Mapping the Neural Substrates of Behavior

Abstract

Assigning behavioral functions to neural structures has long been a central goal in neuroscience and is a necessary first step toward a circuit-level understanding of how the brain generates behavior. Here, we map the neural substrates of locomotion and social behaviors for Drosophila melanogaster using automated machine-vision and machine-learning techniques. From videos of 400,000 flies, we quantified the behavioral effects of activating 2,204 genetically targeted populations of neurons. We combined a novel quantification of anatomy with our behavioral analysis to create brain-behavior correlation maps, which are shared as browsable web pages and interactive software. Based on these maps, we generated hypotheses of regions of the brain causally related to sensory processing, locomotor control, courtship, aggression, and sleep. Our maps directly specify genetic tools to target these regions, which we used to identify a small population of neurons with a role in the control of walking.

Comparative single-cell transcriptomics of complete insect nervous systems

“…These techniques have already revealed striking heterogeneity in cell populations that is lost in bulk samples. In the fruit fly, efforts are already well underway to produce connectomic (4)(5)(6)(7)9), activity (12)(13)(14), and behavior atlases (16)(17)) of the nervous system. Much work has separately revealed the role that genes (18)(19) and circuits (20)(21) play in behavior; a major challenge is to combine genes, circuits, and behavior all at once.…”

The structure of behavioral variation within a genotype

“…We finally examined the structure of behavioral variation in collections of flies where variation came from three additional sources: thermogenetic activation of 2,205 sets of neurons across the brain (Robie et al, 2017), optogenetic activation of 176 sparse populations of descending neurons connecting the brain to the motor centers of the ventral nerve cord (Cande et al, 2018), and variation in genotype across 200 inbred strains derived from wild flies (Mackay et al, 2012). Overall, all of these datasets exhibited high degrees of independence in their behavioral measures.…”

“…Lastly, we examined how the correlation structure of behavior compared between sets of flies with variation coming from different sources. Specifically, we looked at four data sets: 1) the BABAM data set (Robie et al, 2017), in which measures were acquired from groups of flies behaving in open arenas, and variation came from the thermogenetic activation of 2,381 different sets of neurons (the first generation FlyLight Gal4 lines (Jenett et al, 2012); 2) a Drosophila Genome Reference Panel (DGRP; (Mackay et al, 2012)) behavioral data set, in which measures were acquired in behavioral assays similar to the Decathlon experiments (sometimes manually, sometimes automatically), and variation came from the natural genetic variation between lines in the DGRP collection; 3) a DGRP physiological data set, in which measures are physiological or metabolic (e.g., body weight and glucose levels) and variation came from the natural genetic variation between lines in the DGRP collection; and 4) the split-Gal4 Descending Neuron (DN) data set (Cande et al, 2018) in which measures came from the same unsupervised cluster approach as Figure 2, and variation comes from the optogenetic activation of specific sets of descending neurons projecting from the brain to the ventral nerve cord . We analyzed these data sets with the same tools we used to characterize the structure of behavioral variation in the Decathlon experiments.…”

Quantifying behavior to solve sensorimotor transformations: advances from worms and flies

“…Rapidly screening so many animals would not have been possible without automated methods. A more recent study expanded this approach to adult Drosophila , automatically cataloging the neural correlates of behaviors using 400,000 flies and targeting 2,204 different neuronal populations for activation [28]. Other studies in adult flies have used similar automated strategies to identify, for example, the connection between specific visual neurons and behavior [29] or that the same subset of neurons can induce different behaviors (courtship vs. aggression), depending on levels of neural activation [30].…”