TAKO ranks gene knockout effects from wild-type single-cell data in seconds, no training required
Researchers introduced TAKO, a fast and interpretable method that ranks likely targets of a single-gene knockout using only wild-type single-cell ...
Researchers introduced TAKO, a fast and interpretable method that ranks likely targets of a single-gene knockout using only wild-type single-cell ...
A new graph-attention-based generative adversarial network called GARAGE produces realistic synthetic single-cell RNA-seq data that preserves rare cell types and ...
A new calibration framework shows that common benchmarking metrics masked real perturbation-specific signal, and that deep learning models can outperform ...
Researchers in Hong Kong and Japan have developed an integer programming method that constrains single-cell trajectory trees to binary branching, ...
A new benchmark of pathway activity scoring methods in single-cell RNA-seq reveals that top-rank-window tools like AUCell and UCell can ...
A new unsupervised deep learning framework called SaDGAE uses a self-adaptive cell graph and a doubly enhanced graph autoencoder to ...
A new graph neural network framework called DrugSAGE predicts cancer drug responses by aggregating transcriptomic features from tumors and their ...
A new Bayesian framework called MANTRA jointly analyzes three-dimensional drug-response data and two-dimensional gene expression matrices, revealing clinically relevant leukemia ...
Researchers at Genentech have developed Fishash, a fast Fisher's Exact Test-based method that improves the accuracy and speed of assigning ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine