primer, 5 µL of DNA, and 3 µL of nuclease-free water. Amplification was carried out for 35 cycles with an initial denaturation at 95 °C for 3 min, followed by 98 °C for 20 s, Ta = 55 °C for rbcL and Ta = 65 °C for ITS2 for 15 s, 72 °C for 30 s and 72 °C for the final elongation time of 1 min. The PCR products were controlled on a 2% agarose gel under UV light and DNA concentrations Figure 4. Barcode characteristics of each rbcL primer pair tested in silico. The mean amplicon length (bp) is shown for each primer pair, with barcode coverage (Bc) and barcode specificity (Bs) are presented as percentages. (*) denotes primers modified in this study from their original sources; (**) indicates the modified reverse primer MrbcL163-R1. 455 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments were measured using the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, USA). Sequencing was performed by Eurofins Genomics Europe (Konstanz, Germany) with Illumina MiSeq 2x300 bp chemistry, with a theoretical sequencing depth of 120000 reads per sample and can be found under the NCBI bioproject PRJNA1301602 (http://www.ncbi.nlm.nih.gov/bioproject/1301602). Bioinformatic analysis was performed as in Espinosa Prieto et al. (2024b) adapting the pipeline available in https://zenodo.org/records/12571296. Hybridisation capture metabarcoding The present study precedes a broad in-depth study on hybridisation capture (HC) for plant eDNA metabarcoding. As such, the HC method used here was designed as a holistic multi-marker plant eDNA approach, including mosses. Preliminary analysis of the whole HC dataset showed promising results for moss detection compared to metabarcoding with universal plant primers (Espinosa Prieto et al. 2024a). Given the specificity of mosses and the lack of dedicated studies, we used a subset of the HC dataset to focus on this group in the present study. Additionally, HC was designed to capture four markers (rbcL, trnL, ITS1, and ITS2), but only the rbcL and ITS2 were considered in this study as they showed the best preliminary results for mosses. From the literature, rbcL was already used for moss DNA barcoding and had good reference data. For the ITS2 marker, barcode specificity (Bs) at the species level was above 90% in most cases according to our in-silico analysis, meaning it had the best taxonomic resolution (Suppl. material 1: table S3). We discarded the ITS1 marker as its length varies considerably between moss taxa and is not compatible with Illumina sequencing of pooled samples, which should be within 100 bp similar in length to reduce sequencing biases, and would have complicated the comparison with the PCR metabarcoding approach. The trnL showed poor taxonomic resolution for mosses compared to the other markers and was excluded. To shorten the method’s section, and because the HC approach is not exclusive to mosses, we presented a detailed description in Suppl. material 3. Given the nature of this study and the data, the baits and the HC raw sequence data will only be shared in the upcoming study. Nonetheless, the reads that yielded alignments with moss taxa after the BLAST can be found in the Zenodo repository (https://doi.org/10.5281/zenodo.16738645) together with the reference database / Accessions of mosses used for the bait design on the rbcL and ITS2 markers and the reference database used for taxonomic assignment of both PCR and HC reads (Suppl. material 4, 5). Statistics The following analyses were performed using R version 4.1.1 (2021.08.10). Venn diagrams were generated using the R package ggVennDiagram (version 1.5.2) to compare the proportion of species recovered across methods and to identify differences between markers in the two eDNA approaches. A Principal Coordinate Analysis (PCoA) based on Sørensen dissimilarity of community composition (beta diversity) at the same sites was conducted to identify patterns in species detection and assess consistency across methods, using the function vegdist from the vegan R package (version 2.6-8). Additionally, a Permutational 456 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Multivariate Analysis of Variance (PERMANOVA) was performed on the PCoA data to confirm distinct clustering among the groups using adonis2.We examined whether variations in community composition across methods could be attributed to differences in their sensitivity to detecting species with distinct functional traits, life forms, and ecological niches. Specifically, we