449
A compa a i e analysis o hyb idisa ion cap u e and PCR-
based eDNA me aba coding o moni o ing b yophy es in
ipa ian ecosys ems
Thomas Reinha 1, A mando Espinosa P ie o1, Thomas Begoc1, Hugues Tinguy2, F ancis Bick2,
E ienne Chanez1, Jean-Nicolas Beisel1, Lau en Ha dion1
1 Uni e si y o S asbou g, CNRS, ENGEES, LIVE UMR 7362, 67000 S asbou g, F ance
2 Socié é Bo anique d’Alsace, Séles a , F ance
Co esponding au ho s: Thomas Reinha ([email p o ec ed]); A mando Espinosa P ie o (a mando.espinosa@li e-cn s.unis a. )
Copy igh : © Thomas Reinha e al.
This is an open access a icle dis ibu ed unde
e ms o he C ea i e Commons A ibu ion
License (A ibu ion 4.0 In e na ional – CC BY 4.0).
Resea ch A icle
Abs ac
Despi e hei ecological impo ance, mosses emain unde - ep esen ed in ecological
s udies due o hei challenging de ec ion in ield su eys and mo phological iden i ica-
ion, exace ba ed by he lack o expe bo anis s. In his s udy, we op imise an en i on-
men al DNA me hod o he de ec ion o b yophy es om i e wa e samples, wi h he
aim o acili a ing hei inclusion in biodi e si y assessmen s. We compa ed h ee di e -
en me hods in e ms o species de ec ion and communi y dissimila i y a se en si es
along a i e . The me hods include (i) isual ansec su eys conduc ed by b yologis s
based on mac o- and mic o-mo phology, (ii) mul i-ma ke PCR me aba coding o he bcL
and he ITS2 ma ke s wi h newly designed p ime s a ge ing b yophy es, and (iii) hyb idi-
sa ion cap u e (HC) o he same ma ke s. We ound ha PCR me aba coding eco e ed
mo e han 50% (n = 37) o he species obse ed in he ield, while hyb idiza ion cap u e
de ec ed only 16% (n = 11). PCR me aba coding iden i ied he mos species, 101 species
compa ed o 68 obse ed in he ield and 27 wi h HC. Bo h he PCR and HC me aba cod-
ing app oaches iden i ied b yophy e species no eco ded in ield su eys bu expec ed in
he ca chmen . Molecula me hods, pa icula ly PCR me aba coding, eco e ed a e and
elusi e species di icul o obse e in he ield and occu ing ou side ou ansec . The
wo ma ke s used in he molecula app oaches con ibu ed uniquely o species de ec-
ion, making a mul i-ma ke app oach necessa y o s udy his g oup. En i onmen al DNA
and ield su eys ep esen in eg a i e me hods ha collec i ely enhance de ec ion o
inconspicuous species and yield he mos comp ehensi e species in en o y.
Key wo ds: B yophy e-op imized p ime s, eshwa e ecosys em, ITS2, mosses, bcL,
a ge ed cap u e
In oduc ion
Mosses, including B yophy a, Ma chan iophy a, and An hoce ophy a, play a
c ucial ole in ecosys ems and biodi e si y conse a ion, especially du ing
he i s s eps o ecological succession as pionee species, bu also o nu-
ien cycling (Ay es e al. 2006; Lindo and Gonzalez 2010), soil mois u e
Academic edi o : Hugo de Boe
Recei ed:
9 Ap il 2025
Accep ed:
11 Sep embe 2025
Published:
8 Oc obe 2025
Ci a ion: Reinha T, Espinosa
P ie o A, Begoc T, Tinguy H, Bick
F, Chanez E, Beisel J-N, Ha dion L
(2025) A compa a i e analysis o
hyb idisa ion cap u e and PCR-based
eDNA me aba coding o moni o ing
b yophy es in ipa ian ecosys ems.
Me aba coding and Me agenomics
9: e154548. h ps://doi.o g/10.3897/
mbmg.9.154548
Me aba coding and Me agenomics 9: 449–467 (2025)
DOI: 10.3897/mbmg.9.154548
450
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
(Go nall e al. 2007) and wa e e en ion (Sha a 1997). Mosses con ibu e
signi ican ly o plan biomass ac oss di e se ecosys ems om dese s o
bo eal and a c ic egions. They also play a c ucial ole in o es s by main-
aining soil mois u e and ai humidi y (Lindo and Gonzalez 2010; Eld idge e
al. 2023). They also se e as a sou ce o mic oo ganisms, while p o iding
habi a o a la ge di e si y o mic oalgae, ungi, and mic o auna (Ushe and
Boo h 1986; Su en 1991; Döbbele 1997; Roge Ande son 2006; Kause ud e
al. 2008). Wi h nume ous axa on ed lis s (Bick e al. 2014), b yophy es pose
signi ican conse a ion challenges, u he complica ed by he sca ci y o
skilled axonomis s able o iden i y he di e en b yophy e axa. In ac , hey
a e equally h ea ened as ascula lo a, and global change nega i ely impac s
b yophy e colonies (Vi anen e al. 2024). Howe e , his g oup is unde - ep e-
sen ed in ecological s udies (Callaghan 2012) and biodi e si y conse a ion
policies, pa icula ly because iden i ica ion o he genus and species le els
elies on mic omo phological c i e ia ha equi es a high le el o expe ise
(Tinguy 2021). Species iden i ica ion is challenging insi u, and hei small
size and low biomass can bias ield assessmen s (Callaghan 2012) and in-
c ease eliance on na u alis expe ise, isking misiden i ica ion (G ace 1995).
