of 23
A population of descending neurons that regulate the flight motor of Drosophila
1
Shigehiro Namiki
1,3
, Ivo G. Ros
2
, Carmen Morrow
1
, William J. Rowell
1
, Gwyneth M. Card
*1
,
2
Wyatt Korff
*1
, Michael H. Dickinson
*2,4
3
1
Janelia Research Campus, Howard Hughes Medical
Institute,
19700 Helix Dr., Ashburn, VA
4
20147, USA
5
2
Division of Biology and Bioengineering, California Institute of Technology, 1200 E. California
6
Blvd., Pasadena CA, 91125, USA
7
3
The University of Tokyo, Research Center for Advanced Science and Technology
,
4
-
6
-
1
8
Komaba, Meguro, Tokyo 153
-
8904, Japan
9
4
Lead Contact
10
*
Authors for correspondence:
cardg@janelia.hhmi.org
,
korffw@janelia.hhmi.org
,
11
flyman@caltech.edu
12
13
Summary
14
Like many insect species,
Drosophila melanogaster
are capable of maintaining a stable flight
15
trajectory for periods lasting up to several hours
(1, 2)
. Because aerodynamic torque is roughly
16
proportional to the fifth power of wing length
(3)
, even small asymmetries in wing size require the
17
maintenance of subtle bilateral differences in flapping motion to maintain a stable path. Flies can
18
even fl
y straight after losing half of a wing, a feat they accomplish via very large, sustained
19
kinematic changes to the both damaged and intact wings
(4)
. Thus, the neural network responsible
20
for stable flight must be capable of sustaining fine
-
scaled control over wing motion ac
ross a large
21
dynamic range. In this paper, we describe an unusual
type
of descending neurons (DNg02) that
22
project directly from visual output regions of the brain to the dorsal flight neuropil of the ventral
23
nerve cord. Unlike most descending neurons, wh
ich exist as single bilateral pairs with unique
24
morphology, there is a
population of
at least 15
DNg02 cell pairs
with nearly identical shape. By
25
optogenetically activating different numbers of DNg02 cells, we demonstrate that these neurons
26
regulate wingbeat a
mplitude over a wide dynamic range via a population code
. Using 2
-
photon
27
functional imaging, we show that DNg02 cells are responsive to visual motion during flight in a
28
manner that would make them well suited to continuously regulate bilateral changes in
wing
29
kinematics. Collectively, we have identified a critical set of DNs that provide the sensitivity and
30
dynamic range required for flight control.
31
32
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
Results
33
Within a fly’s nervous system, sensory informat
ion from the brain is conveyed to motor regions of
34
the ventral nerve cord (VNC) by several hundred pairs of descending neurons (DNs) that are
35
roughly stratified into a dorsal pathway that projects to flight motor neuropils and a ventral pathway
36
that projec
t to leg neuromeres
(5, 6)
(Figure 1A, B). Whereas most of the DNs are single pairs of
37
bilateral cells with unique morphology, a small number
of DNs constitute larger sets of
38
homomorphic neurons. To identify classes of DNs that might be involved in flight control, we
39
conducted an activation screen in which we expressed CsChrimson
(
7)
in one GAL4 line and 48
40
split
-
GAL4 lines
(8)
that collectively target 29 different dorsally projecting DNs that innervate the
41
wing and haltere neuropils and tectulum of the VNC
(6)
, along with one control split
-
GAL4 line that
42
does not drive expression in any neuron
(“empty”)
. In each trial,
we aligned tethered, flying flies
43
within a machine vision system to measure wingbeat amplitude
(9)
. To promote stable flight during
44
each trial, we presented the flies with a pattern of vertical stripes presented on an LED array
(10)
45
that covered
212
o
of their frontal field of view. While the
flies regulated the angular velocity of the
46
visual pattern under closed
-
loop conditions, we presented a brief, 100 ms pulse of 617 nm light
47
to activate the targeted DNs (Fig. 1C, D). These 50 lines varied not only with respect to the cells
48
they labeled, bu
t also the sparsity of expression. Nevertheless, a clear pattern emerged when
49
comparing the results across lines (Fig. 1E). Of the 13 lines in which CsChrimson activation
50
resulted in the largest change in wingbeat amplitude, 11 targeted members of the same
class of
51
neurons, DNg02.
