How snap-stanford/gears works
GEARS turns perturbational single-cell data into predictions of transcriptional responses, including combinatorial perturbations. The main workflow is to load or preprocess data, define perturbation splits and batches, initialize and train the model, then predict and evaluate responses; the API also exposes genetic-interaction analysis. The graph preserves documented dataset, uncertainty, and plotting capabilities. Some subsystem links rely on the README or SOURCE INDEX; unsampled implementation details are not inferred.
Main entry points: pertdata.py PertData API · gears.py GEARS API · model.py GEARS neural model · data_utils.py Data preprocessing
Components 16
Perturbation data
- PertData API
gears/pertdata.py - Data preprocessing
gears/data_utils.py - Cell graph datasets
gears/pertdata.py - Perturbation splits
gears/data_utils.py
Modeling
- GEARS API
gears/gears.py - Training and evaluation
gears/gears.py - GEARS neural model
gears/model.py - Uncertainty modeling
gears/model.py
Prediction and analysis
- Perturbation prediction
gears/gears.py - Inference metrics
gears/inference.py - Genetic interactions
gears/inference.py - Result visualization
gears/gears.py
Data and graph utilities
- Gene graph utilities
gears/utils.py - Data download utilities
gears/utils.py
Other components
- Researcher
- Harvard Dataverse
Connections 19
- Researcher → PertData API: loads data
README.md - PertData API → Data preprocessing: preprocesses data
gears/pertdata.py - PertData API → Perturbation splits: prepares splits
README.md - PertData API → Cell graph datasets: builds batches
gears/pertdata.py - PertData API → Data download utilities: uses utilities
gears/pertdata.py - PertData API → Harvard Dataverse: downloads resources
gears/pertdata.py - Researcher → GEARS API: initializes model
README.md - GEARS API → Training and evaluation: trains model
README.md - Training and evaluation → GEARS neural model: optimizes
gears/gears.py - Training and evaluation → Inference metrics: evaluates with
gears/gears.py - GEARS neural model → Uncertainty modeling: supports
gears/model.py - Researcher → Perturbation prediction: requests predictions
README.md - GEARS API → Perturbation prediction: dispatches prediction
gears/gears.py - Perturbation prediction → GEARS neural model: runs forward pass
gears/gears.py - GEARS API → Inference metrics: uses metrics
gears/gears.py - GEARS API → Genetic interactions: computes interactions
README.md - GEARS API → Result visualization: plots results
README.md - Data preprocessing → Gene graph utilities: uses graph helpers
gears/data_utils.py - Gene graph utilities → Harvard Dataverse: downloads graph data
gears/utils.py