Molecula iden i ica ion ep esen s an al e na i e me hod ha could acili a e
he inclusion o hese axa in biodi e si y moni o ing p og ams.
The eme ging ield o bo anical eDNA p esen s he nex on ie in eDNA
esea ch, o e ing p omising a enues o biodi e si y s udies (Johnson e
al. 2023). Cu en ly, b yophy es ha e only been conside ed wi hin b oade
mul i axon su eys, whe e only a limi ed numbe o moss axa we e iden i-
ied (Cannon e al. 2016; B unbje g e al. 2019; Ca alho-Sil a e al. 2021;
Bane jee e al. 2022; A iza e al. 2023). The lowe biomass o b yophy es
compa ed o ascula plan s likely esul s in educed eDNA abundance in
he en i onmen , po en ially hinde ing hei de ec ion in eDNA samples when
analysed alongside ascula plan s, a challenge u he exace ba ed by he
use o ma ke s and p ime s designed p ima ily o lowe ing plan s, as in
hese s udies. To da e, he e is no consensus on sui able p ime s and ma k-
e s o b yophy e me aba coding (Liu e al. 2010; Epp e al. 2012; Espinosa
P ie o e al. 2023). The ew s udies a ge ing b yophy e eDNA used ma ke s
wi hin he plas id genome, nL and bcL (Liu e al. 2010; Von C äu lein e
al. 2011; Ballin e al. 2019; Yodphaka e al. 2018; Nelson e al. 2021). Epp
e al. (2012) designed a pai o b yophy es p ime s o ancien DNA s udies
ha ampli ied 50 bp o he nL P6 loop wi h a axonomic esolu ion a he
species le el o 30% om a da abase wi h 4020 species. Howe e , ba code
e e ence da abases exhibi a signi ican gap in sequences o many b yo-
phy e species compa ed o lowe ing plan s (Liu e al. 2010). The e ec i e-
ness o me aba coding depends on he axonomic esolu ion o ma ke s
and he uni e sali y o p ime pai s, ha is, hei capaci y o ampli y a b oad
ange o a ge axa (Liu e al. 2010; Hassel e al. 2013; Cheng e al. 2016).
PCR me aba coding o e ampli ies axa wi h he s onges p ime a ini y
and amplicons wi h he lowe GC con en , esul ing in he ampli ica ion bias
ha compounds o e successi e eac ion cycles (Moina d e al. 2023).
Hyb idisa ion cap u e (HC), also known as a ge ed cap u e, a ge en ich-
men , o cap u e en ichmen , is a p omising al e na i e o PCR-based plan
451
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
eDNA me aba coding (Fos e e al. 2021). Unlike PCR-based me aba cod-
ing, HC ci cum en s ampli ica ion biases, which can lead o alse posi i es
and nega i es, and une en species ep esen a ion (K ehenwinkel e al. 2017;
Kelly e al. 2019). The echnique elies on bio inyla ed RNA o DNA molecules
(bai s), complemen a y o a ge DNA egions o selec i e cap u e a ge se-
quences while emo ing non-hyb idised DNA (Gni ke e al. 2009; Mamano a
e al. 2010; Schuenemann e al. 2011; Ca pen e e al. 2013; Ma ciniak e al.
2015). Hyb idisa ion cap u e imp o es species de ec ion accu acy, pa icula -
ly when a ge ing mul iple loci ac oss genomes (Seebe e al. 2019; Jensen
e al. 2021; Li e al. 2023). This me hod is pa icula ly use ul o deg aded
DNA samples, such as ancien DNA (Mu chie e al. 2021; Re é e e al. 2023),
s ool samples (Aylwa d e al. 2018), and in opical en i onmen s, whe e i
acili a ed he de ec ion o low concen a ed DNA and low biomass and a e
species (Li e al. 2023). Addi ionally, i educes PCR ailu es due o inhibi o s
co-ex ac ed wi h eDNA (Jane e al. 2015; Mu chie e al. 2019).
This s udy aims o op imise eDNA me hods o b yophy e de ec ion, a ax-
onomic g oup unde ep esen ed in ecological esea ch. We compa e h ee
me hods based on species ichness and communi y dissimila i y. The me h-
ods include (i) isual ansec su eys conduc ed by b yologis s based on mac-
o- and mic o-mo phology, (ii) PCR me aba coding o cpDNA bcL and n DNA
ITS2 ma ke s wi h newly designed p ime s, and (iii) a hyb idisa ion cap u e ap-
p oach o he same ma ke s. The o e a ching goal is o acili a e he inclusion
o b yophy es in ecological assessmen s and s udies.