52
53
The DNg02s
were previously identified
anatomically
(6)
and
consist o
f a cluster of at least 15 cell
54
pairs
with nearly identical morphology, the largest of the population
-
typ
e class of DNs identified
55
so far
. Their small, spindly cell bodies reside in a cluster at the ventral edge of the gnathal
ganglion
56
(GNG; Figure 2A). The primary neurites of the DNg02s run ventrally along the edge of the GNG
57
before taking a hairpin turn and ascending dorsally, where each cell arborizes in a hemi
-
circle
58
around the esophageal foramen. The cells’ terminals reside
within a set of five contiguous
59
neuropils consisting of the inferior bridge (IB), inferior clamp (ICL), superior posterior slopes
60
(SPS), inferior posterior slope (IPS), as well as the gnathal ganglion (GNG)
(11)
. Synaptotagmin
61
labeling and fine morphology suggest that processes within the IB and GNG are
outputs, whereas
62
those within the IC, SPS, and IPS are inputs
(6)
.
63
64
After descending ipsilaterally down the neck connectives, the DNg02 cells exhibit a distinct pattern
65
of arborization in the dorsal VNC (Figure 2B). Collectively, the population of cells forms a compact
66
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
‘figure
-
of
-
eight’ shape within the dorsal flig
ht neuropil
a pattern also repeated in the haltere
67
neuropil (Figure 2F). Visualization of the morphology of individual neurons using the multi
-
color
68
flip out (MCFO) expression system
(12)
indicates that the arborizations of
DNg02s remain
69
restricted to the ipsilateral side within the brain, whereas the terminals in the VNC cross the
70
midsagittal plane (Figure 2A,B). Large synaptotagmin
-
positive boutons are distributed diffusely
71
throughout the projections in the wing and halter
e neuropils (Figure 2C
-
E), consistent with output
72
synapses in these regions.
73
74
The large number of DNg02 cells suggests that they might underlie some unique and critical
75
function within the flight motor system. One hypothesis is that the DNg02s act on wing
motion via
76
a population code
(13, 14)
, such
that the precise kinematic output of the wings depends in part on
77
the number of DNg02 cells that are active as well as the level of activity within individual cells.
To
78
test this hypothesis, we used 13 separate split
-
GAL4 lines that targeted different sub
sets of
79
DNg02 cells with little or no expression in off
-
target neurons (Figure 2F). In addition, we evaluated
80
one GAL4 line (GMR42
B02) that targets 15 DNg02 cell pa
irs
. As a control, we tested
the empty
81
split
-
GAL4 line (SS03500). Because the number of DNg02 ce
ll pairs labeled in these 14 lines
82
varied from 0 to 15, we were able to test the influence of population activity by driving each line
83
independently and comparing the magnitude of the effect on wingbeat amplitude during flight. As
84
shown in Figure 3A, we fo
und a strongly linear relationship (r
2
= 0.7394) between the number of
85
DNg02 cells present in each line and the magnitude of the change in wingbeat amplitude elicited
86
by activation.
