of 55
1
Structure of a
n
ant
-
myrmecophile
-
microbe community
1
Elena
K.
Perry
1
, Stefanos Siozios
2
, Gregory
D. D.
Hurst
2
and Joseph Parker
1
*
2
1
Division of Biology and Biological Engineering, California Institute of Technology, 1200 E
3
California Boulevard, Pasadena, CA 91125, United States of America
4
2
Institute of Infection, Veterinary and Ecological Sciences, University of Liverpool, Ic2 Liverpool
5
Science Park, 146 Brownlow Hill, Liverpool, L3 5RF, United Kingdom
6
*corresponding autho
r:
joep@caltech.edu
7
Abs
t
r
act
:
S
uper
or
gani
sm
al
ant
col
oni
es
pl
ay
host
t
o
a
m
enager
i
e
of
sym
bi
ot
i
c
ar
t
hr
opods,
8
te
rm
e
d
my
r
me
c
o
p
h
i
l
e
s
,
w
h
ic
h
exhi
bi
t
var
yi
ng
degr
ees
of
soci
al
i
nt
egr
at
i
on
i
nt
o
col
ony
l
i
f
e.
S
uch
9
syst
em
s
per
m
i
t
exam
i
nat
i
on
of
how
ani
m
al
com
m
uni
t
y
i
nt
er
act
i
ons
i
nf
l
uence
m
i
cr
obi
al
10
assem
bl
ages.
H
er
e,
w
e
pr
esent
an
ecol
ogi
cal
l
y
and
phyl
ogenet
i
cal
l
y
com
pr
ehensi
ve
11
char
act
er
i
zat
i
on
of
a
n
ant
-
my
rm
e
c
o
p
h
ile
-
mi
c
r
o
b
e
com
m
uni
t
y
in
S
o
u
th
e
rn
C
a
lifo
rn
ia
.
U
s
in
g
1
6
S
12
rR
N
A
p
ro
filin
g
,
w
e
fin
d
th
a
t
m
ic
ro
b
io
ta
s
o
f
th
e
v
e
lv
e
ty
tre
e
a
n
t
(
Li
om
et
opum
occi
dent
al
e
)
a
n
d
its
13
cohor
t
of
m
yr
m
ecophi
l
es
ar
e
di
st
i
ngui
shed
by
speci
es
-
speci
f
i
c
char
act
er
i
st
i
cs
but
never
t
hel
ess
14
bear
si
gnat
ur
es
of
t
hei
r
behavi
or
al
i
nt
er
act
i
ons.
We
f
o
u
n
d
t
h
a
t
t
h
e
h
o
s
t
a
n
t
mi
c
r
o
b
i
o
me
wa
s
15
di
ver
se
at
al
l
t
axonom
i
c
l
evel
s
;
th
at
of
a
my
r
me
c
o
p
h
i
l
o
u
s
c
ric
k
e
t
wa
s
mo
d
e
r
a
t
e
l
y
d
i
v
e
r
s
e
,
w
h
ile
16
mi
c
r
o
b
i
o
t
a
s
o
f
t
h
r
e
e
my
r
me
c
o
p
h
i
l
o
u
s
r
o
v
e
b
e
e
t
l
e
s
(
S
t
a
p
h
y
l
i
n
i
d
a
e
)
,
wh
i
c
h
h
a
v
e
c
o
n
v
e
r
g
e
n
t
l
y
17
evol
ved
sym
bi
osi
s
w
i
t
h
Li
om
et
opum
,
w
e
re
d
o
m
in
a
te
d
b
y
in
tra
c
e
llu
la
r
endo
sym
bi
ont
s
.
Y
e
t,
18
despi
t
e
t
hese
com
posi
t
i
onal
di
f
f
er
ences,
s
im
ila
rit
ie
s
bet
w
een
ant
and
my
r
me
c
o
p
h
i
l
e
mi
c
r
o
b
i
o
t
a
s
19
cor
r
el
at
ed
w
i
t
h
t
he
nat
ur
e
and
i
nt
i
m
acy
of
t
hei
r
behavi
or
al
r
el
at
i
onshi
p
s
.