Me hods
Field su ey
Ou compa a i e analysis was ca ied ou a se en si es om a p e ious
eDNA s udy (Espinosa P ie o e al. 2024b) along he Falkens eine bach Ri e
in he Vosges du No d Regional Na u e Pa k (Fig. 1). He e, we expand on
ou p e ious s udy in which wo moss species we e de ec ed as byca ch us-
ing eDNA me aba coding wi h ascula plan p ime s (Fon inalis an ipy e ica
Hedw. and Funa ia hyg ome ica Hedw.). Me aba coding da a (PCR and HC)
o his s udy was gene a ed eusing he eDNA samples o Espinosa P ie o
e al. 2024b collec ed in Oc obe 2020. Moss in en o ies we e conduc ed on
Feb ua y 16 and Ma ch 8 2024 along a single 50-me e -long ansec pa al-
lel o he i e ’s cou se, wi h a 5-me e -wide su ey zone co e ing bo h i e -
banks and he i e bed. The su ey p oceeded om he downs eam o he
ups eam o minimise sedimen dis u bance. B yophy e specimens we e col-
lec ed o species iden i ica ion using mic oscopy. To s eng hen con idence
in ou mo phological iden i ica ions, we used DNA ba coding o e i y mo -
phologically challenging axa. These samples a e p ese ed in he He ba ium
o he Uni e si y o S asbou g and in he p i a e collec ions o F ancis Bick
and Hugues Tinguy. Species a i y o he s udy egion was ob ained om he
assessmen made by Tinguy (2021) and used o iden i y di e ences in he
eco e abili y o a e axa be ween me hods. All axonomic assignmen s in
his s udy ollow he axonomic amewo k o TaxRe 17.0.
452
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Designing and adjus ing p ime pai s o b yophy es
We i s e iewed he li e a u e on plan ba coding and me aba coding o
iden i y sui able p ime s o ampli y he cpDNA bcL gene and he n DNA
ITS2 in B yophy a, Ma chan iophy a, and An hoce ophy a axa. We que ied
he Web o Science and PubMed da abases in Feb ua y 2024 (see Suppl.
ma e ial 1: able S1 o a de ailed lis o p ime s). We pe o med in silico
PCRs using ecoPCR om he OBITools package (Boye e al. 2016) o e alu-
a e 17 p ime pai s o bcL and 21 o ITS2, allowing up o h ee misma ch-
es be ween p ime s and binding si es (Fig. 2; Suppl. ma e ial 1: able S2).
Amplicon leng h was se be ween 100 bp and a ma ke -speci ic maximum:
600 bp o ITS2 and 700 bp o bcL. In he absence o a sa is ac o y p ime
pai o ITS2 in he li e a u e, we designed a p ime pai ha p e e en ially
ampli ies b yophy es: B aF 5’-CGCAAGTTGCGCCCGAGGCT-3’ ( o wa d) and
B aR 5’-GTGATATGCTTAAACTCAGCGGG-3’ ( e e se). We manually explo ed
he alignmen and consensus sequences o all a ailable b yophy e ITS2 se-
quences om he NCBI da abase using Geneious P ime (Do ma ics, Bos on,
USA). By compa ing he consensus sequence wi h ha o ascula plan s, we
iden i ied a sec ion wi hin he lanking egions common o all mosses ha di -
e ed by a leas wo nucleo ides om ascula plan s (Fig. 3). The cumula i e
numbe o misma ches be ween he p ime s and non- a ge axa is known
om p e ious s udies (Espinosa P ie o e al. 2024a; Kol e and Gemeinholze
2021) o esul in p e e en ial ampli ica ion o he a ge g oup. Mo eo e ,
he ITS2 lanking egions a e highly conse ed among S ep ophy a suppo -
ing ou design. Conse a ion o he p iming si es was assessed using he
Figu e 1. Map o he s udy a ea wi hin he Vosges du No d Regional Na u e Pa k, show-
ing he se en si es and he se en eDNA sampling loca ions along he Falkens eine -
bach, 1:150000 scale. The map was c ea ed in QGIS V3.32.2 wi h da a om he Da a-
G andEs Qgis plug-in.
453
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
OBITools commands ‘ecopc . o wa d.shanon’ and ‘ecopc . e e se.shanon’,
and he numbe o misma ches agains mosses and ascula plan s was isu-
alised wi h he command ‘misma chplo ’. We applied s anda d p ime design
p inciples, ensu ing a mel ing empe a u e di e ence wi hin 5 °C be ween
p ime s, p ime leng h be ween 18 and 25 nucleo ides, and we e i ied ha
he mel ing empe a u e o p ime dime s and hai pin s uc u es exceeded
he annealing empe a u e. Fu he mo e, we p io i ised GC clamps on he 3’
end while minimising 3’ end misma ches agains b yophy es and p omo ing
such misma ches agains ascula plan s (Suppl. ma e ial 2: ig. S1). The de-
signed p ime pai was hen subjec ed o he same in silico analyses as he
o he p ime s (Suppl. ma e ial 2: ig. S2).
Ini ial in silico analysis o bcL p ime s om he li e a u e also yielded un-
sa is ac o y esul s. Unlike o ITS2, we did no ind a be e p iming si e ha
allowed o he design o me aba coding p ime s speci ic o B yophy es ollow-
ing he s anda ds abo e. Ins ead, we cus omised p ime s om he li e a u e
(ma ked *) and e un he in silico analysis. Ten p ime s we e iden i ied, gene -
a ing amplicons anging om 107 o 695 bp. Mo e han 95% o he species in
he e e ence da abase we e ampli ied by mos p ime pai s (ba code co e age,
Bc) excep o hose pai s wi h he o wa d 640F o he e e se 804 hR (Fig. 4).