87
88
We further explored the effects of optogenetic activation on a subset of
driver lines, focusing on
89
SS02625, a line that targets 8 DNg02 cells (Figure 2F
)
. Whereas individual flies varied with
90
respect to their background level of wingbeat amplitude prior to optogenetic activation (Figure 3B,
91
top traces), the peak level of wingbe
at amplitude typically reached the same approximate value
92
for each fly during activation. This result suggests that the level of activation we applied
eli
c
i
ted
a
93
saturating excitation of the DNg02 neurons within this line. In such cases, we also observe
d an
94
intriguing pattern of changes in wingbeat frequency elicited by the excitation (Figure 3B, bottom
95
traces). In most cases, DNg02 activation resulted in a correlated rise in both wingbeat amplitude
96
and frequency, but in some instances the increase in am
plitude was accompanied by a net
97
decrease in frequency. Whether activation elicited a decrease or increase in wingbeat frequency
98
was not random, but rather depended on the level of wingbeat frequency prior to activation. In
99
particular, individuals that fle
w with a lower wingbeat frequency exhibited a large increase in
100
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
frequency upon activation, whereas those that flew with a higher wingbeat frequency exhibited a
101
decrease (Figure 3C). A likely explanation for this peculiar trend emerges from plotting
102
instant
aneous wingbeat frequency against wingbeat amplitude throughout the time course of
103
optogenetic activation for all the trials of an individual fly (Figure 3C). In each trial, CsChrimson
104
activation evoked an initial rapid rise in both wingbeat amplitude and
frequency; however, once
105
the mean wingbeat amplitude of the two wings reached a value of about 160
o
, further increases
106
in amplitude were accompanied by a small decline in frequency (see also traces in bottom panel
107
of Figure 3B). Thus, when the DNg02 cells
within the fly were maximally activated, we observed
108
an inverse relationship between wingbeat frequency and amplitude.
109
110
The time
-
varying relationship between wingbeat amplitude and frequency (Figure 3D) bear a
111
striking resemblance to data presented in a fo
rmer study on the metabolic power requirements
112
for flight
(15)
, in which changes in wingbeat amplitude and frequency were elicited by upward and
113
downward visual motion rather than opto
genetic activation of DNs.
T
hat study
presented a model
114
in which the mass specific mechanical power (
P
*
mech
) delivered by the flight muscles sustains the
115
sum of induced power (the cost of lift) and profile power (the cost of drag) during fl
ight. Profile
116
power, which is the dominant term, is proportional to the product of the wingbeat frequency and
117
amplitude cubed
(3)
. Thus, their model predicts that when the mechanical power generated by the
118
flight muscles is constant
as it is when the asynchronous flight muscles are maximally
119
activated
any increase in wingbeat amplitude must be accompanied by a
decrease in frequency
120
and vice versa. To
analyze whether
DNg02 activation may elicit near maximal power output, we
121
superimposed isolines for
mechanical power in the frequency
-
amplitude plane, using equations
122
from the prior study
(15)
. The precise values of these isolines are only approximate, because they
123
are based on average morphometric data for wing length, wing mass, and body mass of the flies
124
used in the prior study
(15)
; we did not take those measurements on the flies used in our
125
experiments. However, the salient observation is that the set of amplitude
-
frequency values
126
elicited during peak DNg02 activation ar
e bounded by the shape of the power isolines (e.g.
P
*
mech
~
127
105 W kg
-
1
), which thus enforces the observed inverse relationship between wingbeat frequency
128
and wingbeat amplitude. These data suggest that optogenetic activation of this particular driver
129
line (
SS02625) results in the production of peak mechanical power, presumably via activation of
130
the motor neurons of the large indirect flight muscles.