P
hysi
cal
i
nt
er
act
i
ons
20
such
as
gr
oom
i
ng
and
t
r
ophal
l
axi
s
lik
e
ly
fa
c
ilita
te
cr
oss
-
speci
es
ext
r
acel
l
ul
ar
mi
c
r
o
b
i
a
l
shar
i
ng.
21
Fur
t
her
,
p
hyl
ogenet
i
c
com
par
i
sons
of
mi
c
r
o
b
i
o
t
a
s
fro
m
my
r
me
c
o
p
h
i
l
e
r
o
v
e
b
e
e
t
l
e
s
a
n
d
22
out
gr
oup
s
re
v
e
a
le
d
a
la
c
k
o
f
c
o
-
cl
adogenesi
s
of
beet
l
es
and
in
tra
c
e
llu
la
r
endosym
bi
ont
s
,
and
23
lim
ite
d
evi
dence
f
or
conver
gence
am
ong
t
he
my
r
me
c
o
p
h
i
l
e
s
'
in
tra
c
e
llu
la
r
mi
c
r
o
b
i
o
t
a
s
.
24
Co
m
pa
r
a
t
i
v
e
g
enom
i
c
anal
ys
e
s
of
th
e
d
o
m
in
a
n
t
Ri
c
k
e
t
t
s
i
a
endosym
bi
ont
of
th
e
m
o
s
t
hi
ghl
y
25
soci
al
l
y
i
nt
egr
at
ed
my
r
me
c
o
p
h
i
l
e
im
p
ly
p
o
s
s
ib
le
fu
n
c
tio
n
s
u
n
re
la
te
d
to
nut
r
i
ent
-
pr
ovi
si
oni
ng
in
26
th
e
h
o
s
t
b
e
e
tle
's
s
p
e
c
ia
liz
e
d
life
s
ty
le
.
O
ur
fin
d
in
g
s
in
d
ic
a
te
th
a
t
my
r
me
c
o
p
h
i
l
e
mi
c
r
o
b
i
o
t
a
s
27
evol
ve
l
ar
gel
y
in
d
e
p
e
n
d
e
n
tly
of
th
e
const
r
ai
nt
s
of
deep
evol
ut
i
onar
y
hi
st
or
y
,
and
th
a
t
th
e
28
tra
n
s
itio
n
to
life
in
s
id
e
col
oni
es,
i
ncl
udi
ng
soci
al
in
te
ra
c
tio
n
s
wi
t
h
h
o
s
t
s
,
pl
ay
s
a
si
gni
f
i
cant
r
ol
e
29
in
s
tru
c
tu
rin
g
bact
er
i
al
assem
bl
ages
of
t
hese
sym
bi
ot
i
c
in
s
e
c
ts
.
30
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
2
Introduction
31
Insects constitute the bulk of global animal biodiversity and form integral components of both
32
terrestrial and freshwater ecosystems
(Scudder, 2017)
. Although many factors have been
33
proposed to underlie insect diversification
(Grimaldi & Engel, 2005)
, studies in numerous taxa
34
have revealed that partnerships with symbiotic
bacteria
have been key to facilitating the
35
exploitation of novel habitat
s
and trophic resources
(Engel & Moran, 2013)
. Insect
-
associated
36
microbes have variously been shown to enable host survival on
recalcitrant food sources
37
(Douglas, 1998; Moran, McCutcheon, &
Nakabachi, 2008; Russell et al., 2009; Salem et al., 2017;
38
Zientz, Dandekar, & Gross, 2004)
, to synthesize chemical defenses
(Piel, 2002)
, to influence
39
communication, mating behavior and reproduction
(Engl & Kaltenpoth, 2018; Wada
-
Katsumata
40
et al., 201
5)
, and
to confer protection against parasites
(Kaltenpoth & Engl, 2014; Koch &
41
Schmid
-
Hempel, 2011; Oliver, Russell, Moran, & Hunter, 2003)
. Yet, despite efforts to
42
characterize the many adaptive (and non
-
adaptive) roles that symbiotic microbe communi
ties
43
play in insect biology, knowledge of the converse relationship
how host ecology shapes the
44
assembly and composition of the microbiome
remains scarce.