The h ee p ime pai s con aining he o wa d bcL_aF p ime , which gene a ed
Figu e 3. Pie cha depic ing he numbe o misma ches ( anging om ze o o h ee)
on he o wa d p ime BRaF (x-axis) and he e e se p ime BRaR (y-axis) o he ITS2
ma ke in mosses and ascula plan s. The size o each pie ep esen s he numbe o
sequences success ully ampli ied in silico. G ea e misma ch numbe s on ei he p ime
indica e educed PCR e iciency.
Figu e 2. Map o he bes p ime s es ed in silico. P ime s linked wi h a ba we e used
o he ampli ica ion o eDNA samples.
454
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
amplicons o app oxima ely 500 bp, exhibi ed he highes species-le el ba code
speci ici y (Bs) a a ound 75% (Fig. 4; Suppl. ma e ial 1: able S3). In con as ,
p ime pai s p oducing c. 200 bp amplicons, despi e achie ing 95% ba code
co e age (Bc), demons a ed lowe Bs alues o app oxima ely 30%, wi h he
highes speci ici y a ibu ed o p ime pai s wi h he o wa d bcL265 (Fig. 4).
En i onmen al DNA me hods
PCR me aba coding
The PCR mix u es we e p epa ed unde a UV-s e ilised hood, cleaned wi h DNA/
RNA-Exi usPlus™ IF (PanReac AppliChem, Ge many) and wi h a Bunsen bu n-
e o p e en con amina ion o DNA samples om he labo a o y en i onmen .
We used ou new p ime pai B aF + B aR o ITS2 and o he bcL he p im-
e pai bcL265* + bcL556* wi h ou sugges ed modi ica ions shown in bold
5‘-ATYGCTTAYGTTGCTTAYCC-3’ ( o wa d)+ 5‘-CAYTCRTAWACWGCTCTACC-3’
( e e se) (modi ied om Tsubo a e al. 1997; Aziz e al. 2017). PCRs we e p e-
pa ed in iplica e using usion p ime s wi h inline dual indexes o ag each
sample. PCR eac ions we e p epa ed in a inal olume o 25 µL ollowing he
manu ac u e ’s ins uc ions o he GoTaq® G2 Ho S a Mas e Mix (P omega,
Madison, USA), comp ising 12 µL o GoTaq Mmix, 2.5 µL o o wa d p ime , 2.5 µL
o e e se p ime , 5 µL o DNA, and 3 µL o nuclease- ee wa e . Ampli ica ion
was ca ied ou o 35 cycles wi h an ini ial dena u a ion a 95 °C o 3 min, ol-
lowed by 98 °C o 20 s, Ta = 55 °C o bcL and Ta = 65 °C o ITS2 o 15 s,
72 °C o 30 s and 72 °C o he inal elonga ion ime o 1 min. The PCR p od-
uc s we e con olled on a 2% aga ose gel unde UV ligh and DNA concen a ions
Figu e 4. Ba code cha ac e is ics o each bcL p ime pai es ed in silico. The mean am-
plicon leng h (bp) is shown o each p ime pai , wi h ba code co e age (Bc) and ba code
speci ici y (Bs) a e p esen ed as pe cen ages. (*) deno es p ime s modi ied in his s udy
om hei o iginal sou ces; (**) indica es he modi ied e e se p ime M bcL163-R1.
455
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
we e measu ed using he Qubi dsDNA HS Assay Ki (The mo Fishe Scien i ic,
Wal ham, USA). Sequencing was pe o med by Eu o ins Genomics Eu ope
(Kons anz, Ge many) wi h Illumina MiSeq 2x300 bp chemis y, wi h a heo e ical
sequencing dep h o 120000 eads pe sample and can be ound unde he NCBI
biop ojec PRJNA1301602 (h p://www.ncbi.nlm.nih.go /biop ojec /1301602).
Bioin o ma ic analysis was pe o med as in Espinosa P ie o e al. (2024b) adap -
ing he pipeline a ailable in h ps://zenodo.o g/ eco ds/12571296.
Hyb idisa ion cap u e me aba coding
The p esen s udy p ecedes a b oad in-dep h s udy on hyb idisa ion cap u e
(HC) o plan eDNA me aba coding. As such, he HC me hod used he e was
designed as a holis ic mul i-ma ke plan eDNA app oach, including moss-
es. P elimina y analysis o he whole HC da ase showed p omising esul s
o moss de ec ion compa ed o me aba coding wi h uni e sal plan p ime s
(Espinosa P ie o e al. 2024a). Gi en he speci ici y o mosses and he lack o
dedica ed s udies, we used a subse o he HC da ase o ocus on his g oup in
he p esen s udy. Addi ionally, HC was designed o cap u e ou ma ke s ( bcL,
nL, ITS1, and ITS2), bu only he bcL and ITS2 we e conside ed in his s udy
as hey showed he bes p elimina y esul s o mosses. F om he li e a u e,
bcL was al eady used o moss DNA ba coding and had good e e ence da a.