131
132
So far, our activation experiments suggest that the
population of
DNg02 cells might function
133
together
to re
gulate flight power like a throttle, by controlling wingbeat amplitude in a bilateral
134
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
fashion via symmetric activation of power and steering muscle motor neurons. However, this
135
result may simply reflect the fact that optogenetic activation
simultaneously
e
xcites both the left
136
and right DNs in each fly. To test if left and right DNg02 cells might operate independently
during
137
steering maneuvers in flight
, we performed 2
-
photon functional imaging from the dendritic region
138
of the neurons in one of the DNg02 dri
ver lines (SS02535) in tethered flying flies using GCaMP6f
139
as an activity indicator (Figure 4A). In preparing the flies for recording, we dissected a window in
140
the head capsule just dorsal to the esophageal foramen, which allowed us to image neurons on
141
the
left and right side of the brain simultaneously (Figure 4B). Because our goal was to test
142
whether left and right DNg02 cells might be active independently, we subjected the flies to an
143
array of different visual patterns during flight, chosen to elicit bot
h symmetrical and asymmetrical
144
wingbeat responses. These stimulus epochs included a widefield pattern that moved upward or
145
downward, yaw motion to the left or right, a stripe oscillating on the left or right, roll motion to the
146
left or right, an expanding
object on the left or right, progressive and regressive motion, and closed
147
loop stripe fixation (Figure 4C). The net results from this array of visual patterns was consistent
148
in that they demonstrated unequivocally that at least some DNg02 cells can respon
d differently
149
on the left and right sides of the brain. This result was most apparent in yaw stimuli (Figure 4D,
150
left traces), during which rightward motion elicited an increase in activity of the right DNg02 cells
151
and a simultaneous decrease in activity o
f the left DNg02 cells (and vice versa for leftward
152
motion). The changes in cell fluorescence were accompanied by the expected asymmetric
153
changes in win
g
beat amplitude for a yaw response. In contrast, bilaterally symmetrical visual
154
patterns, such as regres
sive visual motion, elicited synchronous changes in activity of the right
155
and left DNg02 cells, accompanied by symmetrical changes in wingbeat amplitude (Figure 4D,
156
rightmost traces). These results indicate that the DNg02 cells can operate independently on
the
157
lef
t and right side of the brain i
n an asymmetrical or symmetrical fashion, depending on the pattern
158
of visual input. Across all recordings, we measured a strong correlation between the DNg02 cells
159
and the wingbeat amplitude of the contralateral wing,
and a weaker anti
-
correlation with the
160
wingbeat amplitude of the ipsilateral wing (Figure 4E). These patterns were readily apparent in
161
individual recordings, in which we determined the correlation coefficient between changes in
162
fluorescence (
F/F) and win
gbeat amplitude for each pixel in the
fluorescence
image during a 20
163
second flight epoch (Figure 4F). A pixel pattern that corresponds to the arbor of the right DNg02
164
cells was highly correlated with left wingbeat amplitude, whereas a pixel pattern corresp
onding to
165
the left DNg02 cells was highly correlated with right wingbeat amplitude.
166
167
168
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
Discussion
169
Compared to birds, bats, and pterosaurs
the three other groups of organisms capable of
170
sustained active flight
a unique feature of insects is that their win
gs are novel structures that are
171
not modified from prior ambulatory appendages. Insects retained the six legs of their apterogote
172
ancestors, but added two pairs of more dorsally positioned wings
(16)
. This evolutionary quirk has
173
profound consequences for the underlying neuroanatomy of the insect flight system. Within their
174
thoracic ganglia, the sensory
-
motor
neuropil associated with the wings constitutes a thin, dorsal
175
layer sitting atop the larger ventral regions that control leg motion
(6, 17, 18)
. Numerically,
176
however, there appear to be more DNs targeting the flight neuropil than targeting the leg
177
neuromeres
(6)
. This is surprising, given the more ancient status of the leg motor system and the
178
importance of legs in so many essential behaviors. However, the relatively large number of flight
179
DNs may reflect th
e
fact th
at
control of flight requires greater motion precision because even
180
minute changes in wing motion have large consequences on the resulting aerodynamics
(19)
. In
181
this paper, we describe a class of DNs in
Drosophila
(DNg02) that are unusual in that instead of
182
existing a
s a unique bilateral pair, they constitute a large homomorphic population. By
183
optogenetically driving different numbers of cells, we demonstrated that DNg02 cells can regulate
184
wingbeat amplitude over a wide dynamic range (Figure 3A)
and can elicit maximum
power output
185
from the flight motor (Figure 3D). Using 2
-
photon functional imaging, we also show that at least
186
some DNg02 cells are responsive to large field visual motion during flight in a manner that would
187
make them well suited for continuously regulati
ng wing motion in response to both bilaterally
188
symmetrical and bilaterally asymmetrical patterns of optic flow (Figure 4D).
