Towards low-cost and PCR free field-based community metabarcoding

Towards low-cost and PCR free field-based community metabarcoding | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Method Article Towards low-cost and PCR free field-based community metabarcoding Amadeus Plewnia, Henrik Krehenwinkel, Christopher Heine This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5500332/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract 1. DNA metabarcoding has revolutionized our ability to monitor ecosystems. However, the method is still rarely used in developing countries where resources are limited and fieldwork is challenging. 2. To overcome this, we designed a comprehensive workflow allowing rapid community metabarcoding with minimum, self-manufacturable equipment in the field. We combine cellulose-based fast DNA extraction, optimized for environmental samples, with isothermal recombinase polymerase amplification (RPA). We demonstrate that stepwise RPA-based reamplification allows the introduction of non-homologous sequence adapters and index sequences, allowing the complete isothermal preparation of libraries for high-throughput sequencing. 3. Compared to laboratory-dependent approaches, the novel workflow halves costs and reduces hands on time more than fourfold. Our workflow makes metabarcoding broadly accessible for practitioners, conservationists and researchers without extensive laboratory equipment. Using environmental DNA from fish mock communities and a riverine ecotone, we demonstrate the sensitivity of the workflow and its interconnectivity with existing approaches. 4. Making metabarcoding available to researchers around the globe, our approach signifies a critical contribution to the immense task of characterizing and protecting earth’s biodiversity. biomonitoring 3D-printing environmental DNA field laboratory recombinase polymerase amplification third generation sequencing Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Simple, cost efficient and exhaustive characterization of species communities is key to understanding global biodiversity change and ecosystem health. Metabarcoding is a rapidly advancing field, allowing standardized community-wide barcoding from non-invasive samples with broad applications from bulk barcoding to environmental DNA (eDNA) and microbiome research (Lange et al. 2014 ; Creer et al. 2016 ; Compson et al. 2020 ; Keck et al. 2022 ; Gillespie et al. 2023 ). Despite the broad potential and applicability of metabarcoding, the method is still poorly employed by practitioners due to its complexity and is inaccessible to researchers that lack vast laboratory and financial resources. The inaccessibility of DNA metabarcoding technology especially affects developing countries, which at the same time harbor the vast majority of earth’s biodiversity (cf. Collen et al. 2008 ; Geijzendorffer et al. 2015 ). Simplified and cost-efficient DNA metabarcoding approaches are thus direly needed to tackle the immense task of characterizing and safeguarding global biodiversity. Recent technological developments are now contributing to achieve this goal. The increasing miniaturization of laboratory equipment makes entire laboratories field deployable and has resulted in groundbreaking work on field-based DNA barcoding of individual specimens and rapid single-species detection (LaBarre et al. 2011 ; Krehenwinkel et al. 2019 ; Pomerantz et al. 2022 ). While high error rates have precluded field-based metabarcoding (Krehenwinkel et al. 2019 ), recent advancements in sequence quality of third generation sequencing platforms increase the suitability of portable sequencers for metabarcoding (del Socorro Toxqui Rodríguez et al. 2023 ; Ohta et al. 2023 ). However, high-throughput sequencing remains poorly field-deployable to date. Firstly, for community metabarcoding, both on-site sample processing and DNA extraction require vast equipment and hands-on time, particularly since samples are often inhibitor-rich with little and degraded template DNA (Deiner et al. 2018 ). Direct amplification and ready-to-use microfluidic extraction channels are first steps to overcome this (Fukuzawa et al. 2022 ; Röder & Schwenk 2023 ). Fast, cost-efficient and equipment-free DNA extraction, such as with cellulose paper, now enables simplified, laboratory-independent molecular work (Su & Comeau 1999 ; Zou et al. 2017 ). However, likely due to low recovery rates compared to kit-based solutions, it has not been established for high-throughput sequencing technologies yet. Further, metabarcoding applications rely on PCR, requiring extensive equipment and long reaction times, while alternative direct sequencing approaches are associated with strongly increased costs. With the development of isothermal alternatives to PCR such as recombinase polymerase amplification (RPA), loop-mediated isothermal amplification, rolling circle amplification or helicase-dependent amplification, molecular tools became increasingly field-deployable (Zanoli & Spoto 2013 ). For instance, RPA-CRISPR/Cas-based systems have been developed to rapidly detect single species, mostly pathogens, from swab, blood or even environmental samples (Guo et al. 2023 ; Hoenig et al. 2023 , 2024 ; Zhang et al. 2024 ). While these approaches have proven highly sensitive and reliable, they only enable the detection of single or few predefined targets with known sequences and are not yet available for high-throughput sequencing (Hu et al. 2022 ; Xu et al. 2022 ). Here we tackle these limitations of DNA metabarcoding by ( i ) designing a self-manufacturable field laboratory at low cost, ( ii ) developing a simplified, cellulose-based fast DNA extraction protocol for environmental samples and ( iii ) employing a novel step-wise isothermal library-preparation technique for both second and third generation sequencing. Compared to available laboratory-based approaches, the equally sensitive but completely field-deployable workflow significantly reduces hands-on time and halves costs of existing protocols as demonstrated with environmental DNA from fish mock communities and a riverine ecotone. Making metabarcoding broadly accessible, the novel workflow represents a critical step forward in characterizing and protecting earth’s biodiversity. Methods Field laboratories. We designed a simple, low-cost and field-deployable filtration system for water samples based on the Easy Pump IV peristaltic pumphead (Innofluid Co. Ltd.; Fig. 1 ). The pumphead was installed on a weather sealed plastic box and powered by a 12 Volt DM-36RP555 DC motor (Dong Ming Motor Co. Ltd.) which is mounted onto a 3D printed PLA adapter plate inside the box and connected to the pumphead via a brass shaft. The systems runs on 3 rechargeable 18650 lithium-ion batteries (Ledlenser GmbH) but can be powered by external power sources using the weather sealed plug on the outside of the box, providing the flexibility to use external power sources such as common vehicle mounted auxiliary power outlets. If the pump is used for prolonged periods at a time (> 20 minutes), an auxiliary fan can be switched on to regulate the motor temperature. To toggle the pump and auxiliary fan on and off, we designed a custom printed circuit board (PCB) (Supplementary Methods). For mechanic lysis with beads, we employed a cordless oscillating multifunctional construction tool at 11.000–20.000 rpm and an oscillation angle of 3.2° (Varrito, Einhell AG) with tubes placed on the outer corner of the saw blade in a 3D-printed adapter. As a centrifuge in the field, we employed a drone motor (Turnigy, Hobbying Ltd.) with a 3D printed head and housing. Like the pump, the centrifuge is powered by 3 rechargeable lithium-ion batteries (Ledlenser GmbH). With the help of an Arduino Nano microcontroller (Arduino S.r.l.), centrifugation speed can be adjusted. The head has space for eight 2 ml tubes and two strips of eight 0.2 ml tubes. Housing and lid are made from 3D printed PLA while the head is made from UV resin to ensure a higher resolution and equal weight distribution. The centrifuge reaches a maximum speed of ~ 8.000 rpm. To aid in the incubation of amplification reactions and the ligation of Nanopore sequencing adapters, we designed a simple heated bed, utilizing two generic 12 Volt silicone heating mats and an Arduino Nano (Arduino S.r.l.) microcontroller, incorporated into a housing printed from PLA. Technical drawings, mesh files and schematics were designed in Freecad version 0.21.1. and EasyEDA version 6.5.44. Files were printed using generic PLA filament on an Ender-3 Pro or using generic UV resin on an Anycubic Photon Mono. The brass shaft was machined using a lathe and mill. Technical drawings and schematics for all 3D-printed parts and the PCBs as well as step-by-step instructions on rebuilding are available in Supplementary Methods. Sampling, filter preservation and DNA extraction. Assay development and validation were based on a fish community from the universities’ 450 l aquarium housing a varying, known composition of native and introduced central European species. This provided us with a mock community with differences in expected abundance (i.e. ‘rare’ fish with a single, small individual vs. ‘abundant’ species with many or large specimens) which presented semi-natural conditions (i.e. likely presence of inhibitors from the heavily planted and non-target DNA rich environment). For water sampling, we collected a composite sample by mixing several 1L samples from which we subsampled replicates of 500 ml per filter. Water was filtered with the self-manufactured pump (see above) using 0.45 µm nitrocellulose filter discs (47 mm diameter, Sartorius) placed in 250 ml filter funnels (Nalgene, ThermoScientific). For high sample throughput or applications with increased contamination risk it may be favorable to conduct library preparation in a laboratory environment. For optimized storage of processed water filters, we therefore evaluated five temperature-independent preservation methods, which have either been shown to prevent DNA degradation at room temperature or are commonly used lysis buffers for DNA extraction. We stored 5 filters each in I 1.5 ml tubes containing 450 µl of ATL buffer (Qiagen), II 1.5 ml tubes containing 300 µl of DNA/RNA Shield (Zymo research), III falcon tubes containing 20 ml of UV-sterilized silica gel grains, IV 1.5 ml tubes containing 450 µl of cell lysis buffer (homemade as of Puregene kits, Qiagen) and V 1.5 ml tubes containing 450 µl of lysis buffer from Zou et al. ( 2017 ) (hereafter ‘fast lysis buffer’). As a control, we stored 5 filters dried only by the vacuum generated by our pump in 1.5 ml tubes without addition of preservation agents. As most of the used preservation agents induce lysis already, we kept all fluids for extraction. We extracted DNA using the Blood and Tissue Kit (Qiagen). We adapted lysis conditions by adding a mechanic bead-beating step (0.0350–0.0400 g of silica-zirconium beads, 0.5 mm diameter, Roth; mechanic lysis for 45 sec. at 1500 rpm) prior to enzymatic lysis but after addition of ATL. We increased lysis volume to 450 µl of ATL to ensure suitable conditions for proteinase activity (added to all treatments except those already stored in ATL) and 50 µl proteinase K with 2 h lysis duration. All other steps followed the manufacturers protocol. We eluted DNA in 50 µl AE buffer. Fast DNA extraction. Complementary to sample preservation, we adapted a rapid and low-cost on-site DNA extraction protocol (Zou et al. 2017 ). To explore the possibility of increasing DNA yield from environmental samples in comparison to the tissue-based protocol of Zou et al. ( 2017 ), we isolated 10 aquarium mock community filters each with stepwise adapted treatments. Firstly, we ( i ) followed Zou et al. ( 2017 ) by applying a fast lysis for 30 sec in 500 µl homemade lysis buffer to filters, followed by the removal of filters and gentle agitation of an 8mm Whatman paper #1 cellulose filter dipstick as binding agent for 30 sec with subsequent transfer of the filter disc to 200 µl homemade wash buffer for another 15 sec and final elution by storing the filter disc in 50 µl AE buffer (10 mM Tris-Cl, 0.5 mM EDTA pH 9.0) and shaking with 0.035–0.040 g silica-zirconium beads (Roth) for 30 sec to homogenize the filter disc. We compared these samples to ( ii ), for which we adapted ( i ) by placing the Whatman filter disc on a 3D-printed spin column (printed with UV resin on an Anycubic Photon Mono), transferring the lysate on the column and subsequent quick centrifugation until the entire lysate passed the filter disc for initial DNA binding. For ( iii ), we added an additional bead-beating step during lysis to ( ii ), using 0.035–0.040 g silica-zirconium beads (Roth) and shaking for 30 sec with the