assessed the correlation between species detectability by eDNA methods and dispersion traits, defined here as the propensity of taxa to propagate via vegetative propagules or sporophytes as a proxy for higher eDNA abundance, aswell as their preference for moist environments. Trait data were obtained from the BET database (Van Zuijlen et al. 2023). Additionally, we conducted an indicator species analysis using the indicspecies package in R to evaluate the likelihood of species being preferentially detected by specific methods. R scripts can be found in the Zenodo repository (https://doi.org/10.5281/zenodo.16738645). Results Across all methods, a total of 145 species were detected (see Suppl. material 1: table S4 for a detailed list of taxa); however, only nine species were recovered by all methods (Fig. 5). The field survey identified 68 species, 29 of which were exclusively detected through this approach, including eight species that lacked a reference sequence in the database. PCR metabarcoding recovered over 50% (n = 37) of the species observed in the field, whereas hybridisation capture (HC) detected only 16% (n = 11). The latter recovered 27 bryophyte species, with an average of five species per site. Of the nine species only detected by HC (Fig. 5), two are considered very rare in the region, Dicranella cerviculata (Hedw.) Schimp. and Riccia huebeneriana Lindenb., three are quite rare, Aneura pinguis (L.) Dumort., Hookeria lucens (Hedw.) Sm. and Trichocolea tomentella (Ehrh.) Dumort., three fairly common, Frullania tamarisci (L.) Dumort., Heterocladium heteropterum (Brid.) Schimp. and Physcomitrium pyriforme (Hedw.) Bruch & Schimp., and one is common, Didymodon luridus Hornsch. Species recovery differed between the two markers, ITS2 identified three times more species than rbcL (Fig. 5) and only two species were detected by both markers (Fontinalis antipyretica Hedw. and Ptychostomum capillare (Hedw.) Holyoak & N.Pedersen). ITS2 exhibited higher taxonomic resolution than rbcL, with 21 out of 31 unique sequences successfully assigned to a species, compared to only 6 out of 12 for rbcL, despite both markers having the same amplicon size. For example, 520 reads for the rbcL were assigned to Bryaceae at site F6, likely representing multiple taxa from this family observed in the field. PCR metabarcoding identified 101 species, including 57 species not recovered by the other methods (Fig. 5). Rare species were also detected, for example, Anomodon rugelii (Müll.Hal.) Keissl. (Bick et al. 2014; Tinguy 2021) was identified using the rbcL marker at sites F4 and F7 (Suppl. material 1: table S4). Both markers recovered a similar number of species, with rbcL detecting 67 and ITS2 detecting 59 (Fig. 5). However, approximately 40% of the species were uniquely detected by either marker, while only 20% were shared between them. Principal Coordinate Analysis (PCoA) was performed to explore dissimilarity patterns between the three methods. The first two principal coordinates (PC1 and PC2) explained 32.3% and 15.9% of the total variation (Fig. 6). A distinct clustering of samples based to the survey method was observed, indicating methodological 457 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments biases. The PERMANOVA results indicated significant differences between these groups (R2 = 0.898, F = 70.81, p = 0.001), suggesting that the model explains a substantial proportion of the variance in the data. However, our analysis refuted the hypothesis that these biases result from differences in the methods’ sensitivity to detecting species with specific traits and ecological preferences, as demonstrated by the rank-sum test on dispersal traits and preference for moist environments. Nevertheless, the indicator species analysis identified a set of species that were more likely to be detected by one method over the other (see Suppl. material 1: table S5 for detailed list of species). Discussion In this study, we compared three different approaches to study bryophyte communities in riverine and riparian environments. Our study is the first to evaluate hybridisation capture and PCR metabarcoding using bryophyte-enhanced primers, and to compare these to traditional field surveys. PCR metabarcoding has been shown to accurately capture plant species richness and community dissimilarity along ecological gradients (Shackleton et al. 2019; Reji Chacko et al. 2023). However, eDNA metabarcoding in rivers produces substantially