189
190
Straight flight in
Drosophila
is only possible due to the maintenance of subtle and constant bilateral
191
differences in wing motion,
carefully regulated by feedback from sensory structures such as the
192
eyes
(20, 21)
, antennae
(22, 23)
, and halteres
(24, 25)
. The control system necessary for straight
193
flight must permit the
maintenance of very large, yet finely regulated, distortions of wing motion in
194
order to produce perfectly balanced forces and moments. One means of controlling fine
-
scaled
195
sensitivity over a large dynamic range is through the use of a population code with
range
196
fractionation. The use of a population code to specify motor output is a general principle
(14)
that
197
has been
observed
in a wi
de array of species including leech
es
(26)
, crickets
(27)
,
198
cockroaches
(28)
, and monkeys
(29)
. In dragonflies, 8 pairs of DNs
a group of cells roughly
199
comparable in number to the DNg02 cells
pro
ject to the flight neuropil and encode the direction
200
to small visual targets
(30)
.
201
202
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
Although the DNg02 neurons are morphologically simil
ar, we strongly suspect that the population
203
is not functionally homogeneous. To fly straight with perfect aerodynamic trim, an animal needs
204
to zero its angular velocity about the yaw, pitch, and roll axes, in addition to regulating its forward
205
flight speed
, side slip, and elevation. Thus, if the DNg02 cells are the main means by which flies
206
achieve flight trim, one would expect that they would be organized into several functional
207
subpopulations, with each set of cells controlling a different degree of freed
om of the flight motor
208
system. For example, one subpopulation of DNg02 cells might be primarily responsible for
209
regulating roll, while another is responsible for regulating pitch, and yet another regulates forward
210
thrust. Such subpopulations need not const
itute exclusive sets, but rather might overlap in
211
function, collectively operating like a joystick to regulate flight pose. If this hypothesis is correct,
212
we would expect the DNg02 neurons to differ with respect to both upstream inputs from
213
directionally t
uned visual interneurons as well as downstream outputs to power and steering
214
muscle motor neurons. Unfortunately, we could not distinguish individual cell types across the
215
different driver lines we used at the level of light
-
based microscopy. If DNg02 cell
s are further
216
stratified into subclasses, it is likely that each driver line targets a different mixture of cell types.
217
Indeed, the variation we observed in changes in wingbeat amplitude as a function of the number
218
of DNg02 cells activated (Figure 3E) migh
t reflect this variation in the exact complement of cells
219
targeted by the different driver lines.
F
urth
er,
a
lthough one driver line (R42B02) targets 15 DNg02
220
neurons, it is likely that this number underestimates the size of the entire population
, and
we
221
speculate that there
may be
a small
set of
neurons
dedicated to regulating each output degree of
222
freedom. Collectively, our results suggest that we have identified a critical component of the
223
sensory motor pathway for flight control in
D
rosophila,
the precise organization of which is now
224
available for further study using a combination of genetic, physiological, and connectomic
225
approaches.
226
227
228
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
ACKNOWLEDGMENTS
229
A portion of this work was conducted as part of the Descending Interneuron Project
Team at
230
Janelia Research Campus. We would like to thank the Janelia Visiting Science Program for
231
hosting MHD
,
Gudrun Ihrke and The Project Technical Resources group for assistance in
232
coordinating the screening, and the Janelia FlyCore assisted with anima
l preparations. Research
233
reported in this publication was supported by the Howard Hughes Medical Institute (S.N., W.J.R.,
234
G.M.C., W.K.) and the National Institute of Neurological Disorders and Stroke of the National
235
Institutes of Health (I.G.R., M.H.D.) un
der Award U19NS104655.
236
237
AUTHOR
CONTRIBUTIONS
238
(Following
CRediT taxonomy
): Conceptualization: S.N., G.M.C., W.K, and M.H.D.; Methodology:
239
S.N., G.M.C., W.K
.