Fo he ITS2 ma ke , ba code speci ici y (Bs) a he species le el was abo e
90% in mos cases acco ding o ou in-silico analysis, meaning i had he bes
axonomic esolu ion (Suppl. ma e ial 1: able S3). We disca ded he ITS1 ma k-
e as i s leng h a ies conside ably be ween moss axa and is no compa ible
wi h Illumina sequencing o pooled samples, which should be wi hin 100 bp
simila in leng h o educe sequencing biases, and would ha e complica ed
he compa ison wi h he PCR me aba coding app oach. The nL showed poo
axonomic esolu ion o mosses compa ed o he o he ma ke s and was ex-
cluded. To sho en he me hod’s sec ion, and because he HC app oach is no
exclusi e o mosses, we p esen ed a de ailed desc ip ion in Suppl. ma e ial 3.
Gi en he na u e o his s udy and he da a, he bai s and he HC aw sequence
da a will only be sha ed in he upcoming s udy. None heless, he eads ha
yielded alignmen s wi h moss axa a e he BLAST can be ound in he Zenodo
eposi o y (h ps://doi.o g/10.5281/zenodo.16738645) oge he wi h he e e -
ence da abase / Accessions o mosses used o he bai design on he bcL
and ITS2 ma ke s and he e e ence da abase used o axonomic assignmen
o bo h PCR and HC eads (Suppl. ma e ial 4, 5).
S a is ics
The ollowing analyses we e pe o med using R e sion 4.1.1 (2021.08.10).
Venn diag ams we e gene a ed using he R package ggVennDiag am ( e sion
1.5.2) o compa e he p opo ion o species eco e ed ac oss me hods and o
iden i y di e ences be ween ma ke s in he wo eDNA app oaches. A P incipal
Coo dina e Analysis (PCoA) based on Sø ensen dissimila i y o communi y com-
posi ion (be a di e si y) a he same si es was conduc ed o iden i y pa e ns in
species de ec ion and assess consis ency ac oss me hods, using he unc ion
egdis om he egan R package ( e sion 2.6-8). Addi ionally, a Pe mu a ional
456
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Mul i a ia e Analysis o Va iance (PERMANOVA) was pe o med on he PCoA
da a o con i m dis inc clus e ing among he g oups using adonis2.We exam-
ined whe he a ia ions in communi y composi ion ac oss me hods could be
a ibu ed o di e ences in hei sensi i i y o de ec ing species wi h dis inc
unc ional ai s, li e o ms, and ecological niches. Speci ically, we assessed
he co ela ion be ween species de ec abili y by eDNA me hods and dispe -
sion ai s, de ined he e as he p opensi y o axa o p opaga e ia ege a i e
p opagules o spo ophy es as a p oxy o highe eDNA abundance, aswell as
hei p e e ence o mois en i onmen s. T ai da a we e ob ained om he
BET da abase (Van Zuijlen e al. 2023). Addi ionally, we conduc ed an indica o
species analysis using he indicspecies package in R o e alua e he likelihood
o species being p e e en ially de ec ed by speci ic me hods. R sc ip s can be
ound in he Zenodo eposi o y (h ps://doi.o g/10.5281/zenodo.16738645).
Resul s
Ac oss all me hods, a o al o 145 species we e de ec ed (see Suppl. ma e ial
1: able S4 o a de ailed lis o axa); howe e , only nine species we e eco e ed
by all me hods (Fig. 5). The ield su ey iden i ied 68 species, 29 o which we e
exclusi ely de ec ed h ough his app oach, including eigh species ha lacked
a e e ence sequence in he da abase. PCR me aba coding eco e ed o e 50%
(n = 37) o he species obse ed in he ield, whe eas hyb idisa ion cap u e (HC)
de ec ed only 16% (n = 11). The la e eco e ed 27 b yophy e species, wi h
an a e age o i e species pe si e. O he nine species only de ec ed by HC
(Fig. 5), wo a e conside ed e y a e in he egion, Dic anella ce icula a (Hedw.)
Schimp. and Riccia huebene iana Lindenb., h ee a e qui e a e, Aneu a pinguis
(L.) Dumo ., Hooke ia lucens (Hedw.) Sm. and T ichocolea omen ella (Eh h.)
Dumo ., h ee ai ly common, F ullania ama isci (L.) Dumo ., He e ocladium
he e op e um (B id.) Schimp. and Physcomi ium py i o me (Hedw.) B uch
& Schimp., and one is common, Didymodon lu idus Ho nsch. Species eco -
e y di e ed be ween he wo ma ke s, ITS2 iden i ied h ee imes mo e spe-
cies han bcL (Fig. 5) and only wo species we e de ec ed by bo h ma ke s
(Fon inalis an ipy e ica Hedw. and P ychos omum capilla e (Hedw.) Holyoak &
N.Pede sen). ITS2 exhibi ed highe axonomic esolu ion han bcL, wi h 21 ou
o 31 unique sequences success ully assigned o a species, compa ed o only
6 ou o 12 o bcL, despi e bo h ma ke s ha ing he same amplicon size. Fo
example, 520 eads o he bcL we e assigned o B yaceae a si e F6, likely
ep esen ing mul iple axa om his amily obse ed in he ield.