different species lists than field surveys, and studies often highlight Figure 5. Venn diagrams of the proportion of species for the whole dataset (7 sites) detected through field surveys, PCR metabarcoding and hybridisation capture (HC); and by each marker for the two eDNA approaches. Total species richness (gamma diversity) is shown in parenthesis for each method and marker. Figure 6. Principal Coordinate Analysis (PCoA) based on Sørensen dissimilarity index, illustrating differences in community composition. The PCoA compares communities detected by hybridisation capture (HC), PCR metabarcoding, and field surveys across all sites. PERMANOVA confirmed significant differences between methods (R2 = 0.898, F = 70.81, p = 0.001). 464 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Mamanova L, Coffey AJ, Scott CE, Kozarewa I, Turner EH, Kumar A, Howard E, Shendure J, Turner DJ (2010) Target-enrichment strategies for next-generation sequencing. Nature Methods 7(2): 111–118. https://doi.org/10.1038/nmeth.1419 Marciniak S, Klunk J, Devault A, Enk J, Poinar HN (2015) Ancient human genomics: The methodology behind reconstructing evolutionary pathways. Journal of Human Evolution 79: 21–34. https://doi.org/10.1016/j.jhevol.2014.11.003 Moinard S, Piau D, Laporte F, Rioux D, Taberlet P, Gonindard-Melodelima C, Coissac E (2023) Towards quantitative DNA Metabarcoding: A method to overcome PCR amplification bias. PCR Bias in metabarcoding, 1–31 https://doi.org/10.1101/2023.10.03.560640 Moorhouse-Gann RJ, Dunn JC, De Vere N, Goder M, Cole N, Hipperson H, Symondson WOC (2018) New universal ITS2 primers for high-resolution herbivory analyses using DNA metabarcoding in both tropical and temperate zones. Scientific Reports 8(1): 8542. https://doi.org/10.1038/s41598-018-26648-2 Murchie TJ, Kuch M, Duggan AT, Ledger ML, Roche K, Klunk J, Karpinski E, Hackenberger D, Sadoway T, MacPhee R, Froese D, Poinar H (2019) Optimizing extraction and targeted capture of ancient environmental DNA for reconstructing past environments using the PalaeoChip Arctic-1.0 bait-set. Quaternary Research 99: 305–328. https://doi.org/10.1017/qua.2020.59 Murchie TJ, Monteath AJ, Mahony ME, Long GS, Cocker S, Sadoway T, Karpinski E, Zazula G, MacPhee RDE, Froese D, Poinar HN (2021) Collapse of the mammoth-steppe in central Yukon as revealed by ancient environmental DNA. Nature Communications 12(1): 7120. https://doi.org/10.1038/s41467-021-27439-6 Nelson JM, Hauser DA, Li F (2021) The diversity and community structure of symbiotic cyanobacteria in hornworts inferred from long‐read amplicon sequencing. American Journal of Botany 108(9): 1731–1744. https://doi.org/10.1002/ajb2.1729 Reji Chacko M, Altermatt F, Fopp F, Guisan A, Keggin T, Lyet A, Rey P-L, Richards E, Valentini A, Waldock C, Pellissier L (2023) Catchment-based sampling of river eDNA integrates terrestrial and aquatic biodiversity of alpine landscapes. Oecologia 202(4): 699–713. https://doi.org/10.1007/s00442-023-05428-4 Revéret A, Rijal DP, Heintzman PD, Brown AG, Stoof-Leichsenring KR, Alsos IG (2023) Environmental DNA of aquatic macrophytes: The potential for reconstructing past and present vegetation and environments. Freshwater Biology 68(11): 1929–1950. https://doi.org/10.1111/fwb.14158 Roger Anderson O (2006) The Density and Diversity of Gymnamoebae Associated with Terrestrial Moss Communities (Bryophyta: Bryopsida) in a Northeastern U.S. Forest. The Journal of Eukaryotic Microbiology 53(4): 275–279. https://doi.org/10.1111/ j.1550-7408.2006.00103.x Schuenemann VJ, Bos K, DeWitte S, Schmedes S, Jamieson J, Mittnik A, Forrest S, Coombes BK, Wood JW, Earn DJD, White W, Krause J, Poinar HN (2011) Targeted enrichment of ancient pathogens yielding the pPCP1 plasmid of Yersinia pestis from victims of the Black Death. Proceedings of the National Academy of Sciences of the United States of America 108(38): E746–E752. https://doi.org/10.1073/pnas.1105107108 Seeber PA, McEwen GK, Löber U, Förster DW, East ML, Melzheimer J, Greenwood AD (2019) Terrestrial mammal surveillance using hybridization capture of environmental DNA from African waterholes. Molecular Ecology Resources 19(6): 1486–1496. https://doi.org/10.1111/1755-0998.13069 Shackleton ME, Rees GN, Watson G, Campbell C, Nielsen D (2019) Environmental DNA reveals landscape mosaic of wetland plant communities. Global Ecology and Conservation 19: e00689. https://doi.org/10.1016/j.gecco.2019.e00689 465 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Sharratt