45
Both comparative studies, as well as experiments involving a small number of model
46
insect
species, indicate that host microbiot
as can be influenced by
habitat
(Park et al., 2019; Yun
47
et al., 2014)
, diet
(Colman, Toolson, & Takacs
Vesbach, 2012; Majumder et al., 2020; Mason et
48
al., 2020; Renelies
-
Hamilton, Germer, Sillam
-
Dussès,
Bodawatta, & Poulsen, 2021)
,
49
developmental
stage
(Jennings, Korthauer, Hamilton, & Benoit, 2019; Yun et al., 2014)
,
and
50
evolutionary history
(Brooks, Kohl, Brucker, Opstal, & Bordenstein, 2016; R. T. Jones, Sanchez,
51
& Fierer, 2013)
.
Such
studies have typically focused either on a single insect taxon, or have
52
treated multiple insect taxa as separate, non
-
interacting entities. Yet, many of the centr
al roles
53
that insects play within the biosphere involve their interactions with other animal s
pecies
.
Such
54
relationships
from predator
-
prey interactions to parasitic and mutualistic symbioses
are
55
pervasive, and typify the biologies of many speciose insect clades
(Bologna & Pinto, 2001;
56
Feener & Brown, 1997; Godfray, 1994; Hölldobler & Wilson, 1990;
Kathirithamby, 2009; Kistner,
57
1979, 1982; Kovarik & Caterino, 2005; Parker, 2016; Pierce et al., 2002; Stadler & Dixon, 2005)
.
58
The nature of these
interactions has in
many
cases evolved to become obligate, leading to
59
extreme specialization of one insect species
on a single or small number of partner
s
(Beeren et
60
al., 2021; Elmes, Barr, & Thomas, 1999; Hawkins, 1994; Komatsu, Maruyama, & Itino, 2009;
61
López
-
Estrada, Sanmartín,
Uribe, Abalde, & García
-
París, 2021; Maruyama & Parker, 2017;
62
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
3
Strand & Obrycki, 1996)
. How such interspecies dependencies shape the evolution and
63
composition of insect microbiotas remains an open question
.
64
In other animal clades including mammals, growin
g evidence indicates that, in addition
65
to host diet and phylogeny
(Amato et al., 2019; Groussin et al., 2017)
, d
irect social interactions
66
both within and between species can influence host
-
associated microbial communities. In
67
chimpanzees, seasonal increases in social interaction correlate with increased similarity of gut
68
microbiota
(Moeller et al., 2016)
. In baboo
ns, patterns of gut microbiome similarity can be
69
predicted from social networks based on grooming interactions
(Tung et al., 2015)
. Studies in
70
humans have likewise correlated close social relationships and cohabitation with greater
71
similarity of gut or s
kin microbiota
(Dill
-
McFarland et al., 2019; Song et al., 2013)
. Similarly, dog
72
owners share more members of their skin microbiota with their own pets than with other dogs,
73
indicating that contact
-
mediated microbial exchange can occur between species, st
rongly
74
influenc
ing
microbial community structure
(Song et al., 2013)
. These observations imply that the
75
degree of behavioral intimacy between interacting individuals
either within social groups or
76
between species
may be a key parameter shaping animal microbiotas.
77
Here, we ask how the evolution of social behavioral relatio
nships between insect species
78
impacts their symbiotic bacterial communities. To address this question, we exploit a novel
79
system: an ant species that is targeted by a cohort of socially parasitic insects. Many ant
s
play
80
keystone roles in terrestrial ecosys
tems and engage in relationships with diverse other
81
arthropods
(Hölldobler & Wilson, 1990; Parker & Kronauer, 2021)
.