PCR me aba coding iden i ied 101 species, including 57 species no eco -
e ed by he o he me hods (Fig. 5). Ra e species we e also de ec ed, o example,
Anomodon ugelii (Müll.Hal.) Keissl. (Bick e al. 2014; Tinguy 2021) was iden i-
ied using he bcL ma ke a si es F4 and F7 (Suppl. ma e ial 1: able S4). Bo h
ma ke s eco e ed a simila numbe o species, wi h bcL de ec ing 67 and ITS2
de ec ing 59 (Fig. 5). Howe e , app oxima ely 40% o he species we e uniquely
de ec ed by ei he ma ke , while only 20% we e sha ed be ween hem. P incipal
Coo dina e Analysis (PCoA) was pe o med o explo e dissimila i y pa e ns be-
ween he h ee me hods. The i s wo p incipal coo dina es (PC1 and PC2) ex-
plained 32.3% and 15.9% o he o al a ia ion (Fig. 6). A dis inc clus e ing o
samples based o he su ey me hod was obse ed, indica ing me hodological
457
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
biases. The PERMANOVA esul s indica ed signi ican di e ences be ween hese
g oups (R2 = 0.898, F = 70.81, p = 0.001), sugges ing ha he model explains a
subs an ial p opo ion o he a iance in he da a. Howe e , ou analysis e u ed
he hypo hesis ha hese biases esul om di e ences in he me hods’ sen-
si i i y o de ec ing species wi h speci ic ai s and ecological p e e ences, as
demons a ed by he ank-sum es on dispe sal ai s and p e e ence o mois
en i onmen s. Ne e heless, he indica o species analysis iden i ied a se o
species ha we e mo e likely o be de ec ed by one me hod o e he o he (see
Suppl. ma e ial 1: able S5 o de ailed lis o species).
Discussion
In his s udy, we compa ed h ee di e en app oaches o s udy b yophy e com-
muni ies in i e ine and ipa ian en i onmen s. Ou s udy is he i s o e alu-
a e hyb idisa ion cap u e and PCR me aba coding using b yophy e-enhanced
p ime s, and o compa e hese o adi ional ield su eys. PCR me aba cod-
ing has been shown o accu a ely cap u e plan species ichness and com-
muni y dissimila i y along ecological g adien s (Shackle on e al. 2019; Reji
Chacko e al. 2023). Howe e , eDNA me aba coding in i e s p oduces sub-
s an ially di e en species lis s han ield su eys, and s udies o en highligh
Figu e 5. Venn diag ams o he p opo ion o species o he whole da ase (7 si es)
de ec ed h ough ield su eys, PCR me aba coding and hyb idisa ion cap u e (HC); and
by each ma ke o he wo eDNA app oaches. To al species ichness (gamma di e si y)
is shown in pa en hesis o each me hod and ma ke .
Figu e 6. P incipal Coo dina e Analysis (PCoA) based on Sø ensen dissimila i y index,
illus a ing di e ences in communi y composi ion. The PCoA compa es communi ies
de ec ed by hyb idisa ion cap u e (HC), PCR me aba coding, and ield su eys ac oss
all si es. PERMANOVA con i med signi ican di e ences be ween me hods (R2 =
0.898, F = 70.81, p = 0.001).
464
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Mamano a L, Co ey AJ, Sco CE, Koza ewa I, Tu ne EH, Kuma A, Howa d E, Shendu e
J, Tu ne DJ (2010) Ta ge -en ichmen s a egies o nex -gene a ion sequencing. Na-
u e Me hods 7(2): 111–118. h ps://doi.o g/10.1038/nme h.1419
Ma ciniak S, Klunk J, De aul A, Enk J, Poina HN (2015) Ancien human genomics: The
me hodology behind econs uc ing e olu iona y pa hways. Jou nal o Human E olu-
ion 79: 21–34. h ps://doi.o g/10.1016/j.jhe ol.2014.11.003
Moina d S, Piau D, Lapo e F, Rioux D, Tabe le P, Goninda d-Melodelima C, Coissac E (2023)
Towa ds quan i a i e DNA Me aba coding: A me hod o o e come PCR ampli ica ion
bias. PCR Bias in me aba coding, 1–31 h ps://doi.o g/10.1101/2023.10.03.560640
Moo house-Gann RJ, Dunn JC, De Ve e N, Gode M, Cole N, Hippe son H, Symondson
WOC (2018) New uni e sal ITS2 p ime s o high- esolu ion he bi o y analyses using
DNA me aba coding in bo h opical and empe a e zones. Scien i ic Repo s 8(1):
8542. h ps://doi.o g/10.1038/s41598-018-26648-2
Mu chie TJ, Kuch M, Duggan AT, Ledge ML, Roche K, Klunk J, Ka pinski E, Hacken-
be ge D, Sadoway T, MacPhee R, F oese D, Poina H (2019) Op imizing ex ac ion
and a ge ed cap u e o ancien en i onmen al DNA o econs uc ing pas en i on-
men s using he PalaeoChip A c ic-1.0 bai -se . Qua e na y Resea ch 99: 305–328.