, and M.H
.D.; Software: I.G.R., W.J.R.; Validation: W.J.R., C.M
.
, W.K., G.M.C.,
240
I.G.R., M.H.D.; Formal Analysis:
S
.N
.
, I.G.R., C.M
.
, W.J.R., W.K., M.H.D.; Investigation:
S
.N
.
,
241
I.G.R., W.J.R.; Resources: G.M.C. W.K., M.H.D.; Data Curation: C.M., W.J.R., I.G.R.; Wr
iting
242
(Original Draft): S.N., M.H.D.; Writing (Review & Editing): W.K., I.G.R., M.H.D.; Visualization:
243
S.N., I.G.R., C.M., W.K
.
, M.H.D.; Funding Acquisition: G.M.C., W.K
.
, M.H.D.; Supervision:
244
G.M.C., W.K
.
, M.H.D.;
Project Administration:
G.M.C., W.K
.
, M.H
.D.
245
246
DECLARATION OF INTERESTS
247
The authors have no competing interest to declare.
248
249
STAR METHODS
250
251
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
KEY RESOURCES TABLE
252
REAGENT or RESOURCE
SOURCE
IDENTIFIER
Chemicals, Peptides, and Recombinant Proteins
All
-
trans
-
retinal
Sigma
-
Aldrich
CAS: 116
-
31
-
4
Schnei
der’s Insect Medium
Sigma
-
Aldrich
S0146
Triton X
-
100
Sigma
-
Aldrich
X100
Xylene
Thermo Fisher Scientific
x5
-
500
Dibutyl phthalate in xylene (DPX)
Electron Microscopy Sciences
13512
Paraformaldehyde
Electron Microscopy Sciences
15713
-
S
Antibodies
Mouse
mAb anti
-
Bruchpilot (nc82)
Developmental Studies
Hybridoma Bank
nc82; RRID: AB_2314866
rabbit polyclonal anti
-
GFP
Thermo Fisher Scientific
Cat #: A
-
11122; RRID: AB_221569
Alexa Fluor 488 goat anti
-
rabbit
Thermo Fisher Scientific
Cat #: A
-
11034; RRID: AB
_2576217
Alexa Fluor 568 goat anti
-
mouse
Thermo Fisher Scientific
Cat #: A
-
11031; RRID: AB_144696
Deposited Data
Raw and analyzed data
This paper
https://doi.org/10.17632/7g984jm2zc.1
Experimenta
l Models: Organisms/Strains
D. melanogaster: UAS
-
CsChrimson
Bloomington Drosophila Stock
Center
RRID:BDSC_55135
D. melanogaster: UAS
-
OpGCaMP6f
(
20XUAS
-
IVS
-
Syn21
-
OpGCamp6F
-
p10
in
attP5
)
Gift from D. Anderson
N/A
D. melanogaster: UAS
-
tdTomato
(
P{w[+mC]=UA
S
-
tdTom.S}3
)
Bloomington
RRID:BDSC_36328
D. melanogaster: UAS
-
OpGCaMP6f
;
UAS
-
tdTomato
Constructed from above two
lines
N/A
D. melanogaster: pJFRC200
-
10XUASIVS
-
myr::smGFP
-
HA in
attP18
[
(31)
]
N/A
D. melanogaster:
SS01074
Bloomington Drosophila Stock
Center
RRID:BDSC_75840
D. melanogaster:
SS01063
Bloomington Drosophila Stock
Center
RRID:BDSC_75837
D. melanogaster:
SS00735
Bloomington Drosophila Stock
Center
RRI
D:BDSC_75998
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
D. melanogaster:
SS01540
Bloomington Drosophila Stock
Center
RRID:BDSC_75903
D. melanogaster:
SS01052
Bloomington Drosophila Stock
Center
RRID:BDSC_75825
D. melanogaster:
SS01558
Bloomington Drosophila Stock
Center
RRID:BDSC_75946
D. melan
ogaster:
SS02377
Bloomington Drosophila Stock
Center
RRID:BDSC_75874
D. melanogaster:
SS02384