cordless oscillating multifunctional construction tool (see above). We quantified DNA with a Qubit Fluorometer (ThermoFisher) in duplicate measures using the high-sensitivity dsDNA assay following the manufacturer’s instructions. Based on Qubit Fluorometer reads, we selected the best performing rapid DNA extraction protocol for next generation sequencing as described below. PCR-based library preparation and next generation sequencing. We amplified a ~ 170 bp long fragment of the 12S rRNA in duplicates, using the Tele02 primer pair (forward primer 5′- ACACTCTTTCCCTACACGACGCTCTTCCGATCT AAACTCGTGCCAGCCACC-3′, reverse primer 5′- GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GGGTATCTAATCCCAGTTTG-3′, TruSeq adapter italicized) specific to Teleostei (Taberlet et al. 2018 ) in 10µl reactions (TaqMan Multiplex, 35 cycles, annealing at 63°C). We prepared libraries with dual indexing as described in Lange et al. ( 2014 ), pooled samples to equal amounts and purified pools with magnetic beads (NucleoMag, Macherey-Nagel). We sequenced pools on a MiSeq platform (Illumina) with ~ 20.000 reads coverage for each PCR replicate using v2 reagent kits following the manufacturer´s instructions. PCR free amplification and library preparation. We chose recombinase polymerase amplification (RPA) due to its fast and robust performance as well as its reaction activity at ~ 37°C (Piepenburg et al. 2006 ). For amplification, we used the TwistAmp Basic kit (TwistDX) following the manufacturer’s instructions but reducing reaction volume to 10 µl with 1µl template DNA and a reaction time of 20–30 min. Initial attempts to amplify target DNA with RPA using different eDNA primers for fish failed. However, we were able to re-amplify PCR-amplicons with the same primer pairs in RPA. Therefore, we screened available primer pairs (Zhang et al. 2020 ) without TruSeq adapters for their compatibility with RPA. As available eDNA primers are far below the recommended length of at least 30 bases for RPA, we additionally designed extended versions of Tele02 by adding bases on both ends based on visually identified conserved flanking regions from a sequence alignment of all fish species from our aquarium mock community. The resulting primers (MiPez-F 5′- CCGGTA AAACTCGTGCCAGCCACC GCGGT -3′, MiPez-short-F 5′- CCGGTA AAACTCGTGCCAGCCACC GC -3′, MiPez-R 5′- GCATAGTG GGGTATCTAATCCCAGTTTG T -3′, added bases in bold) were checked for dimers using the Multiple Primer Analyzer (ThermoFisher) and blasted to Europe´s most basal ray-finned fish taxon, Acipenseriformes, to ensure complete sequence identity over the phylogeny. For Illumina sequencing, after selection of RPA-compatible primer pairs, we re-amplified successful RPA products with a version of the same primer pair, now carrying the TruSeq adapters for indexing and subsequently barcoded samples with dual indexes in a third RPA similarly as described for PCR (Lange et al. 2014 ). Index-RPA incubation was terminated after 20 min, amplicons pooled and subsequently purified as described above to prevent cross-indexing at ambient temperatures in the pool. Pools were sequenced on a MiSeq as described above. We used both kit and fast extracts from the aquarium filters for MiSeq library preparation. For Oxford nanopore sequencing, we reamplified MiPez-short-F/R RPA products with the same primer pair, now carrying 20 base indices adopted from Gajski et al. ( 2024 ). RPA incubation was terminated after 20 min and products were pooled to equal amounts, followed by subsequent purification to prevent cross-indexing at ambient temperature as described above. We prepared libraries with the SQK-LSK114 kit following the manufacturer’s instructions. Sequencing was conducted for 48 h on a MinION using a Flongle flow cell (FLO-FLG114, Oxford Nanopore Technologies) following the manufacturer´s instructions. We used kit extracts from the aquarium filters to explore MinION library preparation. Sequence processing. Demultiplexed paired-end MiSeq reads were merged using PEAR (Zhang et al. 2014 ). We filtered merged reads for having at least 90% of bases with a quality score of at least Q30 using the FASTX Toolkit (Gordon & Hannon 2010 ). After trimming of primer sequences, we dereplicated and clustered sequences and generated 3% radius OTU tables containing all sequences with at least five reads using USEARCH (Edgar 2010). OTUs were mapped on a local copy of GenBank’s nucleotide database (downloaded 20. December 2023) using BLASTn 2.5.0 (Camacho et al. 2009 ) and assigned to species names using blast2taxonomy (Schöneberg 2023 ). The resulting taxa list was filtered for Teleostei with percent identity ≥ 98% and contamination from marine fish through feeder pellets manually removed. For MinION reads, we conducted basecalling in Guppy 6.5.7 and demultiplexed and trimmed reads in minibar with barcode edit distance value set to 2 and primer edit distance value to 9 (Krehenwinkel et al. 2019 ). We subsequently filtered for Phred scores > 10 and read length between 155 and 175 bases using chopper (De Coster & Rademakers 2023 ). Currently available pipelines for OTU clustering from MinION data are largely optimized for single-species barcoding, long amplicon sequences or simple metagenetic communities. Since our dataset was based on a short marker, a community containing closely related species in unequal abundance and a high expected sequence error rate, we explored the compatibility of the pipelines amplicon_sorter and NGSpeciesID with default and minimal clustering settings using selected samples from our aquarium mock community (Sahlin et al. 2021 ; Vierstraete & Braeckman 2022). Both pipelines were unable to cluster the reads into consensus sequences that would reveal all species from our community, generally only clustering OTUs on higher taxonomic level. Instead, without conducting OTU clustering, we directly mapped the filtered sequences on a local copy of GenBank’s nucleotide database using BLASTn and assigned species names as described above. The resulting taxa list was filtered for percent identity ≥ 95%. This threshold was experimentally evaluated based on known mock community composition using the lowest threshold that did not yield erroneous taxa assignments while still including a high number of sequence reads for abundance counts. Due to the high error rate of MinION sequence data, only few sequences passed higher thresholds. We subsequently counted the number of sequences for each species name and discarded species derived from a single sequence only. Resulting data were further processed and visualized in R Studio (2023.12.1 + 402) using the packages dplyr (Wickham et al. 2023 ), ggplot2 (Wickham 2016 ), circlize (Gu et al. 2014 ). Case study. To validate the performance of our workflow, we collected water from 10 equally distributed sampling sites along a riverine transect spanning 20 km on the German river Ruwer (from 49.6648 N 6.7193 E to 49.7854 N 6.7053 E). The river constitutes a partially restored and well-preserved ecosystem for native fish communities of the lower trout and upper grayling regions. Natural and anthropogenic barriers along its course led us to hypothesize a change in species composition with an increase in species richness downstream. We filtered water in the field using the equipment described above and collected two replicate filters of 1 l each per site. On site, we extracted DNA from filters using the bead- and column-based fast extraction protocol derived in this study and subsequently performed RPA-based Nanopore library preparation using body heat for incubation. For pipetting, we employed scaled microliter capillaries to minimize the amount of lab equipment needed in the field. Libraries were pooled and purified in the field as described above and brought to the desk for MinION sequencing. Results An integrative approach expands the viability of metabarcoding. A schematic overview on the novel metabarcoding workflow is given in Fig. 1 . It combines on-site sample filtration and optimized cellulose-based fast DNA extraction in a miniaturized self-manufacturable laboratory with isothermal stepwise RPA-based library preparation for either lab-based MiSeq or on-site MinION sequencing. The water filtration system developed performed well with no differences in OTU counts and species composition when compared to a stationary laboratory pump and a mobile Vampire Sampler (Supplementary Material). Processing samples with the new workflow from sample collection to ready-to-sequence MiSeq libraries took about 100 min on average and had consumable costs of 7.64 € per replicate (vs. 460 min and 11.52 € for traditional, PCR-based sample processing; Supplementary Material). A complete field protocol combining all optimized steps from our workflow spanning sample collection to library preparation as well as information on field laboratories is available as Supplementary Methods. Optimized DNA storage and on-site isolation from environmental samples for metabarcoding. All treatments except the fast lysis buffer were able to preserve a sufficient amount of target DNA to detect all species from the aquarium mock community in all replicates, while untreated control samples showed substantial degradation and dropout of ‘rare’ taxa (i.e. small species with one or few specimens in the aquarium; Fig. 2 a & b). Absolute read counts were highest for ATL, DNA/RNA Shield and silica gel and lower for cell lysis buffer (Fig. 2 a; Supplementary Material). However, DNA yield measured with a Qubit fluorometer showed significant differences and was highest in ATL and cell lysis buffer (ANOVA, p < 0.001; for Tukey´s HSD for pairwise differences see Supplementary Material; Fig. 2 c). The fast lysis buffer did not preserve any target DNA despite showing a distinct band of genomic DNA in gel electrophoresis. Subsequent testing revealed strong amplification with bacterial 12S primers (F515/R806, Caporaso et al. 2010 ) but not with general vertebrate 12S primers (12S-V5, Riaz et al. 2011 ) hinting to bacterial growth as causative agent of increased degradation during storage. For most metabarcoding applications, however, on-site DNA extraction might be favorable. Therefore, we optimized a readily available cellulose-based extraction method for inhibitor-rich and low-target environmental samples by ( i ) modification of cellulose-based DNA capture and ( ii ) addition of a mechanic lysis step. The protocol takes about 2 minutes hands-on time per sample. In our first trial, the cellulose column based protocol showed significantly higher DNA yield compared to the dipstick method of Zou et al. ( 2017 ) (Fig. 2 d, Bonferroni corrected t-test p = 0.007; Supplementary Material). We subsequently added mechanic lysis with silica-zirconium beads to the better performing protocol which resulted in an additional increase in DNA yield (Fig. 2 d, Bonferroni corrected t-test p = 0.002; Supplementary Material). Although extracts from the optimized fast extraction protocol yielded only roughly 7% the amount of DNA when compared to Blood & Tissue kit isolates (Supplementary Material), fast extracts were sufficient to reliably detect most taxa from our aquarium mock community through MiSeq sequencing of both PCR- and RPA-based libraries (Fig. 3 ; see below). Absolute number of reads from fast extracts, however, were constantly below those of Blood & Tissue kit isolates and not all replicates showed all taxa. Over all replicates, only the PCR-based library from fast extracts was missing one ‘rare’ species, Cottus gobio (Supplementary Material). Stepwise reamplification with RPA primers allows complete isothermal library preparation. Initial attempts to amplify fish DNA in isothermal recombinase polymerase amplification with various metabarcoding primers carrying TruSeq adapters for MiSeq libraries or 20 base indexes for MinION libraries failed. Instead, we were able to amplify templates with the longer primers designed in this study for RPA, only when adapters were not included in the primer sequence (both the combination of MiPez-F and MiPez-short-F with MiPez-R showed amplification with more prominent bands in the latter pair) while available metabarcoding PCR primers (from Zhang et al 2020 ) failed to amplify without TruSeq tails except for one combination (Ac12S-F/12S-V5-R). However, some combinations of those with the RPA primers were able to amplify target molecules (i.e. MiFish-U-F/MiPez-R, MiPez-F/MiFish-U-R, MiPez-short-F/MiFish-U-R). Using MiPez-short-F/MiPez-R amplicons as template, we were able to introduce TruSeq adapters in a second and dual indexes in a third reamplification to construct MiSeq libraries with RPA. RPA- and PCR-based libraries were equally able to detect all species from our aquarium mock community. However, we observed