h ps://doi.o g/10.1017/qua.2020.59
Mu chie TJ, Mon ea h AJ, Mahony ME, Long GS, Cocke S, Sadoway T, Ka pinski E, Za-
zula G, MacPhee RDE, F oese D, Poina HN (2021) Collapse o he mammo h-s eppe
in cen al Yukon as e ealed by ancien en i onmen al DNA. Na u e Communica ions
12(1): 7120. h ps://doi.o g/10.1038/s41467-021-27439-6
Nelson JM, Hause DA, Li F (2021) The di e si y and communi y s uc u e o symbio ic
cyanobac e ia in ho nwo s in e ed om long‐ ead amplicon sequencing. Ame ican
Jou nal o Bo any 108(9): 1731–1744. h ps://doi.o g/10.1002/ajb2.1729
Reji Chacko M, Al e ma F, Fopp F, Guisan A, Keggin T, Lye A, Rey P-L, Richa ds E, Val-
en ini A, Waldock C, Pellissie L (2023) Ca chmen -based sampling o i e eDNA
in eg a es e es ial and aqua ic biodi e si y o alpine landscapes. Oecologia 202(4):
699–713. h ps://doi.o g/10.1007/s00442-023-05428-4
Re é e A, Rijal DP, Hein zman PD, B own AG, S oo -Leichsen ing KR, Alsos IG (2023)
En i onmen al DNA o aqua ic mac ophy es: The po en ial o econs uc ing pas
and p esen ege a ion and en i onmen s. F eshwa e Biology 68(11): 1929–1950.
h ps://doi.o g/10.1111/ wb.14158
Roge Ande son O (2006) The Densi y and Di e si y o Gymnamoebae Associa ed wi h
Te es ial Moss Communi ies (B yophy a: B yopsida) in a No heas e n U.S. Fo es .
The Jou nal o Euka yo ic Mic obiology 53(4): 275–279. h ps://doi.o g/10.1111/
j.1550-7408.2006.00103.x
Schuenemann VJ, Bos K, DeWi e S, Schmedes S, Jamieson J, Mi nik A, Fo es S,
Coombes BK, Wood JW, Ea n DJD, Whi e W, K ause J, Poina HN (2011) Ta ge ed en-
ichmen o ancien pa hogens yielding he pPCP1 plasmid o Ye sinia pes is om ic-
ims o he Black Dea h. P oceedings o he Na ional Academy o Sciences o he Uni -
ed S a es o Ame ica 108(38): E746–E752. h ps://doi.o g/10.1073/pnas.1105107108
Seebe PA, McEwen GK, Löbe U, Fö s e DW, Eas ML, Melzheime J, G eenwood AD
(2019) Te es ial mammal su eillance using hyb idiza ion cap u e o en i onmen-
al DNA om A ican wa e holes. Molecula Ecology Resou ces 19(6): 1486–1496.
h ps://doi.o g/10.1111/1755-0998.13069
Shackle on ME, Rees GN, Wa son G, Campbell C, Nielsen D (2019) En i onmen al DNA
e eals landscape mosaic o we land plan communi ies. Global Ecology and Conse -
a ion 19: e00689. h ps://doi.o g/10.1016/j.gecco.2019.e00689
465
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Sha a BS (1997) The mal conduc i i y and wa e e en ion o a black sp uce o es
loo . Soil Science 162(8): 576–582. h ps://doi.o g/10.1097/00010694-199708000-
00006
Shog en AJ, Tank JL, And uszkiewicz E, Olds B, Mahon AR, Je de CL, Bols e D (2017)
Con ols on eDNA mo emen in s eams: T anspo , Re en ion, and Resuspension.
Scien i ic Repo s 7(1): 5065. h ps://doi.o g/10.1038/s41598-017-05223-1
Su en AM (1991) B yophy es as in e eb a e habi a in wo New Zealand alpine s eams.
F eshwa e Biology 26(3): 399–418. h ps://doi.o g/10.1111/j.1365-2427.1991.
b01407.x
Tinguy H (2021) Ca alogue des b yophy es du Bas Rhin (F ance). Les Nou elles A -
chi es de la Flo e ju assienne e du no d-es de la F ance 19: 117. h ps://cbn c-o i.
o g/especes- ege ales/ca alogue-des-b yophy es-du-bas- hin- ance
Tsubo a H, Nakao N, A ikawa T, Yamaguchi T, Higuchi M, Deguchi H, Seki T (1997) A
p elimina y phylogeny o Hypnales (Musci) as in e ed om chlo oplas bcL se-
quence da a. Biological Resea ch 7: 233–248. h ps://doi.o g/10.24474/b yological-
esea ch.7.8_233
Ushe MB, Boo h RG (1986) A h opod Communi ies in a Ma i ime An a c ic Moss-Tu
Habi a : Mul iple Scales o Pa e n in he Mi es and Collembola. The Jou nal o Ani-
mal Ecology 55(1): 155. h ps://doi.o g/10.2307/4699
Van Zuijlen K, Nobis MP, Hedenäs L, Hodge s N, Calleja Ala cón JA, Albe os B, Be -
nha d ‐Röme mann M, Gab iel R, Ga ille i R, La a F, P es on CD, Simmel J, U mi E,
Bisang I, Be gamini A (2023) B yophy es o Eu ope T ai s (bE ) da a se : A unda-
men al ool o ecological s udies. Jou nal o Vege a ion Science 34(2): e13179.
h ps://doi.o g/10.1111/j s.13179
Vi anen R, Ha pole WS, Dunke S, Eskelinen A (2024) Mul iple global change ac o s
cause declines o a empe a e b yophy e. Plan Ecology & Di e si y 17(1-2): 1–12.
