cross-orientation suppression across
cortical depth in macaque v1

imaging

Two-photon calcium imaging in anesthetized macaque V1 using PHP.eB-CAG-GCaMP6s. Drifting gratings and orthogonal plaids at 50% contrast, 4 Hz, 4 cyc/deg.

• 28 fields of view, 140 to 518 um deep
• Superficial through deep layer 2/3
• Collaboration with Allen Institute and WANPRC

roi extraction

Suite2p for automated cell detection, adapted for primate cortical data. Neuropil correction applied to all traces.

• 4,785 ROIs across 28 depths
• Neuropil subtraction and signal cleanup
• SNR filtering at multiple thresholds

xori metric

For each ROI, build a linear prediction from component grating responses (shifted by 90 degrees, summed with baseline correction). Compare to actual plaid response.

• S: plaid / linear prediction ratio
• S > 1 = facilitation, S < 1 = suppression
• C: Pearson r between predicted and observed tuning

statistics

Mixed-effects models to handle the nested structure: ROIs within depths within sessions. Mediation analysis for covariates.

• Linear mixed-effects (statsmodels)
• Random intercepts for session and depth
• Mediation: spatial frequency (~40%), bandwidth (~21%)

surgical

I assist in non-human primate neurosurgery: craniotomy prep, optical window implantation, perioperative monitoring. This gives us chronic two-photon imaging access to V1.

computational

I built the entire XORI analysis pipeline. ROI extraction, signal processing, cross-orientation index computation, mixed-effects modeling, mediation analysis, all figures. Every line of analysis code is mine.