slight differences in relative abundance of OTUs (Fig. 3 ) with Gasterosteus aculeatus in RPA and Cyprinus carpio in PCR being the least abundant taxa respectively (Fig. 3 ; Supplementary Material). While RPA detected all species from fast extracts, absolute read counts were generally lower in RPA (Supplementary Material). The long RPA primers almost exclusively amplified fish with few human reads while other vertebrates (mammals, birds) were among the most common reads in Tele02-based PCR libraries. However, in unfiltered MiSeq reads from RPA, short dimers of ~ 60 bases containing forward and reverse primer and a fragment of the TruSeq adapter were most common, particularly in fast extracts and negative controls. Similarly, using MiPez-short-F/MiPez-R amplicons as template for MinION library preparation, we were able to introduce dual indexes in a second reamplification with RPA. With direct BLAST of quality filtered reads, the RPA-based MinION library was able to detect all species from our aquarium mock community with all 10 species detected from 9 out of 10 filters and one missing a single ‘rare’ species ( G. aculeatus ). Relative species abundance was in line with the RPA-based MiSeq libraries with Chondrostoma nasus being the most and G. aculeatus the least abundant taxa (Fig. 3 ). However, taxonomic resolution was overall lower in MinION reads with up to 7.85% (mean 4.77%) of erroneous BLAST assignments (names of fish not part of the aquarium mock community; Supplementary Material). Fast extraction and isothermal library preparation for third generation sequencing shows community turnover along a riverine ecotone. Using the workflow described above in the field with third generation sequencing, we detected a total of ten fish species. Our direct BLAST approach largely assigned names of species known to occur in the river system (HK unpubl. data). However, in several cases, proper taxonomic assignment occurred only on the genus level (i.e. various Cottus spp. and various Phoxinus spp. detected while only one species occurs in the study area each; (Supplementary Material). Cottus and Phoxinus were detected in all sampling sites, Salmo trutta and Barbatula barbatula occurred in most samples along the entire river course. Particularly in Cottus spp., we observed a decline in read abundance downstream as approximated from sequence blast counts. Squalius cephalus was exclusively found in the lower sections of the river course and two additional species ( Alburnoides bipunctatus and Rutilus rutilus ) were detected only at the confluence with the Mosel river. Other taxa did not show site-specific trends in occurrence (Fig. 4 ; Supplementary Material). Discussion Through a combination of bead-cellulose-based fast DNA extraction and stepwise RPA library preparation and aided by a complete field laboratory, we developed a metabarcoding workflow that requires little equipment, halves costs and reduces hands-on time more than fourfold in comparison to previous approaches. The cost-effective and rapid assessment of communities is essential for applications such as biodiversity conservation, detection of infectious disease or microbiome research. Exemplified by the successful characterization of a diverse ecotone of central European fish communities and validated by an aquarium mock community, the novel workflow is shown to be as sensitive as currently available protocols. While slightly increasing complexity and hands-on time, the introduction of bead beating and simplified spin columns increased DNA yield from environmental samples in cellulose-based fast extraction (Zou et al. 2017 ). To detect rare taxa, kit-based extraction may still be favorable, while optimized fast extraction is sufficient to approximate community composition. With its low cost, increased numbers of replicates are feasible and can likely in part compensate the lower detection rates of rare taxa. For laboratory-based extraction, we show that multiple treatments can equally preserve sample integrity in the field, with common lysis reagents and silica-gel drying being as effective as costly preservation reagents (Camacho-Sanchez et al. 2013 ; Menke et al. 2017 ). So far, isothermal amplification techniques have not been employed for complete preparation of high throughput sequencing libraries for metabarcoding (Magriñá Lobato et al. 2018; Li et al. 2019 ). In RPA, this is perhaps due to the non-homologous bases preventing binding of recombinase/primer filaments to template DNA, as the latter is only locally displaced in RPA. This might explain why only stepwise RPA-based reamplification was able to introduce adapters. Despite the additional amplification step, RPA-based libraries did not show increased amplification bias and we demonstrated that isothermal library preparation recovered complete species compositions and highly similar community structure from our aquarium mock community. The slight differences in relative OTU abundance are difficult to assess and may require comparison with direct sequencing of genomic DNA from isolates to approach amplification bias for the two methods (Krehenwinkel et al. 2017 ). We did not observe target sequences in negative controls, suggesting that cross-indexing in RPA pools at ambient temperature is not an issue. The higher specificity of RPA to our target taxon (Teleostei) is likely explained by the longer RPA primers. As metabarcoding primers are generally short but RPA performs best with primers > 30 bases, design of novel RPA primers for higher taxa may be hampered by the lack of long conserved regions and they may be more prone to forming artifacts. As such, the reduced number of absolute MiSeq target reads in our RPA libraries compared to PCR is likely attributed to random amplification and dimer production, particularly with little template in environmental samples (Magriñá Lobato et al. 2018; Munawar et al. 2020 ; Wu et al. 2020 ). The use of self-avoiding molecular recognition systems (SAMRS) in the RPA primers (Li et al. 2019 ) may overcome this risk and size selection with magnetic beads of pooled libraries might be able to reduce short fragments as competitors in sequencing pools to further increase the performance of RPA-based libraries. While our workflow combined with portable third generation sequencing devices allows complete metabarcoding in the field, we present alternatives for optimized sample storage and library preparation for laboratory-based high throughput sequencing. Both allow for integration in currently established metabarcoding workflows. Sequencing with portable Oxford Nanopore platforms is still scarcely used for metabarcoding due to its higher error rates (Chang et al. 2023 ). The high similarity in relative abundance as shown here between MiSeq and MinION libraries is particularly promising for even quantitative metabarcoding with Oxford Nanopore platforms. With complete on-site library preparation, MinION sequencing consistently recovered the characteristic community composition of our riverine ecosystem and detected a downstream increase in species richness but decrease of sequence reads from characteristic members of the upper trout region. However, with direct BLAST instead of clustering, assessing species communities deeply relies on near-complete sequencing databases. Future studies may fill these gaps and focus on systematically assessing error rates of MinIon metabarcoding as well as on the design of novel RPA primers amplifying longer fragments (Hatfield et al. 2023 ; Huggins et al. 2024a , b ). RPA has been shown to amplify fragments of up to 1,500 bases (Piepenburg et al. 2006 ), which can increase taxonomic resolution and reduce false positives from direct BLAST in metabarcoding. Our workflow constitutes a first step towards automated preparation of rapid metabarcoding libraries when integrated into emerging lab-on-a-chip techniques. Particularly in microbiome research where bacterial contamination is a widespread problem, complete on-site processing could help conserve the site specific OTU composition and prevent contamination during transport or with other samples in the lab. Research that is slowed or prevented by sample accessibility, such as sample shipment under the Nagoya protocol, will benefit from on-site diagnostics. In laboratory settings, the novel workflow may still be favored due to its strong reduction of hands-on time and its cost efficiency. Most importantly, we hope our study will democratize metabarcoding by making it more accessible for researchers in developing countries where threatened biodiversity richness peaks and standardized monitoring tools, such as eDNA metabarcoding, are needed more than ever. Declarations Data availability for Peer review. References Lange V, Böhme I, Hofmann J, Lang K, Sauter J, Schöne B, Paul P, Albrecht V, Andreas JM, Baier DM, Nething J, Ehninger U, Schwarzelt C, Pingel J, Ehninger G, Schmidt AH (2014) Cost-efficient high-throughput HLA typing by MiSeq amplicon sequencing. BMC Genomics 15:63 Creer S, Deiner K, Frey S, Porazinska D, Taberlet P, Thomas WK, Potter C, Bik HM (2016) The ecologist´s field guide to sequence-based identification of biodiversity. Methods Ecol Evol 7:1008–1018 Compson ZG, McClenaghan B, Singer GAC, Fahner NA, Hajibabaei M (2020) Metabarcoding from microbes to mammals: comprehensive bioassessment on a global scale. Front Ecol Evol 8:581835 Keck F, Blackman RC, Bossart R, Brantschen J, Couton M, Hürlemann S et al (2022) Meta-analysis shows both congruence and complementarity of DNA and eDNA metabarcoding to traditional methods for biological community assessment. Mol Ecol 31:1820–1835 Gillespie RG, Bik HM, Hickerson MJ, Krehenwinkel H, Overcast I, Rominger AJ (2023) Insigths into ecological & evolutionary processes via community metabarcoding. Mol Ecol 32:6083–6092 Collen B, Ram M, Zamin T, McRae L (2008) The tropical biodiversity data gap: addressing disparity in global monitoring. Trop Conserv Sci 1:75–88 Geijzendorffer IR, Regan EC, Pereira HM, Brotons L, Brummitt N, Gavish Y et al (2015) Bridging the gap between biodiversity data and policy reporting needs: An Essential Biodiversity Variables perspective. J Appl Ecol 53:1341–1350 LaBarre P et al (2011) A simple, inexpensive device for nucleic acid amplification without electricity – toward instrument-free molecular diagnostics in low-resource settings. PLoS ONE 6:e19738 Krehenwinkel H et al (2019) Nanopore sequencing of long ribosomal DNA amplicons enables portable and simple biodiversity assessments with high phylogenetic resolution across broad taxonomic scale. Gigascience 8:giz006 Pomerantz A et al (2022) Rapid in situ identification of biological specimens via DNA amplicon sequencing using miniaturized laboratory equipment. Nat Protoc 17:1415–1443 del Rodríguez ST, Vanhollebeke M, J., Derycke S (2023) Evaluation of DNA metabarcoding using Oxford Nanopore sequencing for authentication of mixed seafood products. Food Control 145:109388 Ohta A, Nishi K, Hirota K, Matsuo Y (2023) DNA metabarcoding workflow utilizing nanopore long-read sequencing and consensus generation for rapid identification of fungal taxa with high phylogenetic resolution. BioRxiv preprint Deiner K et al (2018) Optimising the detection of marine taxonomic richness using environmental DNA metabarcoding: the effects of filter material, pore size and extraction method. MBMG 2:e28963 Fukuzawa T, Kameda Y, Nagata H, Nishizawa N, Doi H (2022) Filtration extraction method using a microfluidic channel for measuring environmental DNA. Mol Ecol Res 22:2651–2661 Röder N, Schwenk K (2023) Direct PCR meets high-throughput sequencing – metabarcoding of chironomid communities without DNA extraction. MBMG 7:e102455 Su X, Comeau AM (1999) Cellulose as a matrix for nucleic acid purification. Anal Biochem 267:418–420 Zou Y et al (2017) Nucleic acid purification from plants, animals and microbes in under 30 seconds. PLoS Biol 15:e2003916 Zanoli LM, Spoto G (2013) Isothermal Amplification Methods for the Detection of Nucleic Acids in Microfluidic Devices. Biosensors 3:18–43 Guo Y, Xia H, Dai T, Liu T (2023) RPA-CRISPR/Cas12a mediated isothermal amplification for visual detection of Phytophthora sojae . Front Cell Infect Microbiol 13:1208837 Hoenig BD et al (2023) FINDeM: A CRISPR-based, molecular method for rapid, inexpensive and field-deployable organism detection. Methods Ecol Evol 14:3055–3067 Hoenig BD, Böning P, Plewnia A (2024) & Richards-Zawacki, C. Using FINDeM for rapid, CRISPR-based detection of the emerging salamandrid fungal pathogen, Batrachochytrium salamandrivorans . Ecohealth in press Zhang W, Qu H, Wu X, Shi J, Wang X (2024) Rapid, sensitive, and user-friendly detection of Pseudomonas aeruginosa using the RPA/CRISPR/Cas12a system. BMC Infect Dis 24:458 Hu T et al (2022) CRISPR/Cas12a-enabled multiplex biosensing strategy via an affordable and visual nylon membrane readout. Adv Sci 10:2204689 Xu Z et al (2022) Microfluidic space coding for multiplexed nucleic acid detection via CRISPR-Cas12a and recombinase polymerase amplification. Nat Commun 13:6480 Caporaso JG, Lauber CL, Walters WA, Knight R (2010) Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. PNAS 108:4516–4522 Riaz T, Shehzad W, Viari A, Pompanon F, Taberlet P, Coissac E (2011) ecoPrimers: inference of new DNA barcode markers from whole genome sequence analysis. Nucleic Acids Res 39:e145 Zhang S, Zhao J, Yao M (2020) A comprehensive and comparative evaluation of primers for metabarcoding eDNA from fish. Methods Ecol Evol 11:1609–1625 Camacho-Sanchez M, Burraco P, Gomez-Mestre I, Leonard JA (2013) Preservation of RNA and DNA from mammal samples under field conditions. Mol Ecol Resour 13:663–673 Menke S, Gillingham MAF, Wilhelm K, Sommer S (2017) Home-made cost effective preservation buffer is a better alternative to commercial preservation methods for microbiome research. Front microbiol 8:102 Magriñá Lobato I, O’Sullivan CK (2018) Recombinase polymerase amplification: Basics, applications and recent advances. Trends Anal Chem 98:19–35 Li J, Macdonald J, von Stetten F (2019) Review: a comprehensive summary of a decade development of the recombinase polymerase amplification. Analyst 144:31–67 Krehenwinkel H, Wolf M, Lim JY, Rominger AJ, Simison WB, Gillespie RG (2017) Estimating and mitigating amplification bias in qualitative and quantitative arthropod metabarcoding. Sci Rep 7:17668 Munawar MA, Martin F, Toljamo A, Kokko H, Oksanen E (2020) RPA-PCR couple: an approach to expedite plant diagnostics and overcome PCR inhibitors. Biotechniques 69:271–280 Wu H, Zhao P, Yang X, Li J, Zhang J, Zhang X, Zeng Z, Dong J, Gao S, Lu C (2020) A recombinase polymerase amplification and lateral flow strip combined method that detects Salmonella enterica serotype typhimurium with no worry of primer-dependent artifacts. Front Microbiol 11:1015 Chang JJM, Ip YCA, Neo WL, Mowe MAD, Jaafar Z, Huang D (2023) Primed and ready: nanopore metabarcoding can now recover highly accurate consensus barcodes that are generally indel-free. BioRxiv preprint Hatfield RG, Ryder D, Tidy AM, Hartnell DM, Dean KJ, Batista FM (2023) combining nanopore sequencing with recombinase polymerase amplification enables identification of dinoflagellates from the Alexandrium genus, providing a rapid, field deployable tool. Toxins 15:372 Huggins LG, Colella V, Young ND, Traub RJ (2024a) Metabarcoding using nanopore long-read sequencing for the unbiased characterization of apicomplexan haemoparasites. Mol Ecol Res 24:e13878 Huggins LG, Atapattu U, Young ND, Traub RJ, Colella V (2024b) Development and validation of a long-read metabarcoding platform for the detection of filarial worm pathogens of animals and humans. BMC Microbiol 24:28 Piepenburg O, Williams CH, Stemple DL, Armes (2006) N.A. DNA detection using recombination proteins. PLoS Biol 4:e204 Taberlet P, Bonin A, Zinger L, Coissac E, Environmental (2018) DNA: for biodiversity research and monitoring. Oxford University Press, Oxford, UK, p 252 Gajski D, Wolff JO, Melcher A, Weber S, Prost S, Krehenwinkel H, Kennedy S (2024) Facilitating taxonomy and phylogenetics: An informative and cost-effective protocol integrating long amplicon PCRs and third-generation sequencing. Mol Phylogenet Evol 192:107988 Zhang J, Kobert K, Flouri T, Stamatakis A (2014) PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics 30:614–620 Gordon A, Hannon GJ (2010) Fastx-toolkit. FASTQ/A short-reads pre-processing tools. Unpublished Http://Hannonlab.Cshl.Edu/Fastx_Toolkit Edgar RC (2009) Search and clustering orders of magnitude faster than BLAST. Bioinformatics, 26, 2460–2461. BMC Bioinformatics 10, 421 Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL (2009) BLAST+: architecture and applications. BMC Bioinformatics 15:421 Schöneberg Y (2023) yschoeneberg/blast2taxonomy: v1.3.4. 10.5281/zenodo.10009721 . Zenodo De Coster W, Rademakers R (2023) NanoPack2: population-scale evaluation of long-read sequencing data. Bioinform 39:btad311 Sahlin K, Lim MCW, Prost S (2021) NGSpeciesID: DNA barcode and amplicon consensus generation from long-read sequencing data. Ecol Evol 11:1392–1398 Vierstraete AR, Braeckman BP, Amplicon_sorter: (2022) A tool for reference-free amplicon sorting based on sequence similarity and for building consensus sequences. Ecol Evol 12:e8603 Wickham H, François R, Henry L, Müller K (2023) & Vaughan, D. dplyr: A grammar of data manipulation. R package version 1.1.4 Wickham H (2016) ggplot2: Elegant Graphics for Data Analysis. Springer-, New York Gu Z, Gu L, Eils R, Schlesner M, Brors B (2014) Circlize implements and enhances circular visualization in R. Bioinform 30:2811–2812 Supplementary Materials Supplementary Materials are not available with this version. Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5500332","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Method Article","associatedPublications":[],"authors":[{"id":381115194,"identity":"da17f5a1-3cac-44ca-8d74-94f68cb1bbe4","order_by":0,"name":"Amadeus Plewnia","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0001-8731-6312","institution":"Trier University","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Amadeus","middleName":"","lastName":"Plewnia","suffix":""},{"id":381115195,"identity":"cf29b576-a21b-4563-9d08-c345ad70a416","order_by":1,"name":"Henrik Krehenwinkel","email":"","orcid":"https://orcid.org/0000-0001-5069-8601","institution":"Trier University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Henrik","middleName":"","lastName":"Krehenwinkel","suffix":""},{"id":381115196,"identity":"30ae390f-0c38-46aa-b3e2-1c238b47420d","order_by":2,"name":"Christopher Heine","email":"","orcid":"https://orcid.org/0009-0002-6103-2391","institution":"Trier University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Heine","suffix":""}],"badges":[],"createdAt":"2024-11-21 21:21:34","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false,"coiExplicitlySet":false},"doi":"10.21203/rs.3.rs-5500332/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5500332/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":69855030,"identity":"1e28ec78-f703-4324-b30b-8a0a1cbdd66f","added_by":"auto","created_at":"2024-11-26 02:50:34","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":719130,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSchematic workflow of PCR-free metabarcoding with cellulose column-based fast DNA extraction and RPA-based library preparation for next or third generation sequencing.\u003c/strong\u003eField sampling is based on a self-manufacturable filtration system (upper row, center). Library preparation from fast extracts constitutes three (Illumina) or two (Nanopore) stepwise RPAs with the first generating short amplicons with a completely homologous primer followed by reamplification(s) with primers carrying adapter and/or index sequences; blue = template strand, red = primer sequence, pink = primer and binding site, orange = barcoding adapters, purple = index, for Illumina with flowcell adapter, violet = recombinase proteins, green = single-stranded binding proteins. Created in BioRender. https://BioRender.com/j55c700\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-5500332/v1/43b394e72f90ca635b12df57.png"},{"id":69855029,"identity":"fa15150f-b368-4e2d-851b-6b521a763a73","added_by":"auto","created_at":"2024-11-26 02:50:33","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":125236,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOptimizing DNA preservation and extraction methods for field settings. a \u003c/strong\u003eAbsolute and \u003cstrong\u003eb\u003c/strong\u003erelative OTU counts and \u003cstrong\u003ec\u003c/strong\u003e DNA yield (ng/µl) from eDNA filters collected from an aquarium mock community and stored at ambient conditions for six weeks in different temperature-independent preservation agents: ATL (Qiagen), DNA/RNA Shield (Zymo research), silica gel, cell lysis buffer (Puregene) and fast lysis buffer (Zou et al. 2017) with untreated filters as degradation control. \u003cstrong\u003ed\u003c/strong\u003eDNA yield (ng/µl) from stepwise improved fast DNA extraction protocols that are field-deployable.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-5500332/v1/ba91bfc7138b3260449b6519.png"},{"id":69855032,"identity":"176a1af8-a593-489e-af5a-709d9f215388","added_by":"auto","created_at":"2024-11-26 02:50:34","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":372629,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePerformance of RPA-based libraries for next and third generation sequencing from mock community samples. \u003c/strong\u003eAbsence of differences in community composition but in relative abundance among RPA- and PCR-based MiSeq and MinION libraries from fast and Qiagen extracts.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-5500332/v1/689f71d49d33876ac19528f6.png"},{"id":69855042,"identity":"2124f9ce-741c-4eac-acb1-0bfc6e20d940","added_by":"auto","created_at":"2024-11-26 02:50:34","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":486791,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCommunity turnover along a riverine ecotone derived from on-site metabarcoding. \u003c/strong\u003eRead counts along the river Ruwer as detected with fast extraction, RPA library preparation and MinION sequencing with direct BLAST. Note read counts presented in logarithmic scale.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-5500332/v1/3770ad5c9f1241d15c5d9cd4.png"},{"id":69856046,"identity":"82c52045-653a-4ec0-8f1f-ee20bc5f8679","added_by":"auto","created_at":"2024-11-26 03:14:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2462619,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5500332/v1/4bd6d606-1b47-4e74-962c-c3affbf6b539.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eTowards low-cost and PCR free field-based community metabarcoding\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSimple, cost efficient and exhaustive characterization of species communities is key to understanding global biodiversity change and ecosystem health. Metabarcoding is a rapidly advancing field, allowing standardized community-wide barcoding from non-invasive samples with broad applications from bulk barcoding to environmental DNA (eDNA) and microbiome research (Lange et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Creer et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Compson et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Keck et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Gillespie et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Despite the broad potential and applicability of metabarcoding, the method is still poorly employed by practitioners due to its complexity and is inaccessible to researchers that lack vast laboratory and financial resources. The inaccessibility of DNA metabarcoding technology especially affects developing countries, which at the same time harbor the vast majority of earth\u0026rsquo;s biodiversity (cf. Collen et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Geijzendorffer et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Simplified and cost-efficient DNA metabarcoding approaches are thus direly needed to tackle the immense task of characterizing and safeguarding global biodiversity.\u003c/p\u003e \u003cp\u003eRecent technological developments are now contributing to achieve this goal. The increasing miniaturization of laboratory equipment makes entire laboratories field deployable and has resulted in groundbreaking work on field-based DNA barcoding of individual specimens and rapid single-species detection (LaBarre et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Krehenwinkel et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Pomerantz et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). While high error rates have precluded field-based metabarcoding (Krehenwinkel et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), recent advancements in sequence quality of third generation sequencing platforms increase the suitability of portable sequencers for metabarcoding (del Socorro Toxqui Rodr\u0026iacute;guez et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ohta et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, high-throughput sequencing remains poorly field-deployable to date. Firstly, for community metabarcoding, both on-site sample processing and DNA extraction require vast equipment and hands-on time, particularly since samples are often inhibitor-rich with little and degraded template DNA (Deiner et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Direct amplification and ready-to-use microfluidic extraction channels are first steps to overcome this (Fukuzawa et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; R\u0026ouml;der \u0026amp; Schwenk \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Fast, cost-efficient and equipment-free DNA extraction, such as with cellulose paper, now enables simplified, laboratory-independent molecular work (Su \u0026amp; Comeau \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Zou et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). However, likely due to low recovery rates compared to kit-based solutions, it has not been established for high-throughput sequencing technologies yet.