Fig. 1: Reducing SV release probability reduces coding reliability at high stimulus intensities.
From: Homeostatic synaptic plasticity rescues neural coding reliability
A cacRNAi was expressed in ORNs and recordings were made from PNs. PN responses were measured for high odor concentration in 2–4-day-old flies. B Top, Raster plot of PN population odor responses to isopentyl acetate in wt flies (left, n = 72 flies) and cacRNAi in ORNs (right, n = 50 flies). Bottom, the 10 repetitions of a single PN are presented. The shaded area indicates the odor stimulus. C Peristimulus time histogram (PSTH) of PN population responses to five odors (shaded areas represent SEM, odor pulse is labeled with a black bar). Knockdown of cac (green) in ORNs resulted in decreased PN odor responses. (n = 48-72 flies). D Temporal reliability analysis. Pairwise correlations for each odor-neuron combination were pooled across all odors for the data presented in Fig. S1. cacRNAi in ORNs reduces correlation values. E Left, the curve saturation point was calculated for each odor-neuron combination and pooled across all odors for the data presented in Fig. S1. Right, Left curves are presented at a smaller scale. cacRNAi in ORNs shifts the optimal temporal integration window of PNs as evident by a larger integration window spread (left) and a peak shift of ~10 ms (right). Two-sided permutation test, p < 0.001. F Firing-rate reliability analysis for the data presented in Fig. S1. Increased rate variability for cacRNAi flies is observed for high firing rates. Error bands represent SEM. G Example recordings performed in current clamp (IC, black) and in voltage clamp (VC, red). VC trace is inverted for presentation purposes. H PN firing-rate during the first 25 ms of the response to isopentyl acetate in current clamp (IC, black) vs. voltage clamp (VC, red). Each dot represents an individual fly (n = 22). The VC configuration enabled detection of more spikes. Two-sided two sample paired t-test, p = 0.001. I First spike latency of PNs in response to isopentyl acetate in IC (black) vs. VC (red). Dots represent individual flies (n = 22). Two-sided two sample paired t-test, p < 0.001. J First spike latency of PNs in response to the indicated odors for wt and cacRNAi flies (n = 50). Each dot represents the mean first spike latency for 10 trials of a given neuron. Data were obtained in VC and PNs that did not spike within 100 ms after stimulus onset were omitted. K First spike jitter (standard deviation, STD) of PN odor responses, pooled across all odors, for wt and cacRNAi flies (n = 50). Two-sided permutation test, p < 0.001. For all panels ***p < 0.001, for detailed statistical analysis, see Table S1.
