A two-timepoint framework for sensitive and specific single-cell activity screening

Methods to identify brain areas where neurons are activated in response to stimuli or states typically rely on immediate early gene (IEG) expression. However, variability in IEG expression across subjects and brain areas often requires large sample sizes. Further, IEG expression alone cannot determine whether the same or different neurons are activated in response to two distinct stimuli or states. To overcome these challenges, we developed a brain screening method-two-timepoint statistical inference with subtraction (TTP-S)-to identify brain areas on the basis of activity at two timepoints, not one. Applying this approach across 500+ brain areas increases the sensitivity and specificity of screens for engaged brain areas, thereby reducing the required sample size. Coupled with graph theoretical analyses, we use this method to analyze hunger, satiety, food cue presentation, drug treatment, and alcohol consumption, identifying known relevant brain regions and implicating new areas.