\u003c/p\u003e \u003cp\u003eFurther, metabarcoding applications rely on PCR, requiring extensive equipment and long reaction times, while alternative direct sequencing approaches are associated with strongly increased costs. With the development of isothermal alternatives to PCR such as recombinase polymerase amplification (RPA), loop-mediated isothermal amplification, rolling circle amplification or helicase-dependent amplification, molecular tools became increasingly field-deployable (Zanoli \u0026amp; Spoto \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). For instance, RPA-CRISPR/Cas-based systems have been developed to rapidly detect single species, mostly pathogens, from swab, blood or even environmental samples (Guo et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Hoenig et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Zhang et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). While these approaches have proven highly sensitive and reliable, they only enable the detection of single or few predefined targets with known sequences and are not yet available for high-throughput sequencing (Hu et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Xu et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHere we tackle these limitations of DNA metabarcoding by (\u003cb\u003ei\u003c/b\u003e) designing a self-manufacturable field laboratory at low cost, (\u003cb\u003eii\u003c/b\u003e) developing a simplified, cellulose-based fast DNA extraction protocol for environmental samples and (\u003cb\u003eiii\u003c/b\u003e) employing a novel step-wise isothermal library-preparation technique for both second and third generation sequencing. Compared to available laboratory-based approaches, the equally sensitive but completely field-deployable workflow significantly reduces hands-on time and halves costs of existing protocols as demonstrated with environmental DNA from fish mock communities and a riverine ecotone. Making metabarcoding broadly accessible, the novel workflow represents a critical step forward in characterizing and protecting earth\u0026rsquo;s biodiversity.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e \u003cb\u003eField laboratories.\u003c/b\u003e We designed a simple, low-cost and field-deployable filtration system for water samples based on the Easy Pump IV peristaltic pumphead (Innofluid Co. Ltd.; Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The pumphead was installed on a weather sealed plastic box and powered by a 12 Volt DM-36RP555 DC motor (Dong Ming Motor Co. Ltd.) which is mounted onto a 3D printed PLA adapter plate inside the box and connected to the pumphead via a brass shaft. The systems runs on 3 rechargeable 18650 lithium-ion batteries (Ledlenser GmbH) but can be powered by external power sources using the weather sealed plug on the outside of the box, providing the flexibility to use external power sources such as common vehicle mounted auxiliary power outlets. If the pump is used for prolonged periods at a time (\u0026gt;\u0026thinsp;20 minutes), an auxiliary fan can be switched on to regulate the motor temperature. To toggle the pump and auxiliary fan on and off, we designed a custom printed circuit board (PCB) (Supplementary Methods).\u003c/p\u003e \u003cp\u003eFor mechanic lysis with beads, we employed a cordless oscillating multifunctional construction tool at 11.000\u0026ndash;20.000 rpm and an oscillation angle of 3.2\u0026deg; (Varrito, Einhell AG) with tubes placed on the outer corner of the saw blade in a 3D-printed adapter.\u003c/p\u003e \u003cp\u003eAs a centrifuge in the field, we employed a drone motor (Turnigy, Hobbying Ltd.) with a 3D printed head and housing. Like the pump, the centrifuge is powered by 3 rechargeable lithium-ion batteries (Ledlenser GmbH). With the help of an Arduino Nano microcontroller (Arduino S.r.l.), centrifugation speed can be adjusted. The head has space for eight 2 ml tubes and two strips of eight 0.2 ml tubes. Housing and lid are made from 3D printed PLA while the head is made from UV resin to ensure a higher resolution and equal weight distribution. The centrifuge reaches a maximum speed of ~\u0026thinsp;8.000 rpm.\u003c/p\u003e \u003cp\u003eTo aid in the incubation of amplification reactions and the ligation of Nanopore sequencing adapters, we designed a simple heated bed, utilizing two generic 12 Volt silicone heating mats and an Arduino Nano (Arduino S.r.l.) microcontroller, incorporated into a housing printed from PLA.\u003c/p\u003e \u003cp\u003eTechnical drawings, mesh files and schematics were designed in Freecad version 0.21.1. and EasyEDA version 6.5.44. Files were printed using generic PLA filament on an Ender-3 Pro or using generic UV resin on an Anycubic Photon Mono. The brass shaft was machined using a lathe and mill.\u003c/p\u003e \u003cp\u003eTechnical drawings and schematics for all 3D-printed parts and the PCBs as well as step-by-step instructions on rebuilding are available in Supplementary Methods.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSampling, filter preservation and DNA extraction.\u003c/b\u003e Assay development and validation were based on a fish community from the universities\u0026rsquo; 450 l aquarium housing a varying, known composition of native and introduced central European species. This provided us with a mock community with differences in expected abundance (i.e. \u0026lsquo;rare\u0026rsquo; fish with a single, small individual vs. \u0026lsquo;abundant\u0026rsquo; species with many or large specimens) which presented semi-natural conditions (i.e. likely presence of inhibitors from the heavily planted and non-target DNA rich environment). For water sampling, we collected a composite sample by mixing several 1L samples from which we subsampled replicates of 500 ml per filter. Water was filtered with the self-manufactured pump (see above) using 0.45 \u0026micro;m nitrocellulose filter discs (47 mm diameter, Sartorius) placed in 250 ml filter funnels (Nalgene, ThermoScientific).\u003c/p\u003e \u003cp\u003eFor high sample throughput or applications with increased contamination risk it may be favorable to conduct library preparation in a laboratory environment. For optimized storage of processed water filters, we therefore evaluated five temperature-independent preservation methods, which have either been shown to prevent DNA degradation at room temperature or are commonly used lysis buffers for DNA extraction. We stored 5 filters each in \u003cb\u003eI\u003c/b\u003e 1.5 ml tubes containing 450 \u0026micro;l of ATL buffer (Qiagen), \u003cb\u003eII\u003c/b\u003e 1.5 ml tubes containing 300 \u0026micro;l of DNA/RNA Shield (Zymo research), \u003cb\u003eIII\u003c/b\u003e falcon tubes containing 20 ml of UV-sterilized silica gel grains, \u003cb\u003eIV\u003c/b\u003e 1.5 ml tubes containing 450 \u0026micro;l of cell lysis buffer (homemade as of Puregene kits, Qiagen) and \u003cb\u003eV\u003c/b\u003e 1.5 ml tubes containing 450 \u0026micro;l of lysis buffer from Zou et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) (hereafter \u0026lsquo;fast lysis buffer\u0026rsquo;). As a control, we stored 5 filters dried only by the vacuum generated by our pump in 1.5 ml tubes without addition of preservation agents.\u003c/p\u003e \u003cp\u003eAs most of the used preservation agents induce lysis already, we kept all fluids for extraction. We extracted DNA using the Blood and Tissue Kit (Qiagen). We adapted lysis conditions by adding a mechanic bead-beating step (0.0350\u0026ndash;0.0400 g of silica-zirconium beads, 0.5 mm diameter, Roth; mechanic lysis for 45 sec. at 1500 rpm) prior to enzymatic lysis but after addition of ATL. We increased lysis volume to 450 \u0026micro;l of ATL to ensure suitable conditions for proteinase activity (added to all treatments except those already stored in ATL) and 50 \u0026micro;l proteinase K with 2 h lysis duration. All other steps followed the manufacturers protocol. We eluted DNA in 50 \u0026micro;l AE buffer.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFast DNA extraction.\u003c/b\u003e Complementary to sample preservation, we adapted a rapid and low-cost on-site DNA extraction protocol (Zou et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). To explore the possibility of increasing DNA yield from environmental samples in comparison to the tissue-based protocol of Zou et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), we isolated 10 aquarium mock community filters each with stepwise adapted treatments. Firstly, we (\u003cb\u003ei\u003c/b\u003e) followed Zou et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) by applying a fast lysis for 30 sec in 500 \u0026micro;l homemade lysis buffer to filters, followed by the removal of filters and gentle agitation of an 8mm Whatman paper #1 cellulose filter dipstick as binding agent for 30 sec with subsequent transfer of the filter disc to 200 \u0026micro;l homemade wash buffer for another 15 sec and final elution by storing the filter disc in 50 \u0026micro;l AE buffer (10 mM Tris-Cl, 0.5 mM EDTA pH 9.0) and shaking with 0.035\u0026ndash;0.040 g silica-zirconium beads (Roth) for 30 sec to homogenize the filter disc. We compared these samples to (\u003cb\u003eii\u003c/b\u003e), for which we adapted (\u003cb\u003ei\u003c/b\u003e) by placing the Whatman filter disc on a 3D-printed spin column (printed with UV resin on an Anycubic Photon Mono), transferring the lysate on the column and subsequent quick centrifugation until the entire lysate passed the filter disc for initial DNA binding. For (\u003cb\u003eiii\u003c/b\u003e), we added an additional bead-beating step during lysis to (\u003cb\u003eii\u003c/b\u003e), using 0.035\u0026ndash;0.040 g silica-zirconium beads (Roth) and shaking for 30 sec with the cordless oscillating multifunctional construction tool (see above). We quantified DNA with a Qubit Fluorometer (ThermoFisher) in duplicate measures using the high-sensitivity dsDNA assay following the manufacturer\u0026rsquo;s instructions. Based on Qubit Fluorometer reads, we selected the best performing rapid DNA extraction protocol for next generation sequencing as described below.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePCR-based library preparation and next generation sequencing.\u003c/b\u003e We amplified a\u0026thinsp;~\u0026thinsp;170 bp long fragment of the 12S rRNA in duplicates, using the Tele02 primer pair (forward primer 5\u0026prime;-\u003cem\u003eACACTCTTTCCCTACACGACGCTCTTCCGATCT\u003c/em\u003eAAACTCGTGCCAGCCACC-3\u0026prime;, reverse primer 5\u0026prime;-\u003cem\u003eGTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT\u003c/em\u003eGGGTATCTAATCCCAGTTTG-3\u0026prime;, TruSeq adapter italicized) specific to Teleostei (Taberlet et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) in 10\u0026micro;l reactions (TaqMan Multiplex, 35 cycles, annealing at 63\u0026deg;C). We prepared libraries with dual indexing as described in Lange et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), pooled samples to equal amounts and purified pools with magnetic beads (NucleoMag, Macherey-Nagel). We sequenced pools on a MiSeq platform (Illumina) with ~\u0026thinsp;20.000 reads coverage for each PCR replicate using v2 reagent kits following the manufacturer\u0026acute;s instructions.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePCR free amplification and library preparation.