Modern Python implementation for CRISPRi/a sgRNA activity prediction, based on Horlbeck et al., eLife 2016.
- Python 3.9+ compatible with type hints
- Multiple ML models: ElasticNet, Random Forest, Gradient Boosting, MLP
- Full feature set (~800 features) including sequence, positional, and chromatin features
- Clean, modular API
conda create -n crispria python=3.11
conda activate crispria
pip install -r crispria_modern/requirements.txt| Notebook | Description |
|---|---|
Demo_CRISPRia_Modern.ipynb |
Quick start demo showing basic usage of the toolkit |
Demo_CRISPRa_Full.ipynb |
Full training pipeline with ~800 features, model comparison (ElasticNet, Gradient Boosting, MLP), and evaluation |
from crispria_modern import DataLoader, ActivityPredictor
# Load data
loader = DataLoader("path/to/data_files")
data = loader.load_training_data("CRISPRa")
# Train model
predictor = ActivityPredictor(model_type="elasticnet")
predictor.fit(X_train, y_train)
# Predict
predictions = predictor.predict(X_new)Horlbeck MA, et al. Compact and highly active next-generation libraries for CRISPR-mediated gene repression and activation. eLife 2016;5:e19760.