odol2qs8rsg.html

<div>Overall, despite these limitations, our dataset provides a starting point and a framework from which to build hypotheses and ask questions.&nbsp;</div><div></div><div></div><div></div><div><b>The significance metric/directness etc...</b></div><div>The connectivity matrix obtained is, unsurprisingly, sparser than what the light level anatomy would predict. This has a number of interesting consequences.&nbsp; First, the&nbsp;&nbsp;Δ7 interneurons appear to be a bottleneck for information processed in the protocerebral bridge. This is all the more interesting given the range of responses evoked by&nbsp;&nbsp;Δ7 stimulation (supplementary figure&nbsp;<b>XX</b>). Potentially, synaptic properties at the&nbsp;Δ7 synapses could play a primary role in the way the heading emerges in the EB columnar system and how it is transferred to the FB columnar system. Moreover, it seems that the network contains very few (<b>XX</b> in the pairs tested) excitatory feedforward chains or reciprocal excitatory connections. This raises the question of how activity is maintained in the structure given the wealth of inhibitory inputs it receives. Two candidates are neuronal intrinsic properties -- some types showed robust post stimulus rebounds -- and excitatory inputs in the PB.&nbsp;</div><div>In general, the inputs revealed here are calling for investigation. Whereas some ring neuron subtypes have received some attention (<b>refs</b>), the PB inputs and noduli interneurons are for now unexplored venues. A recent study reported that one of the noduli interneurons -- a likely input to the FB system -- carries a representation of progressive motion in the wasp (<b>latest Heinze ref</b>) .&nbsp;</div><div>The last troubling observation we made is the fact that the output channel of the central complex seems to be very narrow. Unless a lot of outputs have been missed, this would suggest that the information transmitted by the CX is relatively low dimensional (as proposed by some models). Finally, the output region, the LAL, also acts has an input region (via ring neurons or FB-LAL neurons), which points to the LAL as an extra computational layer to the central complex module.</div><div>Optional : network motifs ?</div><div></div><h1 data-label="249564" class="ltx_title_section">Materials and methods</h1><h2 data-label="398564" class="ltx_title_subsection">Fly stocks and crosses</h2><div>For every LexA driver used, we prepared two stocks containing GCaMP6-m <cite class="ltx_cite raw v1">\cite{chen_ultrasensitive_2013}</cite> and CsChrimson <cite class="ltx_cite raw v1">\cite{klapoetke_independent_2014}</cite> under LexAop (resp. UAS) or UAS (resp. LexAop) control : <i>XXX-LexA;13XLexAop2-IVS-p10-GCaMP6m 50.629 in VK00005, 20xUAS-CsChrimson-mCherry-trafficked in su(Hw)attP1</i> and <i>XXX-LexA;20xUAS-IVS-GCaMP6m 15.629 in attP2, 13XLexAop2-CsChrimson-tdTomato in VK00005</i>. Those stocks were then crossed to a Gal4 driver or a split-Gal4 <cite class="ltx_cite raw v1">\cite{luan_refined_2006}</cite> driver for the experiment. In the split-Gal4 case, the two split halves are inserted in attP40 and attP2 respectively. To avoid transvection between the split and the LexA driver <cite class="ltx_cite raw v1">\cite{mellert_transvection_2012}</cite>, we inserted the LexA drivers in alternative sites, either su(Hw)attP5 <cite class="ltx_cite raw v1">\cite{pfeiffer_refinement_2010}</cite>&nbsp;or VK22 <cite class="ltx_cite raw v1">\cite{venken_pacman:_2006}</cite>, and used the splits exclusively in combination with those lines. The list of drivers used and the corresponding cell types are given in Table &nbsp;<span class="au-ref raw v1">\ref{786393}</span>. We generally follow the nomenclature proposed in&nbsp;<cite class="ltx_cite raw v1">\cite{wolff_neuroarchitecture_2015}</cite>.</div><div></div>