BS (1997) Thermal conductivity and water retention of a black spruce forest floor. Soil Science 162(8): 576–582. https://doi.org/10.1097/00010694-19970800000006 Shogren AJ, Tank JL, Andruszkiewicz E, Olds B, Mahon AR, Jerde CL, Bolster D (2017) Controls on eDNA movement in streams: Transport, Retention, and Resuspension. Scientific Reports 7(1): 5065. https://doi.org/10.1038/s41598-017-05223-1 Suren AM (1991) Bryophytes as invertebrate habitat in two New Zealand alpine streams. Freshwater Biology 26(3): 399–418. https://doi.org/10.1111/j.1365-2427.1991. tb01407.x Tinguy H (2021) Catalogue des bryophytes du Bas Rhin (France). Les Nouvelles Archives de la Flore jurassienne et du nord-est de la France 19: 117. https://cbnfc-ori. org/especes-vegetales/catalogue-des-bryophytes-du-bas-rhin-france Tsubota H, Nakao N, Arikawa T, Yamaguchi T, Higuchi M, Deguchi H, Seki T (1997) A preliminary phylogeny of Hypnales (Musci) as inferred from chloroplast rbcL sequence data. Biological Research 7: 233–248. https://doi.org/10.24474/bryologicalresearch.7.8_233 Usher MB, Booth RG (1986) Arthropod Communities in a Maritime Antarctic Moss-Turf Habitat: Multiple Scales of Pattern in the Mites and Collembola. The Journal of Animal Ecology 55(1): 155. https://doi.org/10.2307/4699 Van Zuijlen K, Nobis MP, Hedenäs L, Hodgetts N, Calleja Alarcón JA, Albertos B, Bernhardt‐Römermann M, Gabriel R, Garilleti R, Lara F, Preston CD, Simmel J, Urmi E, Bisang I, Bergamini A (2023) Bryophytes of Europe Traits (bEt) data set: A fundamental tool for ecological studies. Journal of Vegetation Science 34(2): e13179. https://doi.org/10.1111/jvs.13179 Virtanen R, Harpole WS, Dunker S, Eskelinen A (2024) Multiple global change factors cause declines of a temperate bryophyte. Plant Ecology & Diversity 17(1-2): 1–12. https://doi.org/10.1080/17550874.2024.2330659 Von Cräutlein M, Korpelainen H, Pietiläinen M, Rikkinen J (2011) DNA barcoding: A tool for improved taxon identification and detection of species diversity. Biodiversity and Conservation 20(2): 373–389. https://doi.org/10.1007/s10531-010-9964-0 Yodphaka S, Boonpragob K, Lumbsch HT, Kraichak E (2018) Evaluation of six regions for their potential as DNA barcodes in epiphyllous liverworts from Thailand. Applications in Plant Sciences 6(8): e01174. https://doi.org/10.1002/aps3.1174 Supplementary material 1 Tables summarizing primer details, in silico analyses, detected taxa, and indicator species analysis results to support the findings presented in the main manuscript Authors: Thomas Reinhart, Armando Espinosa Prieto, Thomas Begoc, Hugues Tinguy, Francis Bick, Etienne Chanez, Jean-Nicolas Beisel, Laurent Hardion Data type: xlsx Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.154548.suppl1 466 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Supplementary material 2 Figures illustrating DNA logoplots and barcode characteristics for ITS2 primer pair tested in silico Authors: Thomas Reinhart, Armando Espinosa Prieto, Thomas Begoc, Hugues Tinguy, Francis Bick, Etienne Chanez, Jean-Nicolas Beisel, Laurent Hardion Data type: pdf Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.154548.suppl2 Supplementary material 3 Methods for Hybridisation capture metabarcoding and figures for read abundance thresholds Authors: Thomas Reinhart, Armando Espinosa Prieto, Thomas Begoc, Hugues Tinguy, Francis Bick, Etienne Chanez, Jean-Nicolas Beisel, Laurent Hardion Data type: docx Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.154548.suppl3 Supplementary material 4 Krona plot visualisations of the curated rbcL reference databases used for taxonomic assignment and primers/baits design in the study Authors: Thomas Reinhart, Armando Espinosa Prieto, Thomas Begoc, Hugues Tinguy, Francis Bick, Etienne Chanez, Jean-Nicolas Beisel, Laurent Hardion Data type: html Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.154548.suppl4 467 Metabarcoding and Metagenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548 Thomas Reinhart et al.: Comparing eDNA approaches for moss detection in lotic environments Supplementary material 5 Krona plot visualisations of the curated ITS2 reference databases used for taxonomic assignment and primers/baits design in the study Authors: Thomas Reinhart, Armando Espinosa Prieto, Thomas Begoc, Hugues Tinguy, Francis Bick, Etienne Chanez, Jean-Nicolas Beisel, Laurent Hardion Data type: html Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.154548.suppl5