Description of BlueBerry wireless optogenetic system.

a) Front and back of the BlueBerry device shown beside a real blueberry for scale (left). Image of a mouse carrying a BlueBerry device connected to the multichannel LED implant fixed on the skull (right). b) Exploded 3D view of the BlueBerry device illustrating all miniature electronic components. c) Simplified schematic of the control BlueHub control unit, the BlueBerry device and LED implants. Each BlueHub input channel (BNC1–4) can be assigned to a function that includes the BlueBerry (BlueBerry id) device to activate and the corresponding stimulation parameters (stimulation channel, frequency, pulse width and number of pulses). When a channel is triggered, the BlueHub sends a Bluetooth packet to the assigned BlueBerry device containing the specified stimulation protocol. On the BlueBerry side, Bluetooth packets are decoded inside the bluetooth module and communicated with the microcontroller where each output stimulation channel is configured in terms of frequency, pulse width and number of pulses. d) Multichannel implants targeting superficial cortical layers (top) through surface mount micro LEDs or deep brain regions through custom made fiber coupled LED (bottom).

Closed-loop control of large-scale navigation using the BlueBerry.

a) Coronal section of mouse brain with bilateral ChR2 expression on the barrel sensory field and the LED implant setup. b) Top view of the arrangement of the two-channel LED implant on the skull before covering with dental acrylic. c & d) Detailed configuration of the infinit Y-maze along with an example highlighting the trial structure. Mice arriving at the intersection zones received optogenetic stimulation based on a randomly chosen path (contralateral barrel sensory field) that leads to the arm containing a rewarded port. Reward is only delivered when mice choose the correct path and arrive in the corresponding reward zone. e) Performance of mice in first (no stimulation) versus the last session (4ms - 20Hz) to associate artificial sensory cues with the correct chosen path at every intersection in both small and large infinit Y-maze (N =6 ChR2 expressing mice, blue lines, 3 mice per maze; N=5 control mice not expressing ChR2, grey lines, 3 for small and 2 for large maze; two-way repeated measure ANOVA, main and interaction effect P< 0.01, pairwise two-sample t-test,Tukey post hoc test, ***P < 0.001). Bars represent standard error of mean.

Closed-loop shaping of social behavior in groups of mice.

a) Top: Fiber-coupled LED setup for freely moving mice to optogenetically stimulate DA neurons expressing ChR2 in the VTA. Bottom: Representative histology of the optical fiber trace and virus expression in the VTA region. b) Example trajectory of a mouse during place preference tasks (two chambers setting) where they received VTA optogenetic stimulation using the BlueBerry device upon entering a conditioned chamber. c) Performance of all mice in the real-time place preference task (N=9, pairwise student t-test, ***P<0.001) d) Schematic of the tube test used to determine dominance hierarchy within each cage. e) Color coded hierarchy (ranks 1-3) for each cage based on tube test outcomes. f) Schema of real-time tracking and interaction detection. Simultaneous pose estimation of multiple freely moving mice is used to detect social interaction in real-time. Unilateral interaction is detected based on the angle (<110 degree) and distance (<5cm) of one mouse’s nose to another’s mouse tail. g) Right: Expected structure for baseline social interactions within a triad, with dominant interactions shown by thick arrows and submissive interactions by thin arrows. Left: Percentage of unilateral interactions initiated by submissive and dominant mice during baseline. The y-axis is scaled logarithmically (3 cages, 3 males per cage; pairwise t-test on log-transformed interaction durations, *P < 0.05). h) Behavior-triggered reinforcement strategy: rank 2 (green) and rank 3 (blue) mice received optogenetic VTA stimulation upon initiating unilateral interactions toward rank 1 (red) and rank 2 (green) mice, respectively. Bottom: Experimental timeline showing pre-stimulation, stimulation, and post-stimulation phases. i) Percentage of unilateral dyadic interactions initiated by the submissive member of each pair across pre-stimulation (first day), stimulation (last day) and post-stimulation (last day) sessions. The y-axis is scaled logarithmically (N=9 interaction, 3 interactions per cage, one-way repeated measure ANOVA, main effect ***P<0.001, Bonferroni-corrected paired t-tests for post hoc test on log-transformed interaction durations, ***P < 0.001). j) Train formation probability across pre-stimulation, stimulation and post-stimulation sessions. Each black line represents a cage. The y-axis is scaled logarithmically. The pie-chart represents the distribution of all possible trains (with the target train shown in orange) averaged across three cages during the optogenetic stimulation session. (Top) Image of multiple mice during an optogenetic stimulation session forming a moving train. The plotted dots on each mouse shows the location of their body parts (nose, body and tail) and the color indicates their natural hierarchy rank as shown in panel (e).