One pervasive mode of
82
interaction occurs within the ant colony itself
a sheltered microhabitat that houses a
83
concentration of resources i
n the form of ant brood and harvested food. Although ruthlessly
84
policed against intruders, colonies are vulnerable to exploitation by an array of specialized
85
arthropods that have evolved ways to evade recognition and gain entry
(Kistner, 1979, 1982;
86
Parker
, 2016)
. Such organisms, termed “myrmecophiles”, are often obligately dependent on a
87
single host ant
species
(Beeren et al., 2021; Elmes et al., 1999; Komatsu et al., 2009; Maruyama
88
& Parker, 2017)
. To infiltrate colonies, myrmecophiles commonly employ
deceptive strategies
89
that permit them to forge intimate relationships with their unknowing hosts. Social integration of
90
myrmecophiles typically hinges on chemical and behavioral adaptations, including mimicry of
91
host ant pheromones (cuticular hydrocarbons
; CHCs)
(Akino, 2002; Bagnères, Blomquist,
92
Bagnères, & Lorenzi, 2010; Beeren et al., 2018; Beeren, Schulz, Hashim, & Witte, 2011; Lenoir
93
et al., 2012; Maruyama, Akino, Hashim, & Komatsu, 2009; Meer & Wojcik, 1982; Parker, 2016)
,
94
or the secretion of so
-
ca
lled “appeasement compounds” that attenuate ant aggression and
95
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
4
foster the myrmecophile’s acceptance into the nest
(Akre & Hill, 1973; Cammaerts, 1992;
96
Hölldobler, 1967, 1970; Hölldobler & Kwapich
, 2019; Jordan, 1913; Parker & Grimaldi, 2014;
97
Stoeffler, Tolasch, & Steidle, 2011)
. Once integrated into
colonies
, myrmecophiles can engage
98
in intimate physical interactions with host
s,
includ
ing
reciprocal grooming, phoresy, and mouth
-
99
to
-
mouth feeding
(trophallaxis)
(Akre & Hill, 1973; Hölldobler, 1971; Hölldobler & Wilson, 1990;
100
Kistner, 1979, 1982; Leschen, 1991; Parker, 2016)
. The obligate nature and extreme closeness
101
of some ant
-
myrmecophile relationships provides a paradigm to explore how intersp
ecies
102
relationships influence microbial community structure in both the host and symbiont insects.
103
Our model ant
-
myrmecophile network centers on the ecologically dominant native ant of
104
Southern California: the velvety tree ant,
Liometopum occidentale
(For
micidae: Dolichoderinae)
105
(
Fig. 1A
)
.
Liometopum
ants form huge colonies numbering
over one
million workers, and patrol
106
sectors of low
-
to mid
-
elevation
oak and pine
forest floor that can span
tens to
hundreds of
107
meters in diameter
(Hoey
-
Chamberlain, Rust, &
Klotz, 2013; Wang, Patel, Vu, & Nonacs, 2010)
.
108
Colonies of this ant house a large menagerie of myrmecophiles with different socially parasitic
109
lif
estyles and integrating strategies
. By virtue of their distinct behaviors, these species permit
110
insight into how different modes of interspecies interaction can impact symbiotic microbial
111
assemblages.
Prominent within
Liometopum
colonies are multiple species of rove beetle
112
(Staphylinidae)
that are obligately associated with this ant, and for which detailed understanding
113
of ethology and chemical ecology has been obtained
(Danoff
-
Burg, 1996)
.
Crucially, each rove
114
beetle species has independently evolved to socially parasitize
Liometopum
col
onies: the
115
species all belong to the same rove beetle subfamily, Aleocharinae (Staphylinidae), but have
116
evolved into myrmecophiles from phylogenetically distant free
-
living ancestors belonging to
117
different taxonomic tribes
. This property permits comparison
s of microbi
ota
between
the
118
convergent myrmecophile
s
and outgroup lineages, potentially illuminating how the evolution of
119
behavioral symbioses shapes the microbiome.
120
In this study, we harness these attributes of the
Liometopum
-
myrmecophile network to
121
address how the social behaviors, evolutionary histories and microhabitats of interacting insect
122
species contribute to structuring their symbiotic microbiotas. By extensive community sampling
123
of core members of the myrmecophile net
work across colonies and localities, we have
124
constructed a comprehensive, quantitative picture of the microbiome of an ant colony and its
125
main social parasite symbionts. Incorporating knowledge of both myrmecophile behavior and
126
microbiome data from outgrou
p relatives of key members of the myrmecophile network, we
127
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
5
present evidence that the intimacy of interspecies social relationships is a major determinant
128
shaping the evolution of insect bacterial communities.