\u003c/b\u003e We chose recombinase polymerase amplification (RPA) due to its fast and robust performance as well as its reaction activity at ~\u0026thinsp;37\u0026deg;C (Piepenburg et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). For amplification, we used the TwistAmp Basic kit (TwistDX) following the manufacturer\u0026rsquo;s instructions but reducing reaction volume to 10 \u0026micro;l with 1\u0026micro;l template DNA and a reaction time of 20\u0026ndash;30 min. Initial attempts to amplify target DNA with RPA using different eDNA primers for fish failed. However, we were able to re-amplify PCR-amplicons with the same primer pairs in RPA. Therefore, we screened available primer pairs (Zhang et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) without TruSeq adapters for their compatibility with RPA. As available eDNA primers are far below the recommended length of at least 30 bases for RPA, we additionally designed extended versions of Tele02 by adding bases on both ends based on visually identified conserved flanking regions from a sequence alignment of all fish species from our aquarium mock community. The resulting primers (MiPez-F 5\u0026prime;-\u003cb\u003eCCGGTA\u003c/b\u003eAAACTCGTGCCAGCCACC\u003cb\u003eGCGGT\u003c/b\u003e-3\u0026prime;, MiPez-short-F 5\u0026prime;-\u003cb\u003eCCGGTA\u003c/b\u003eAAACTCGTGCCAGCCACC\u003cb\u003eGC\u003c/b\u003e-3\u0026prime;, MiPez-R 5\u0026prime;-\u003cb\u003eGCATAGTG\u003c/b\u003eGGGTATCTAATCCCAGTTTG\u003cb\u003eT\u003c/b\u003e-3\u0026prime;, added bases in bold) were checked for dimers using the Multiple Primer Analyzer (ThermoFisher) and blasted to Europe\u0026acute;s most basal ray-finned fish taxon, Acipenseriformes, to ensure complete sequence identity over the phylogeny.\u003c/p\u003e \u003cp\u003eFor Illumina sequencing, after selection of RPA-compatible primer pairs, we re-amplified successful RPA products with a version of the same primer pair, now carrying the TruSeq adapters for indexing and subsequently barcoded samples with dual indexes in a third RPA similarly as described for PCR (Lange et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Index-RPA incubation was terminated after 20 min, amplicons pooled and subsequently purified as described above to prevent cross-indexing at ambient temperatures in the pool. Pools were sequenced on a MiSeq as described above. We used both kit and fast extracts from the aquarium filters for MiSeq library preparation.\u003c/p\u003e \u003cp\u003eFor Oxford nanopore sequencing, we reamplified MiPez-short-F/R RPA products with the same primer pair, now carrying 20 base indices adopted from Gajski et al. (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). RPA incubation was terminated after 20 min and products were pooled to equal amounts, followed by subsequent purification to prevent cross-indexing at ambient temperature as described above. We prepared libraries with the SQK-LSK114 kit following the manufacturer\u0026rsquo;s instructions. Sequencing was conducted for 48 h on a MinION using a Flongle flow cell (FLO-FLG114, Oxford Nanopore Technologies) following the manufacturer\u0026acute;s instructions. We used kit extracts from the aquarium filters to explore MinION library preparation.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSequence processing.\u003c/b\u003e Demultiplexed paired-end MiSeq reads were merged using PEAR (Zhang et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). We filtered merged reads for having at least 90% of bases with a quality score of at least Q30 using the FASTX Toolkit (Gordon \u0026amp; Hannon \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). After trimming of primer sequences, we dereplicated and clustered sequences and generated 3% radius OTU tables containing all sequences with at least five reads using USEARCH (Edgar 2010). OTUs were mapped on a local copy of GenBank\u0026rsquo;s nucleotide database (downloaded 20. December 2023) using BLASTn 2.5.0 (Camacho et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) and assigned to species names using blast2taxonomy (Sch\u0026ouml;neberg \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The resulting taxa list was filtered for Teleostei with percent identity\u0026thinsp;\u0026ge;\u0026thinsp;98% and contamination from marine fish through feeder pellets manually removed.\u003c/p\u003e \u003cp\u003eFor MinION reads, we conducted basecalling in Guppy 6.5.7 and demultiplexed and trimmed reads in minibar with barcode edit distance value set to 2 and primer edit distance value to 9 (Krehenwinkel et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We subsequently filtered for Phred scores\u0026thinsp;\u0026gt;\u0026thinsp;10 and read length between 155 and 175 bases using chopper (De Coster \u0026amp; Rademakers \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Currently available pipelines for OTU clustering from MinION data are largely optimized for single-species barcoding, long amplicon sequences or simple metagenetic communities. Since our dataset was based on a short marker, a community containing closely related species in unequal abundance and a high expected sequence error rate, we explored the compatibility of the pipelines amplicon_sorter and NGSpeciesID with default and minimal clustering settings using selected samples from our aquarium mock community (Sahlin et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Vierstraete \u0026amp; Braeckman 2022). Both pipelines were unable to cluster the reads into consensus sequences that would reveal all species from our community, generally only clustering OTUs on higher taxonomic level. Instead, without conducting OTU clustering, we directly mapped the filtered sequences on a local copy of GenBank\u0026rsquo;s nucleotide database using BLASTn and assigned species names as described above. The resulting taxa list was filtered for percent identity\u0026thinsp;\u0026ge;\u0026thinsp;95%. This threshold was experimentally evaluated based on known mock community composition using the lowest threshold that did not yield erroneous taxa assignments while still including a high number of sequence reads for abundance counts. Due to the high error rate of MinION sequence data, only few sequences passed higher thresholds. We subsequently counted the number of sequences for each species name and discarded species derived from a single sequence only.\u003c/p\u003e \u003cp\u003eResulting data were further processed and visualized in R Studio (2023.12.1\u0026thinsp;+\u0026thinsp;402) using the packages dplyr (Wickham et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), ggplot2 (Wickham \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), circlize (Gu et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eCase study.\u003c/b\u003e To validate the performance of our workflow, we collected water from 10 equally distributed sampling sites along a riverine transect spanning 20 km on the German river Ruwer (from 49.6648 N 6.7193 E to 49.7854 N 6.7053 E). The river constitutes a partially restored and well-preserved ecosystem for native fish communities of the lower trout and upper grayling regions. Natural and anthropogenic barriers along its course led us to hypothesize a change in species composition with an increase in species richness downstream. We filtered water in the field using the equipment described above and collected two replicate filters of 1 l each per site. On site, we extracted DNA from filters using the bead- and column-based fast extraction protocol derived in this study and subsequently performed RPA-based Nanopore library preparation using body heat for incubation. For pipetting, we employed scaled microliter capillaries to minimize the amount of lab equipment needed in the field. Libraries were pooled and purified in the field as described above and brought to the desk for MinION sequencing.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eAn integrative approach expands the viability of metabarcoding.\u003c/b\u003e A schematic overview on the novel metabarcoding workflow is given in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. It combines on-site sample filtration and optimized cellulose-based fast DNA extraction in a miniaturized self-manufacturable laboratory with isothermal stepwise RPA-based library preparation for either lab-based MiSeq or on-site MinION sequencing. The water filtration system developed performed well with no differences in OTU counts and species composition when compared to a stationary laboratory pump and a mobile Vampire Sampler (Supplementary Material). Processing samples with the new workflow from sample collection to ready-to-sequence MiSeq libraries took about 100 min on average and had consumable costs of 7.64 \u0026euro; per replicate (vs. 460 min and 11.52 \u0026euro; for traditional, PCR-based sample processing; Supplementary Material). A complete field protocol combining all optimized steps from our workflow spanning sample collection to library preparation as well as information on field laboratories is available as Supplementary Methods.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eOptimized DNA storage and on-site isolation from environmental samples for metabarcoding.\u003c/b\u003e All treatments except the fast lysis buffer were able to preserve a sufficient amount of target DNA to detect all species from the aquarium mock community in all replicates, while untreated control samples showed substantial degradation and dropout of \u0026lsquo;rare\u0026rsquo; taxa (i.e. small species with one or few specimens in the aquarium; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea \u0026amp; b). Absolute read counts were highest for ATL, DNA/RNA Shield and silica gel and lower for cell lysis buffer (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea; Supplementary Material). However, DNA yield measured with a Qubit fluorometer showed significant differences and was highest in ATL and cell lysis buffer (ANOVA, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; for Tukey\u0026acute;s HSD for pairwise differences see Supplementary Material; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). The fast lysis buffer did not preserve any target DNA despite showing a distinct band of genomic DNA in gel electrophoresis. Subsequent testing revealed strong amplification with bacterial 12S primers (F515/R806, Caporaso et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) but not with general vertebrate 12S primers (12S-V5, Riaz et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) hinting to bacterial growth as causative agent of increased degradation during storage.\u003c/p\u003e \u003cp\u003eFor most metabarcoding applications, however, on-site DNA extraction might be favorable. Therefore, we optimized a readily available cellulose-based extraction method for inhibitor-rich and low-target environmental samples by (\u003cb\u003ei\u003c/b\u003e) modification of cellulose-based DNA capture and (\u003cb\u003eii\u003c/b\u003e) addition of a mechanic lysis step. The protocol takes about 2 minutes hands-on time per sample. In our first trial, the cellulose column based protocol showed significantly higher DNA yield compared to the dipstick method of Zou et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed, Bonferroni corrected t-test p\u0026thinsp;=\u0026thinsp;0.007; Supplementary Material). We subsequently added mechanic lysis with silica-zirconium beads to the better performing protocol which resulted in an additional increase in DNA yield (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed, Bonferroni corrected t-test p\u0026thinsp;=\u0026thinsp;0.002; Supplementary Material). Although extracts from the optimized fast extraction protocol yielded only roughly 7% the amount of DNA when compared to Blood \u0026amp; Tissue kit isolates (Supplementary Material), fast extracts were sufficient to reliably detect most taxa from our aquarium mock community through MiSeq sequencing of both PCR- and RPA-based libraries (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; see below). Absolute number of reads from fast extracts, however, were constantly below those of Blood \u0026amp; Tissue kit isolates and not all replicates showed all taxa. Over all replicates, only the PCR-based library from fast extracts was missing one \u0026lsquo;rare\u0026rsquo; species, \u003cem\u003eCottus gobio\u003c/em\u003e (Supplementary Material).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eStepwise reamplification with RPA primers allows complete isothermal library preparation.