h ps://doi.o g/10.1080/17550874.2024.2330659
Von C äu lein M, Ko pelainen H, Pie iläinen M, Rikkinen J (2011) DNA ba coding: A ool
o imp o ed axon iden i ica ion and de ec ion o species di e si y. Biodi e si y and
Conse a ion 20(2): 373–389. h ps://doi.o g/10.1007/s10531-010-9964-0
Yodphaka S, Boonp agob K, Lumbsch HT, K aichak E (2018) E alua ion o six egions o
hei po en ial as DNA ba codes in epiphyllous li e wo s om Thailand. Applica ions
in Plan Sciences 6(8): e01174. h ps://doi.o g/10.1002/aps3.1174
Supplemen a y ma e ial 1
Tables summa izing p ime de ails, in silico analyses, de ec ed axa, and
indica o species analysis esul s o suppo he indings p esen ed in he
main manusc ip
Au ho s: Thomas Reinha , A mando Espinosa P ie o, Thomas Begoc, Hugues Tinguy,
F ancis Bick, E ienne Chanez, Jean-Nicolas Beisel, Lau en Ha dion
Da a ype: xlsx
Copy igh no ice: This da ase is made a ailable unde he Open Da abase License
(h p://openda acommons.o g/licenses/odbl/1.0/). The Open Da abase License
(ODbL) is a license ag eemen in ended o allow use s o eely sha e, modi y, and
use his Da ase while main aining his same eedom o o he s, p o ided ha he
o iginal sou ce and au ho (s) a e c edi ed.
Link: h ps://doi.o g/10.3897/mbmg.9.154548.suppl1
466
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Supplemen a y ma e ial 2
Figu es illus a ing DNA logoplo s and ba code cha ac e is ics o ITS2
p ime pai es ed in silico
Au ho s: Thomas Reinha , A mando Espinosa P ie o, Thomas Begoc, Hugues Tinguy,
F ancis Bick, E ienne Chanez, Jean-Nicolas Beisel, Lau en Ha dion
Da a ype: pd
Copy igh no ice: This da ase is made a ailable unde he Open Da abase License
(h p://openda acommons.o g/licenses/odbl/1.0/). The Open Da abase License
(ODbL) is a license ag eemen in ended o allow use s o eely sha e, modi y, and
use his Da ase while main aining his same eedom o o he s, p o ided ha he
o iginal sou ce and au ho (s) a e c edi ed.
Link: h ps://doi.o g/10.3897/mbmg.9.154548.suppl2
Supplemen a y ma e ial 3
Me hods o Hyb idisa ion cap u e me aba coding and igu es o ead
abundance h esholds
Au ho s: Thomas Reinha , A mando Espinosa P ie o, Thomas Begoc, Hugues Tinguy,
F ancis Bick, E ienne Chanez, Jean-Nicolas Beisel, Lau en Ha dion
Da a ype: docx
Copy igh no ice: This da ase is made a ailable unde he Open Da abase License
(h p://openda acommons.o g/licenses/odbl/1.0/). The Open Da abase License
(ODbL) is a license ag eemen in ended o allow use s o eely sha e, modi y, and
use his Da ase while main aining his same eedom o o he s, p o ided ha he
o iginal sou ce and au ho (s) a e c edi ed.
Link: h ps://doi.o g/10.3897/mbmg.9.154548.suppl3
Supplemen a y ma e ial 4
K ona plo isualisa ions o he cu a ed bcL e e ence da abases used o
axonomic assignmen and p ime s/bai s design in he s udy
Au ho s: Thomas Reinha , A mando Espinosa P ie o, Thomas Begoc, Hugues Tinguy,
F ancis Bick, E ienne Chanez, Jean-Nicolas Beisel, Lau en Ha dion
Da a ype: h ml
Copy igh no ice: This da ase is made a ailable unde he Open Da abase License
(h p://openda acommons.o g/licenses/odbl/1.0/). The Open Da abase License
(ODbL) is a license ag eemen in ended o allow use s o eely sha e, modi y, and
use his Da ase while main aining his same eedom o o he s, p o ided ha he
o iginal sou ce and au ho (s) a e c edi ed.
Link: h ps://doi.o g/10.3897/mbmg.9.154548.suppl4
467
Me aba coding and Me agenomics 9: 449–467 (2025), DOI: 10.3897/mbmg.9.154548
Thomas Reinha e al.: Compa ing eDNA app oaches o moss de ec ion in lo ic en i onmen s
Supplemen a y ma e ial 5
K ona plo isualisa ions o he cu a ed ITS2 e e ence da abases used o
axonomic assignmen and p ime s/bai s design in he s udy
Au ho s: Thomas Reinha , A mando Espinosa P ie o, Thomas Begoc, Hugues Tinguy,
F ancis Bick, E ienne Chanez, Jean-Nicolas Beisel, Lau en Ha dion
Da a ype: h ml
Copy igh no ice: This da ase is made a ailable unde he Open Da abase License
(h p://openda acommons.o g/licenses/odbl/1.0/). The Open Da abase License
(ODbL) is a license ag eemen in ended o allow use s o eely sha e, modi y, and
use his Da ase while main aining his same eedom o o he s, p o ided ha he
o iginal sou ce and au ho (s) a e c edi ed.
Link: h ps://doi.o g/10.3897/mbmg.9.154548.suppl5