Bloomington Drosophila Stock
Center
RRID:BDSC_75958
D. melanogaster:
SS01053
Bloomington Drosophila Stock
Center
RRID:BDSC_86728
D. melanogaster:
SS02631
Bloom
ington Drosophila Stock
Center
RRID:BDSC_75976
D. melanogaster:
SS02536
Bloomington Drosophila Stock
Center
RRID:BDSC_75940
D. melanogaster:
SS01069
Bloomington Drosophila Stock
Center
RRID:BDSC_75828
D. melanogaster:
SS02542
Bloomington Drosophila Stoc
k
Center
RRID:BDSC_75941
D. melanogaster:
SS01546
Bloomington Drosophila Stock
Center
RRID:BDSC_75944
D. melanogaster:
SS02392
Bloomington Drosophila Stock
Center
RRID:BDSC_75878
D. melanogaster:
SS1075
Bloomington Drosophila Stock
Center
RRID:BDSC_7584
1
D. melanogaster:
SS01056
Bloomington Drosophila Stock
Center
RRID:BDSC_75818
D. melanogaster:
SS01556
Bloomington Drosophila Stock
Center
RRID:BDSC_75953
D. melanogaster:
SS02393
Bloomington Drosophila Stock
Center
RRID:BDSC_75933
D. melanogaster:
SS02608
Bloomington Drosophila Stock
Center
RRID:BDSC_75966
D. melanogaster:
SS01058
[
(6)
]
N/A
D. melanogaster:
SS02383
Bloomington Drosophila Stock
Center
RRID:BDSC_75888
D. melanogaster:
SS02552
Bloomington Drosophila Stock
Center
RRID:BDSC_75942
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint
D. melanogaster:
SS02379
Bloomington Drosophila Stock
Center
RR
ID:BDSC_75963
D. melanogaster:
SS02635
Bloomington Drosophila Stock
Center
RRID:BDSC_75969
D. melanogaster:
SS03500
This paper
N/A
D. melanogaster:
SS02111
Bloomington Drosophila Stock
Center
RRID:BDSC_75935
D. melanogaster:
SS02396
Bloomington Drosoph
ila Stock
Center
RRID:BDSC_75882
D. melanogaster:
SS01579
Bloomington Drosophila Stock
Center
RRID:BDSC_75902
D. melanogaster:
SS02634
Bloomington Drosophila Stock
Center
RRID:BDSC_75970
D. melanogaster:
SS02617
Bloomington Drosophila Stock
Center
RRID:
BDSC_75967
D. melanogaster:
SS01061
Bloomington Drosophila Stock
Center
RRID:BDSC_75836
D. melanogaster:
SS02551
This paper
N/A
D. melanogaster:
SS02553
Bloomington Drosophila Stock
Center
RRID:BDSC_75936
D. melanogaster:
SS02627
This paper
N/A
D. mel
anogaster:
SS01049
Bloomington Drosophila Stock
Center
RRID:BDSC_75833
D. melanogaster:
SS01541
Bloomington Drosophila Stock
Center
RRID:BDSC_75891
D. melanogaster:
SS01577
This paper
N/A
D. melanogaster:
SS01578
This paper
N/A
D. melanogaster:
SS01560
Bloomington Drosophila Stock
Center
RRID:BDSC_75948
D. melanogaster:
SS02535
This paper
N/A
D. melanogaster:
SS01073
Bloomington Drosophila Stock
Center
RRID:BDSC_75839
D. melanogaster:
SS02625
Bloomington Drosophila Stock
Center
RRID:BDSC_75974
D.
melanogaster:
SS01563
This paper
N/A
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted August 6, 2021.
;
https://doi.org/10.1101/2021.08.05.455281
doi:
bioRxiv preprint