\u003c/b\u003e Initial attempts to amplify fish DNA in isothermal recombinase polymerase amplification with various metabarcoding primers carrying TruSeq adapters for MiSeq libraries or 20 base indexes for MinION libraries failed. Instead, we were able to amplify templates with the longer primers designed in this study for RPA, only when adapters were not included in the primer sequence (both the combination of MiPez-F and MiPez-short-F with MiPez-R showed amplification with more prominent bands in the latter pair) while available metabarcoding PCR primers (from Zhang et al \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) failed to amplify without TruSeq tails except for one combination (Ac12S-F/12S-V5-R). However, some combinations of those with the RPA primers were able to amplify target molecules (i.e. MiFish-U-F/MiPez-R, MiPez-F/MiFish-U-R, MiPez-short-F/MiFish-U-R). Using MiPez-short-F/MiPez-R amplicons as template, we were able to introduce TruSeq adapters in a second and dual indexes in a third reamplification to construct MiSeq libraries with RPA. RPA- and PCR-based libraries were equally able to detect all species from our aquarium mock community. However, we observed slight differences in relative abundance of OTUs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) with \u003cem\u003eGasterosteus aculeatus\u003c/em\u003e in RPA and \u003cem\u003eCyprinus carpio\u003c/em\u003e in PCR being the least abundant taxa respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; Supplementary Material). While RPA detected all species from fast extracts, absolute read counts were generally lower in RPA (Supplementary Material). The long RPA primers almost exclusively amplified fish with few human reads while other vertebrates (mammals, birds) were among the most common reads in Tele02-based PCR libraries. However, in unfiltered MiSeq reads from RPA, short dimers of ~\u0026thinsp;60 bases containing forward and reverse primer and a fragment of the TruSeq adapter were most common, particularly in fast extracts and negative controls. Similarly, using MiPez-short-F/MiPez-R amplicons as template for MinION library preparation, we were able to introduce dual indexes in a second reamplification with RPA. With direct BLAST of quality filtered reads, the RPA-based MinION library was able to detect all species from our aquarium mock community with all 10 species detected from 9 out of 10 filters and one missing a single \u0026lsquo;rare\u0026rsquo; species (\u003cem\u003eG. aculeatus\u003c/em\u003e). Relative species abundance was in line with the RPA-based MiSeq libraries with \u003cem\u003eChondrostoma nasus\u003c/em\u003e being the most and \u003cem\u003eG. aculeatus\u003c/em\u003e the least abundant taxa (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). However, taxonomic resolution was overall lower in MinION reads with up to 7.85% (mean 4.77%) of erroneous BLAST assignments (names of fish not part of the aquarium mock community; Supplementary Material).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eFast extraction and isothermal library preparation for third generation sequencing shows community turnover along a riverine ecotone.\u003c/b\u003e Using the workflow described above in the field with third generation sequencing, we detected a total of ten fish species. Our direct BLAST approach largely assigned names of species known to occur in the river system (HK unpubl. data). However, in several cases, proper taxonomic assignment occurred only on the genus level (i.e. various \u003cem\u003eCottus\u003c/em\u003e spp. and various \u003cem\u003ePhoxinus\u003c/em\u003e spp. detected while only one species occurs in the study area each; (Supplementary Material). \u003cem\u003eCottus\u003c/em\u003e and \u003cem\u003ePhoxinus\u003c/em\u003e were detected in all sampling sites, \u003cem\u003eSalmo trutta\u003c/em\u003e and \u003cem\u003eBarbatula barbatula\u003c/em\u003e occurred in most samples along the entire river course. Particularly in \u003cem\u003eCottus\u003c/em\u003e spp., we observed a decline in read abundance downstream as approximated from sequence blast counts. \u003cem\u003eSqualius cephalus\u003c/em\u003e was exclusively found in the lower sections of the river course and two additional species (\u003cem\u003eAlburnoides bipunctatus\u003c/em\u003e and \u003cem\u003eRutilus rutilus\u003c/em\u003e) were detected only at the confluence with the Mosel river. Other taxa did not show site-specific trends in occurrence (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e; Supplementary Material).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThrough a combination of bead-cellulose-based fast DNA extraction and stepwise RPA library preparation and aided by a complete field laboratory, we developed a metabarcoding workflow that requires little equipment, halves costs and reduces hands-on time more than fourfold in comparison to previous approaches. The cost-effective and rapid assessment of communities is essential for applications such as biodiversity conservation, detection of infectious disease or microbiome research. Exemplified by the successful characterization of a diverse ecotone of central European fish communities and validated by an aquarium mock community, the novel workflow is shown to be as sensitive as currently available protocols.\u003c/p\u003e \u003cp\u003eWhile slightly increasing complexity and hands-on time, the introduction of bead beating and simplified spin columns increased DNA yield from environmental samples in cellulose-based fast extraction (Zou et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). To detect rare taxa, kit-based extraction may still be favorable, while optimized fast extraction is sufficient to approximate community composition. With its low cost, increased numbers of replicates are feasible and can likely in part compensate the lower detection rates of rare taxa. For laboratory-based extraction, we show that multiple treatments can equally preserve sample integrity in the field, with common lysis reagents and silica-gel drying being as effective as costly preservation reagents (Camacho-Sanchez et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Menke et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSo far, isothermal amplification techniques have not been employed for complete preparation of high throughput sequencing libraries for metabarcoding (Magri\u0026ntilde;\u0026aacute; Lobato et al. 2018; Li et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In RPA, this is perhaps due to the non-homologous bases preventing binding of recombinase/primer filaments to template DNA, as the latter is only locally displaced in RPA. This might explain why only stepwise RPA-based reamplification was able to introduce adapters. Despite the additional amplification step, RPA-based libraries did not show increased amplification bias and we demonstrated that isothermal library preparation recovered complete species compositions and highly similar community structure from our aquarium mock community. The slight differences in relative OTU abundance are difficult to assess and may require comparison with direct sequencing of genomic DNA from isolates to approach amplification bias for the two methods (Krehenwinkel et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). We did not observe target sequences in negative controls, suggesting that cross-indexing in RPA pools at ambient temperature is not an issue. The higher specificity of RPA to our target taxon (Teleostei) is likely explained by the longer RPA primers. As metabarcoding primers are generally short but RPA performs best with primers\u0026thinsp;\u0026gt;\u0026thinsp;30 bases, design of novel RPA primers for higher taxa may be hampered by the lack of long conserved regions and they may be more prone to forming artifacts. As such, the reduced number of absolute MiSeq target reads in our RPA libraries compared to PCR is likely attributed to random amplification and dimer production, particularly with little template in environmental samples (Magri\u0026ntilde;\u0026aacute; Lobato et al. 2018; Munawar et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Wu et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The use of self-avoiding molecular recognition systems (SAMRS) in the RPA primers (Li et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) may overcome this risk and size selection with magnetic beads of pooled libraries might be able to reduce short fragments as competitors in sequencing pools to further increase the performance of RPA-based libraries.\u003c/p\u003e \u003cp\u003eWhile our workflow combined with portable third generation sequencing devices allows complete metabarcoding in the field, we present alternatives for optimized sample storage and library preparation for laboratory-based high throughput sequencing. Both allow for integration in currently established metabarcoding workflows. Sequencing with portable Oxford Nanopore platforms is still scarcely used for metabarcoding due to its higher error rates (Chang et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The high similarity in relative abundance as shown here between MiSeq and MinION libraries is particularly promising for even quantitative metabarcoding with Oxford Nanopore platforms. With complete on-site library preparation, MinION sequencing consistently recovered the characteristic community composition of our riverine ecosystem and detected a downstream increase in species richness but decrease of sequence reads from characteristic members of the upper trout region. However, with direct BLAST instead of clustering, assessing species communities deeply relies on near-complete sequencing databases. Future studies may fill these gaps and focus on systematically assessing error rates of MinIon metabarcoding as well as on the design of novel RPA primers amplifying longer fragments (Hatfield et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Huggins et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2024a\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003eb\u003c/span\u003e). RPA has been shown to amplify fragments of up to 1,500 bases (Piepenburg et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), which can increase taxonomic resolution and reduce false positives from direct BLAST in metabarcoding.\u003c/p\u003e \u003cp\u003eOur workflow constitutes a first step towards automated preparation of rapid metabarcoding libraries when integrated into emerging lab-on-a-chip techniques. Particularly in microbiome research where bacterial contamination is a widespread problem, complete on-site processing could help conserve the site specific OTU composition and prevent contamination during transport or with other samples in the lab. Research that is slowed or prevented by sample accessibility, such as sample shipment under the Nagoya protocol, will benefit from on-site diagnostics. In laboratory settings, the novel workflow may still be favored due to its strong reduction of hands-on time and its cost efficiency. Most importantly, we hope our study will democratize metabarcoding by making it more accessible for researchers in developing countries where threatened biodiversity richness peaks and standardized monitoring tools, such as eDNA metabarcoding, are needed more than ever.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eData availability\u003c/h2\u003e \u003cp\u003efor Peer review.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLange V, B\u0026ouml;hme I, Hofmann J, Lang K, Sauter J, Sch\u0026ouml;ne B, Paul P, Albrecht V, Andreas JM, Baier DM, Nething J, Ehninger U, Schwarzelt C, Pingel J, Ehninger G, Schmidt AH (2014) Cost-efficient high-throughput HLA typing by MiSeq amplicon sequencing. BMC Genomics 15:63\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCreer S, Deiner K, Frey S, Porazinska D, Taberlet P, Thomas WK, Potter C, Bik HM (2016) The ecologist\u0026acute;s field guide to sequence-based identification of biodiversity. Methods Ecol Evol 7:1008\u0026ndash;1018\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCompson ZG, McClenaghan B, Singer GAC, Fahner NA, Hajibabaei M (2020) Metabarcoding from microbes to mammals: comprehensive bioassessment on a global scale. Front Ecol Evol 8:581835\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeck F, Blackman RC, Bossart R, Brantschen J, Couton M, H\u0026uuml;rlemann S et al (2022) Meta-analysis shows both congruence and complementarity of DNA and eDNA metabarcoding to traditional methods for biological community assessment. Mol Ecol 31:1820\u0026ndash;1835\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillespie RG, Bik HM, Hickerson MJ, Krehenwinkel H, Overcast I, Rominger AJ (2023) Insigths into ecological \u0026amp; evolutionary processes via community metabarcoding. Mol Ecol 32:6083\u0026ndash;6092\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollen B, Ram M, Zamin T, McRae L (2008) The tropical biodiversity data gap: addressing disparity in global monitoring. Trop Conserv Sci 1:75\u0026ndash;88\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeijzendorffer IR, Regan EC, Pereira HM, Brotons L, Brummitt N, Gavish Y et al (2015) Bridging the gap between biodiversity data and policy reporting needs: An Essential Biodiversity Variables perspective. J Appl Ecol 53:1341\u0026ndash;1350\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaBarre P et al (2011) A simple, inexpensive device for nucleic acid amplification without electricity \u0026ndash; toward instrument-free molecular diagnostics in low-resource settings. PLoS ONE 6:e19738\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrehenwinkel H et al (2019) Nanopore sequencing of long ribosomal DNA amplicons enables portable and simple biodiversity assessments with high phylogenetic resolution across broad taxonomic scale. Gigascience 8:giz006\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePomerantz A et al (2022) Rapid in situ identification of biological specimens via DNA amplicon sequencing using miniaturized laboratory equipment. Nat Protoc 17:1415\u0026ndash;1443\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003edel Rodr\u0026iacute;guez ST, Vanhollebeke M, J., Derycke S (2023) Evaluation of DNA metabarcoding using Oxford Nanopore sequencing for authentication of mixed seafood products. Food Control 145:109388\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOhta A, Nishi K, Hirota K, Matsuo Y (2023) DNA metabarcoding workflow utilizing nanopore long-read sequencing and consensus generation for rapid identification of fungal taxa with high phylogenetic resolution. BioRxiv preprint\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeiner K et al (2018) Optimising the detection of marine taxonomic richness using environmental DNA metabarcoding: the effects of filter material, pore size and extraction method. MBMG 2:e28963\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFukuzawa T, Kameda Y, Nagata H, Nishizawa N, Doi H (2022) Filtration extraction method using a microfluidic channel for measuring environmental DNA. Mol Ecol Res 22:2651\u0026ndash;2661\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR\u0026ouml;der N, Schwenk K (2023) Direct PCR meets high-throughput sequencing \u0026ndash; metabarcoding of chironomid communities without DNA extraction. MBMG 7:e102455\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu X, Comeau AM (1999) Cellulose as a matrix for nucleic acid purification. Anal Biochem 267:418\u0026ndash;420\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZou Y et al (2017) Nucleic acid purification from plants, animals and microbes in under 30 seconds. PLoS Biol 15:e2003916\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZanoli LM, Spoto G (2013) Isothermal Amplification Methods for the Detection of Nucleic Acids in Microfluidic Devices. Biosensors 3:18\u0026ndash;43\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo Y, Xia H, Dai T, Liu T (2023) RPA-CRISPR/Cas12a mediated isothermal amplification for visual detection of \u003cem\u003ePhytophthora sojae\u003c/em\u003e. Front Cell Infect Microbiol 13:1208837\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoenig BD et al (2023) FINDeM: A CRISPR-based, molecular method for rapid, inexpensive and field-deployable organism detection. Methods Ecol Evol 14:3055\u0026ndash;3067\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoenig BD, B\u0026ouml;ning P, Plewnia A (2024) \u0026amp; Richards-Zawacki, C. Using FINDeM for rapid, CRISPR-based detection of the emerging salamandrid fungal pathogen, \u003cem\u003eBatrachochytrium salamandrivorans\u003c/em\u003e. Ecohealth in press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang W, Qu H, Wu X, Shi J, Wang X (2024) Rapid, sensitive, and user-friendly detection of \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e using the RPA/CRISPR/Cas12a system. BMC Infect Dis 24:458\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu T et al (2022) CRISPR/Cas12a-enabled multiplex biosensing strategy via an affordable and visual nylon membrane readout. Adv Sci 10:2204689\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu Z et al (2022) Microfluidic space coding for multiplexed nucleic acid detection via CRISPR-Cas12a and recombinase polymerase amplification. Nat Commun 13:6480\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaporaso JG, Lauber CL, Walters WA, Knight R (2010) Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. PNAS 108:4516\u0026ndash;4522\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiaz T, Shehzad W, Viari A, Pompanon F, Taberlet P, Coissac E (2011) ecoPrimers: inference of new DNA barcode markers from whole genome sequence analysis. Nucleic Acids Res 39:e145\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang S, Zhao J, Yao M (2020) A comprehensive and comparative evaluation of primers for metabarcoding eDNA from fish. Methods Ecol Evol 11:1609\u0026ndash;1625\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCamacho-Sanchez M, Burraco P, Gomez-Mestre I, Leonard JA (2013) Preservation of RNA and DNA from mammal samples under field conditions. Mol Ecol Resour 13:663\u0026ndash;673\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMenke S, Gillingham MAF, Wilhelm K, Sommer S (2017) Home-made cost effective preservation buffer is a better alternative to commercial preservation methods for microbiome research. Front microbiol 8:102\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMagri\u0026ntilde;\u0026aacute; Lobato I, O\u0026rsquo;Sullivan CK (2018) Recombinase polymerase amplification: Basics, applications and recent advances. Trends Anal Chem 98:19\u0026ndash;35\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, Macdonald J, von Stetten F (2019) Review: a comprehensive summary of a decade development of the recombinase polymerase amplification. Analyst 144:31\u0026ndash;67\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrehenwinkel H, Wolf M, Lim JY, Rominger AJ, Simison WB, Gillespie RG (2017) Estimating and mitigating amplification bias in qualitative and quantitative arthropod metabarcoding. Sci Rep 7:17668\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMunawar MA, Martin F, Toljamo A, Kokko H, Oksanen E (2020) RPA-PCR couple: an approach to expedite plant diagnostics and overcome PCR inhibitors. Biotechniques 69:271\u0026ndash;280\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu H, Zhao P, Yang X, Li J, Zhang J, Zhang X, Zeng Z, Dong J, Gao S, Lu C (2020) A recombinase polymerase amplification and lateral flow strip combined method that detects \u003cem\u003eSalmonella enterica\u003c/em\u003e serotype typhimurium with no worry of primer-dependent artifacts. Front Microbiol 11:1015\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang JJM, Ip YCA, Neo WL, Mowe MAD, Jaafar Z, Huang D (2023) Primed and ready: nanopore metabarcoding can now recover highly accurate consensus barcodes that are generally indel-free. BioRxiv \u003cem\u003epreprint\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHatfield RG, Ryder D, Tidy AM, Hartnell DM, Dean KJ, Batista FM (2023) combining nanopore sequencing with recombinase polymerase amplification enables identification of dinoflagellates from the \u003cem\u003eAlexandrium\u003c/em\u003e genus, providing a rapid, field deployable tool. Toxins 15:372\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuggins LG, Colella V, Young ND, Traub RJ (2024a) Metabarcoding using nanopore long-read sequencing for the unbiased characterization of apicomplexan haemoparasites. Mol Ecol Res 24:e13878\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuggins LG, Atapattu U, Young ND, Traub RJ, Colella V (2024b) Development and validation of a long-read metabarcoding platform for the detection of filarial worm pathogens of animals and humans. BMC Microbiol 24:28\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiepenburg O, Williams CH, Stemple DL, Armes (2006) N.A. DNA detection using recombination proteins. PLoS Biol 4:e204\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaberlet P, Bonin A, Zinger L, Coissac E, Environmental (2018) DNA: for biodiversity research and monitoring. Oxford University Press, Oxford, UK, p 252\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGajski D, Wolff JO, Melcher A, Weber S, Prost S, Krehenwinkel H, Kennedy S (2024) Facilitating taxonomy and phylogenetics: An informative and cost-effective protocol integrating long amplicon PCRs and third-generation sequencing. Mol Phylogenet Evol 192:107988\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Kobert K, Flouri T, Stamatakis A (2014) PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics 30:614\u0026ndash;620\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGordon A, Hannon GJ (2010) Fastx-toolkit. FASTQ/A short-reads pre-processing tools. Unpublished Http://Hannonlab.Cshl.Edu/Fastx_Toolkit\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdgar RC (2009) Search and clustering orders of magnitude faster than BLAST. Bioinformatics, 26, 2460\u0026ndash;2461. BMC Bioinformatics 10, 421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCamacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL (2009) BLAST+: architecture and applications. BMC Bioinformatics 15:421\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSch\u0026ouml;neberg Y (2023) yschoeneberg/blast2taxonomy: v1.3.4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.10009721\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.10009721\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Zenodo\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Coster W, Rademakers R (2023) NanoPack2: population-scale evaluation of long-read sequencing data. Bioinform 39:btad311\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSahlin K, Lim MCW, Prost S (2021) NGSpeciesID: DNA barcode and amplicon consensus generation from long-read sequencing data. Ecol Evol 11:1392\u0026ndash;1398\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVierstraete AR, Braeckman BP, Amplicon_sorter: (2022) A tool for reference-free amplicon sorting based on sequence similarity and for building consensus sequences. Ecol Evol 12:e8603\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H, Fran\u0026ccedil;ois R, Henry L, M\u0026uuml;ller K (2023) \u0026amp; Vaughan, D. dplyr: A grammar of data manipulation. R package version 1.1.4\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWickham H (2016) ggplot2: Elegant Graphics for Data Analysis. Springer-, New York\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGu Z, Gu L, Eils R, Schlesner M, Brors B (2014) Circlize implements and enhances circular visualization in R. Bioinform 30:2811\u0026ndash;2812\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Supplementary Materials","content":"\u003cp\u003eSupplementary Materials are not available with this version.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Trier University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"biomonitoring, 3D-printing, environmental DNA, field laboratory, recombinase polymerase amplification, third generation sequencing","lastPublishedDoi":"10.21203/rs.3.rs-5500332/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5500332/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e1. DNA metabarcoding has revolutionized our ability to monitor ecosystems. However, the method is still rarely used in developing countries where resources are limited and fieldwork is challenging.\u003c/p\u003e \u003cp\u003e2. To overcome this, we designed a comprehensive workflow allowing rapid community metabarcoding with minimum, self-manufacturable equipment in the field. We combine cellulose-based fast DNA extraction, optimized for environmental samples, with isothermal recombinase polymerase amplification (RPA). We demonstrate that stepwise RPA-based reamplification allows the introduction of non-homologous sequence adapters and index sequences, allowing the complete isothermal preparation of libraries for high-throughput sequencing.\u003c/p\u003e \u003cp\u003e3. Compared to laboratory-dependent approaches, the novel workflow halves costs and reduces hands on time more than fourfold. Our workflow makes metabarcoding broadly accessible for practitioners, conservationists and researchers without extensive laboratory equipment. Using environmental DNA from fish mock communities and a riverine ecotone, we demonstrate the sensitivity of the workflow and its interconnectivity with existing approaches.\u003c/p\u003e \u003cp\u003e4. Making metabarcoding available to researchers around the globe, our approach signifies a critical contribution to the immense task of characterizing and protecting earth\u0026rsquo;s biodiversity.\u003c/p\u003e","manuscriptTitle":"Towards low-cost and PCR free field-based community metabarcoding","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-11-26 02:50:29","doi":"10.21203/rs.3.rs-5500332/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6e1358b2-ac2a-4a97-8aca-36ea140f5f06","owner":[],"postedDate":"November 26th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-11-26T02:50:29+00:00","versionOfRecord":[],"versionCreatedAt":"2024-11-26 02:50:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5500332","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5500332","identity":"rs-5500332","version":["v1"]},"buildId":"FbvkV6FR0MCFSLy54lSbu","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}