ORBIT: Learning Gene Program Co-Activation Structure for Cell-Type-Stratified Pathway Rewiring Analysis in Single-Cell Transcriptomics

TL;DR

ORBIT is a transformer-based method that learns asymmetric gene program influences from single-cell RNA data, revealing cell-type-specific pathway rewiring and improving classification accuracy.

Contribution

It introduces a self-supervised transformer with an intervention-based training objective to model directional gene program dependencies from observational data.

Findings

Recovers co-activation structures aligned with Alzheimer's signatures.

Identifies cell-type-specific pathway rewiring invisible to differential expression.

Achieves high cell-type classification accuracy close to gene-based classifiers.

Abstract

Gene programs co-activate within cells, but existing single-cell methods either treat programs independently or require experimental perturbation data to model their interactions. We introduce ORBIT, a self-supervised transformer that learns asymmetric dependencies among gene programs from observational single-cell RNA-sequencing data alone, quantifying how strongly each program influences every other program. The key mechanism is an intervention-consistent training objective: the model learns each program's directional influence on every other program by predicting how the others change when that program is removed, yielding attention weights that reflect asymmetric influence rather than symmetric co-occurrence. Applied to 191,890 prefrontal cortex nuclei across three pathway vocabularies, ORBIT recovers co-activation structure consistent with established Alzheimer's disease vulnerability signatures, identifies cell-type-specific rewiring invisible to differential expression, and achieves 0.984 macro F1 on cell-type classification from 220 pathway scores, which is within 0.3 points of a state-of-the-art classifier using all 22,088 genes.