Design, Engineering, Specified Complexity: Appreciating the Fruit Fly Brain
Groundbreaking new research has documented the complexity and design of the brains of fruit flies (Drosphila melanogaster). Many of the results were published in a series of papers in the journal Nature. The basis for the research is the completion of the entire wiring diagram (called a connectome) of the fruit fly brain, which consists of 140,000 neurons.1 In addition, it includes more than 50 million connections (chemical synapses).2Keep in mind that, despite the number of neurons and connections, fruit fly brains are tiny, smaller than a poppy seed. Previously, researchers had mapped the brains of a few other organisms, including the roundworm C. elegans, however their brains consist of only 302 neurons.
Most of the work was conducted by a group of researchers called the FlyWire consortium. The completion of the project and ongoing research is expected to result in a revolution in neuroscience. Previously it was believed that brains with hundreds of thousands of neurons were too large to map and assess function in much detail. But the results are a first step toward being able to do so, and potentially toward mapping at least segments of larger brains (including humans with more than 80 billion neurons and 100 trillion connections). The research has already revealed a number of important, and in some cases, surprising findings.
Neuron Types
The research has identified at least 8,453 neuronal cell types.3 A neuron cell type is a group that has similar morphology and connectivity. This compares with the worm C. elegans which has 118 cell types.4 The research also identified different classes of neurons, depending upon their function. Examples include sensory neurons (labeled afferent) that send signals from sensory organs to the brain. Motor and endocrine neurons (labeled efferent) send signals from the brain to muscles and other organs.5
Previously, some theorized that brain neurons might be like “snowflakes,” that is, each one is unique. That would imply their development and connections are essentially a random process. However, the research confirms that is generally not the case. There is some evidence of randomness, as one analysis shows that, “Over 50% of the connectome graph is a snowflake. Of course, these non-reproducible edges [connections] are mostly weak.”6 The analysis does show that, “Neurons occasionally do something unexpected (take a different route or make an extra branch on one side of the brain). We hypothesize that such stochastic differences are unnoticed variability present in most brains…In conclusion, we have not collected a snowflake.”7 This means that the stronger connections are largely stereotyped and do not vary in a random manner significantly. Conversely, the findings show convincingly that neither is the brain structure a regular lattice type, as in crystals.
Complexity
Fruit flies exhibit a number of complex behaviors, including flight control (hovering, rapid changes in direction), navigation, mating courtship using pheromones, and swarming. Therefore, it isn’t that surprising that their brains show complexity. The average fruit fly neuron connection consists of 12.6 synapses.8 Individual neurons typically have less than 10 connections, but some have more than 100, and even a few have 1,000.9This means that there isn’t a uniform distribution of neurons or a uniform distribution of connections. The research has even been able to map the flow of information throughout the brain. The fruit fly brain consists of areas of specialized functions. These include visual processing, olfactory, auditory, mechanical sensors, and temperature sensors. A further indication of specialized functions is the report of one research project that analyzed 78 anatomically distinct “subnetworks” in the brain.10 This same analysis concluded, “The local structure of the brain displays a high degree of non-randomness, consistent with previous studies in C. elegans and in the mouse cortex.”11
The overall structure of the brain is consistent among fruit flies, based on the finding of “[a] high degree of stereotypy at every level; neuron counts are highly consistent between brains, as are connections above a certain weight.”12 This is consistent with previous research with different insect brains.13
Another finding from the research is that the fruit fly brain exhibits the characteristics of is what is called a “small-world network,” where the “nodes are highly clustered and path lengths are short.”14 Other examples of small-world networks are power grids, train routes, and electronic circuits. The brain of C. elegans was the first example identified of a small-world neural network. Characteristics of small-world networks include “enhanced signal-propagation, computational power, and synchronizability.”15 The key benefit for brain function is that it provides “highly effective global communication among neurons.”16
Overall, the research shows that the fruit fly brain has a high degree of complexity, but more importantly, much of it is specified complexity. This includes the engineering design of the various specialized neural networks and subnetworks. Some of the engineering design principles that are evident in aspects of the brain include optimization, efficiency, and coherence. As complex as the brain is shown to be so far from the research, it is likely even more complex than currently appears to be the case since the electrical connections have yet to be fully mapped in a similar way to the chemical connections
