Mitral/tufted cells of the olfactory bulb receive odorant information from receptor neurons and transmit this information to the cortex. Studies in awake behaving animals have found that sustained responses of mitral cells to odorants are rare, suggesting sparse combinatorial representation of the odorants. Careful alignment of mitral cell firing with the phase of the respiration cycle revealed brief transient activity in the larger population of mitral cells, which respond to odorants during a small fraction of the respiration cycle. Responses of these cells are therefore temporally sparse. Here, we propose a mathematical model for the olfactory bulb network that can reproduce both combinatorially and temporally sparse mitral cell codes. We argue that sparse codes emerge as a result of the balance between mitral cells’ excitatory inputs and inhibition provided by the granule cells. Our model suggests functional significance for the dendrodendritic synapses mediating interactions between mitral and granule cells.

We propose a version of least-mean-square (LMS) algorithm for sparse system identification. Our algorithm called online linearized Bregman iteration (OLBI) is derived from minimizing the cumulative prediction error squared along with an l 1 -l 2 norm regularizer. By systematically treating the non-differentiable regularizer we arrive at a simple two-step iteration. We demonstrate that OLBI is bias free and compare its operation with existing sparse LMS algorithms by rederiving them in the online convex optimization framework. We perform convergence analysis of OLBI for white input signals and derive theoretical expressions for the steady state mean square deviations (MSD). We demonstrate numerically that OLBI improves the performance of LMS type algorithms for signals generated from sparse tap weights.

Electrical microstimulation can establish causal links between the activity of groups of neurons and perceptual and cognitive functions. However, the number and identities of neurons microstimulated, as well as the number of action potentials evoked, are difficult to ascertain. To address these issues we introduced the light-gated algal channel channelrhodopsin-2 (ChR2) specifically into a small fraction of layer 2/3 neurons of the mouse primary somatosensory cortex. ChR2 photostimulation in vivo reliably generated stimulus-locked action potentials at frequencies up to 50 Hz. Here we show that naive mice readily learned to detect brief trains of action potentials (five light pulses, 1 ms, 20 Hz). After training, mice could detect a photostimulus firing a single action potential in approximately 300 neurons. Even fewer neurons (approximately 60) were required for longer stimuli (five action potentials, 250 ms). Our results show that perceptual decisions and learning can be driven by extremely brief epochs of cortical activity in a sparse subset of supragranular cortical pyramidal neurons.

Sparse coding may be a general strategy of neural systems for augmenting memory capacity. In Drosophila melanogaster, sparse odor coding by the Kenyon cells of the mushroom body is thought to generate a large number of precisely addressable locations for the storage of odor-specific memories. However, it remains untested how sparse coding relates to behavioral performance. Here we demonstrate that sparseness is controlled by a negative feedback circuit between Kenyon cells and the GABAergic anterior paired lateral (APL) neuron. Systematic activation and blockade of each leg of this feedback circuit showed that Kenyon cells activated APL and APL inhibited Kenyon cells. Disrupting the Kenyon cell-APL feedback loop decreased the sparseness of Kenyon cell odor responses, increased inter-odor correlations and prevented flies from learning to discriminate similar, but not dissimilar, odors. These results suggest that feedback inhibition suppresses Kenyon cell activity to maintain sparse, decorrelated odor coding and thus the odor specificity of memories.

Lipid droplets (LDs) are neutral lipid storage organelles that transfer lipids to various organelles including peroxisomes. Here, we show that the hereditary spastic paraplegia protein M1 Spastin, a membrane-bound AAA ATPase found on LDs, coordinates fatty acid (FA) trafficking from LDs to peroxisomes through two inter-related mechanisms. First, M1 Spastin forms a tethering complex with peroxisomal ABCD1 to promote LD-peroxisome contact formation. Second, M1 Spastin recruits the membrane-shaping ESCRT-III proteins IST1 and CHMP1B to LDs via its MIT domain to facilitate LD-to-peroxisome FA trafficking, possibly through IST1 and CHMP1B modifying LD membrane morphology. Furthermore, M1 Spastin, IST1 and CHMP1B are all required to relieve LDs of lipid peroxidation. The roles of M1 Spastin in tethering LDs to peroxisomes and in recruiting ESCRT-III components to LD-peroxisome contact sites for FA trafficking may help explain the pathogenesis of diseases associated with defective FA metabolism in LDs and peroxisomes.

Tissue and organ function has been conventionally understood in terms of the interactions among discrete and homogeneous cell types. This approach has proven difficult in neuroscience due to the marked diversity across different neuron classes, but it may be further hampered by prominent within-class variability. Here, we considered a well-defined canonical neuronal population-hippocampal CA1 pyramidal cells (CA1 PCs)-and systematically examined the extent and spatial rules of transcriptional heterogeneity. Using next-generation RNA sequencing, we identified striking variability in CA1 PCs, such that the differences within CA1 along the dorsal-ventral axis rivaled differences across distinct pyramidal neuron classes. This variability emerged from a spectrum of continuous gene-expression gradients, producing a transcriptional profile consistent with a multifarious continuum of cells. This work reveals an unexpected amount of variability within a canonical and narrowly defined neuronal population and suggests that continuous, within-class heterogeneity may be an important feature of neural circuits.