The geometry of robustness in spiking neural networks.

Neural systems are remarkably robust against various perturbations, a phenomenon that still requires a clear explanation. Here, we graphically illustrate how neural networks can become robust. We study spiking networks that generate low-dimensional representations, and we show that the neurons' subthreshold voltages are confined to a convex region in a lower-dimensional voltage subspace, which we call a 'bounding box'. Any changes in network parameters (such as number of neurons, dimensionality of inputs, firing thresholds, synaptic weights, or transmission delays) can all be understood as deformations of this bounding box. Using these insights, we show that functionality is preserved as long as perturbations do not destroy the integrity of the bounding box. We suggest that the principles underlying robustness in these networks - low-dimensional representations, heterogeneity of tuning, and precise negative feedback - may be key to understanding the robustness of neural systems at the circuit level.

Efficacy of Tricaine (MS-222) and Hypothermia as Anesthetic Agents for Blocking Sensorimotor Responses in Larval Zebrafish.

Tricaine, or MS-222, is the most commonly used chemical anesthetic in zebrafish research. It is thought to act via blocking voltage-gated sodium channels, though its mechanism of action, particularly at the neuronal level, is not yet fully understood. Here, we first characterized the effects of tricaine on both body balance and touch responses in freely swimming animals, before determining its effect on the neural activity underlying the optokinetic response at the level of motion perception, sensorimotor signaling and the generation of behavior in immobilized animals. We found that the standard dose for larvae (168 mg/L) induced loss of righting reflex within 30 seconds, which then recovered within 3 minutes. Optokinetic behavior recovered within 15 minutes. Calcium imaging showed that tricaine interferes with optokinetic behavior by interruption of the signals between the pretectum and hindbrain. The motion sensitivity indices of identified sensory neurons were unchanged in larvae exposed to tricaine, though fewer such neurons were detected, leaving a small population of active sensory neurons. We then compared tricaine with gradual cooling, a potential non-chemical alternative method of anesthesia. While neuronal tuning appeared to be affected in a similar manner during gradual cooling, gradual cooling induced a surge in calcium levels in both the pretectum and hindbrain. This calcium surge, alongside a drop in heartrate, is potentially associated with harmful changes in physiology and suggests that tricaine is a better anesthetic agent than gradual cooling for zebrafish laboratory research.

Spherical arena reveals optokinetic response tuning to stimulus location, size, and frequency across entire visual field of larval zebrafish.

Many animals have large visual fields, and sensory circuits may sample those regions of visual space most relevant to behaviours such as gaze stabilisation and hunting. Despite this, relatively small displays are often used in vision neuroscience. To sample stimulus locations across most of the visual field, we built a spherical stimulus arena with 14,848 independently controllable LEDs. We measured the optokinetic response gain of immobilised zebrafish larvae to stimuli of different steradian size and visual field locations. We find that the two eyes are less yoked than previously thought and that spatial frequency tuning is similar across visual field positions. However, zebrafish react most strongly to lateral, nearly equatorial stimuli, consistent with previously reported spatial densities of red, green, and blue photoreceptors. Upside-down experiments suggest further extra-retinal processing. Our results demonstrate that motion vision circuits in zebrafish are anisotropic, and preferentially monitor areas with putative behavioural relevance.

Publisher Correction: Evoking and tracking zebrafish eye movement in multiple larvae with ZebEyeTrack.

The version of this paper originally published contained the following text errors: (1) In the abstract, "(ii) visual stimulation with moving bars; (ii) eye detection and tracking, as well as general motion detection" should have been "(ii) visual stimulation with moving bars; (iii) eye detection and tracking, as well as general motion detection." (2) In the legend for Table 1, "vertical pixel coordinate; LE, left eye; RE, right eye; x, horizontal pixel coordinate; y" should have read "LE, left eye; RE, right eye; x, horizontal pixel coordinate; y, vertical pixel coordinate." These errors have been corrected in the HTML and PDF versions of the paper.

Evoking and tracking zebrafish eye movement in multiple larvae with ZebEyeTrack.

Reliable measurement of spontaneous and evoked eye movement is critical for behavioral vision research. Zebrafish are increasingly used as a model organism for visual neural circuits, but ready-to-use eye-tracking solutions are scarce. Here, we present a protocol for automated real-time measurement of angular horizontal eye position in up to six immobilized larval fish using a custom-built LabVIEW-based software, ZebEyeTrack. We provide its customizable source code, as well as a streamlined and compiled version, ZebEyeTrack Light. The full version of ZebEyeTrack controls all required hardware and synchronizes six essential aspects of the experiment: (i) stimulus design; (ii) visual stimulation with moving bars; (ii) eye detection and tracking, as well as general motion detection; (iv) real-time analysis; (v) eye-position-dependent closed-loop event control; and (vi) recording of external event times. This includes optional integration with external hardware such as lasers and scanning microscopes. Once installation is complete, experiments, including stimulus design, can be completed in <10 min, and recordings can last anywhere between seconds and many hours. Results include digitized angular eye positions and hardware status, which can be used to compute tuning curves, optokinetic gain, and other custom data analysis. After the experiment, or based on existing videos, optokinetic response (OKR) performance can be analyzed semi-automatically via the graphical user interface, and results can be exported. ZebEyeTrack has been used successfully for psychophysics experiments, for optogenetic stimulation, and in combination with calcium imaging.

Force-driven separation of short double-stranded DNA.

Short double-stranded DNA is used in a variety of nanotechnological applications, and for many of them, it is important to know for which forces and which force loading rates the DNA duplex remains stable. In this work, we develop a theoretical model that describes the force-dependent dissociation rate for DNA duplexes tens of basepairs long under tension along their axes ("shear geometry"). Explicitly, we set up a three-state equilibrium model and apply the canonical transition state theory to calculate the kinetic rates for strand unpairing and the rupture-force distribution as a function of the separation velocity of the end-to-end distance. Theory is in excellent agreement with actual single-molecule force spectroscopy results and even allows for the prediction of the rupture-force distribution for a given DNA duplex sequence and separation velocity. We further show that for describing double-stranded DNA separation kinetics, our model is a significant refinement of the conventionally used Bell-Evans model.