129
Materials and Methods
130
Sample
collection
131
Specimens of
Liometopum
,
Sceptobius
,
Myrmecophilus
,
Platyusa
, and
Liometoxenus
were
132
collected in Southern California in May and July, 2019 (see “Collection localities” below). The
133
specimens were collected into sterile 50 mL conical tubes, with e
ach species in separate tubes,
134
and stored at 4°C until processing later the same day. For dissected samples, specimens were
135
transferred to sterile 1.5 mL microcentrifuge tubes and washed by vortexing at high speed for
136
10 s in 527 μL of sterile TE buffer (1
0 mM Tris
-
HCl, 1 mM EDTA, pH 8) just prior to dissection;
137
the TE buffer fractions were retained as “body wash” samples and stored at
-
80°C. Whole body
138
specimens were washed in the same manner
(except for
Pella
,
Lasius
,
Drusilla
, and
Lissagria
)
,
139
transferred to fresh sterile microcentrifuge tubes, and stored at
-
80°C.
Sceptobius
and
Platyusa
140
were dissected to separate the head and the gut from the rest of the body, while for
Liometopum
141
only the gut was separated from the rest of the body. For
Sce
ptobius
,
due their small body size,
142
t
hree
individuals were pooled per dissected
body part
sample (single individuals were used
in
143
the case of
whole body samples) to ensure sufficient DNA yields due to the small size of the
144
beetles.
The dissections
were
per
formed in sterile Petri dishes using forceps and dissecting
145
Figure 1. The
Liometopum
myrmecophile commun
ity. A:
Workers of the velvety tree ant (
Liometopum occidentale
)
foraging on a tree trunk. Credit: Kim Moore.
B:
Sceptobius lativentris
rove beetle mounted on a
Liometopum
worker,
performing grooming behavior.
C:
Platyusa sonomae
rove beetle exuding appeas
ement secretion from abdominal
gland to a
Liometopum
worker's mouthparts.
D:
Liometoxenus newtonarum
rove beetle with
Liometopum
nearby.
E:
Myrmecophilus cf. manni
cricket interacting with a
Liometopum
worker.
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
6
scissors that were sterilized in between samples with 10% bleach and flaming with ethanol.
146
Dissected body parts were placed in fresh sterile 1.5 mL microcentrifuge tubes and immediately
147
stored at
-
80°C until DNA extraction.
148
Collection localities
149
Post Tree: USA, California, Altadena, Chaney Trail, 34.216293
-
118.145922, 26 vii 2019, coll. E.
150
Perry and J. Parker
151
First Tree: USA, California, Altadena, Chaney Trail, 34.216436
-
118.148452, 26 vii 2019
, coll. E.
152
Perry and J. Parker
153
Rocks Tree: USA, California, Altadena, Chaney Trail, 34.217124
-
118.154182, 26 vii 2019, coll.
154
E. Perry and J. Parker
155
Sleepy Tree: USA, California, Altadena, Chaney Trail, 34.217456
-
118.154426, 26 vii 2019, coll.
156
E. Perry an
d J. Parker
157
Mile High Tree: USA, California, Altadena, Rubio Canyon Trail, 34.2053886,
-
118.1176927, 22 v
158
2019, coll. E. Perry and J. Parker
159
DNA extraction
and
sequencing
160
DNA extractions were performed in batches of 11 samples, with one blank extraction per batch.
161
To minimize the possibility of confounding batch effects with true differences between samples,
162
each batch consisted of a random assortment of different sample ty
pes. To maximize bacterial
163
DNA yield, the extractions were performed using a multistage protocol that involved grinding
164
samples in liquid nitrogen, bead beating, and several rounds of phenol
-
chloroform extraction as
165
described below.
166
Prior to starting extr
actions, 0.1 mm glass beads were sterilized by baking at 280°C for 4
167
hours. Working in a biological safety cabinet to maintain sterility, approximately 100 μL of beads
168
were then transferred into each sample tube using a spatula sterilized with 50% bleach.
After
169
adding the beads, the samples (except for body wash samples) were ground to a fine powder in
170
liquid nitrogen using blunt metal forceps that were sterilized in between samples with 50%
171
bleach and flaming with ethanol. 527 μL of TE buffer was added to
each tube immediately after
172
grinding, except for body wash samples, which already consisted of the same volume of TE
173
buffer; the latter were thawed on ice. Subsequently, 60 μL of 10% sodium dodecyl sulfate (SDS),
174
7.5 μL of proteinase K (20 mg/mL), and 6 μL
of lysozyme (100 mg/mL) were added to each
175
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint
7
sample. The tubes were subjected to bead beating in a Disruptor Genie (
Scientific Industries,
176
Inc.)
for 1 min at 4°C, and then incubated overnight at 37°C with shaking at 250 rpm.
177
The following day, 100 μL of 5
M NaCl was added and mixed thoroughly, followed by the
178
addition of 80 μL of CTAB/NaCL solution (10% cetyltrimethylammonium bromide, 0.7 M NaCl)
179
preheated to 65°C. The samples were incubated at 65°C for 10 min with periodic mixing by
180
inversion. Next, 650 μL
of 25:24:1 phenol/chloroform/isoamyl alcohol (pH 8) was added and
181
mixed thoroughly by inversion, and the samples were spun at 13,000 rpm for 5 min in a
182
microcentrifuge. The aqueous supernatant was transferred to a fresh 1.5 mL microcentrifuge
183
tube and the
extraction was repeated with 700 μL of 24:1 chloroform/isoamyl alcohol. Finally,
184
the aqueous layer was extracted a third time with 500 μL of 24:1 chloroform/isoamyl alcohol. In
185
addition, 650 μL of fresh TE buffer was added to the first round of phenol/chl
oroform/isoamyl
186
alcohol tubes and the entire three
-
step extraction process was repeated, such that each organic
187
fraction was extracted twice. Both aqueous fractions were then combined for each sample in a
188
1.5 mL DNA LoBind microcentrifuge tube (
Eppendorf
)
.
DNA was precipitated overnight at
-
20°C
189
with 0.6 vol isopropanol and 1 μL GenElute linear polyacrylamide (
MilliporeSigma
). Following
190
precipitation, the DNA was pelleted by centrifuging at 13,000 rpm for 30 min at 4°C. The pellets
191
were washed twice with 70
% ethanol, then air dried and resuspended in 25 μL of 10 mM Tris
192
buffer (pH 8). Resuspension was allowed to proceed overnight at 4°C prior to quantification of
193
DNA yield using the Qubit dsDNA HS assay kit (
ThermoFisher Scientific
).
194
16S V4 amplification a
nd Illumina library preparation were performed by Laragen, Inc.,
195
following the protocol recommended by the Earth Microbiome Project
(Ul
-
Hasan et al., 2019)
.
196
For approximately half of the samples (not limited to particular sample types), PCR amplification
197
initially failed. Most of these samples ultimately yielded a detectable gel electrophoresis band
198
upon the inclusion of 100 mg/mL bovine serum albumin in the reaction, or after further sample
199
purification using the DNeasy PowerClean CleanUp kit (
QIAGEN
) an
d/or use of the KAPA3G
200
Plant PCR kit (Roche) instead of Platinum Hot Start PCR Master Mix (
ThermoFisher Scientific
).
201
The full list of sample treatments
(excluding blank samples)
is provided in
Table S1
.
Amplified
202
samples were sequenced on the Illumina MiSe
q platform using the MiSeq v2 300 cycles kit
203
(
Illumina
).
204
16S rRNA gene amplicon sequence processing and curation
205
MiSeq data were processed using the dada2 R package (version 1.14.1)
(Callahan et al., 2016)
206
to perform quality control (trimming and filtering) on sequences, infer exact amplicon sequence
207
(which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this preprint
this version posted October 5, 2021.
;
https://doi.org/10.1101/2021.10.04.462948
doi:
bioRxiv preprint