BRAID is a department within Genentech dedicated to advancing biological and clinical sciences through artificial intelligence. Our core focus is on developing foundation models—general-purpose AI models trained on large-scale biological datasets—which we fine-tune for specialized applications.
At BRAID, we not only develop cutting-edge AI tools but also actively use them to advance scientific discovery. All of the tools we make available to the scientific community are linked on this page. We always welcome feedback—feel free to contact Tommaso or the BRAID leadership team with any input or suggestions.
Foundation Models for TranscriptomicsSCimilarity : Unifying representation of single cell expression profiles
PaSCient: Multi-cellular representations of single-cell transcriptomics data
DNA/RNA Sequence2FunctiongReLU: Train, interpret, and apply deep learning models to DNA sequences
Decima: DNA sequence models on single-cell RNA-seq data
Polygraph: Evaluation and comaprison of nucleic acid sequences for regulatory element design
regLM: hyenaDNA-based autoregressive language models on DNA for generation of novel regulatory elements
ARGMINN: Mechanistically interpretable neural network for regulatory genomics
Virtual Screens / Drug discoveryGNEprop: Small molecule property predictor which we used for antibiotic discovery
MolCap-Arena: Benchmark on language-enhanced molecular property prediction
Foundational Machine LearningVCI: Estimating individual counterfactual treatment effects
BwR: Efficient and expressive graph generative modeling
GraphGUIDE: Interpretable and controllable conditional graph generation
SPICE: Uncertainty estimation via conformal prediction for deep learning models
CTRL: RL-based conditional control for diffusion models
CellArr: TileDB-backed store for large collections of genomic experimental data
Prior software from current BRAID team membersTangram: Spatial alignment of single cell and spatial transcriptomics
MiloDE: Sensitive DE testing using cell neighborhoods
Cumulus: A series of Cloud-based scalable and efficient single-cell genomics data analysis workflows
Pegasus: A scalable and efficient tool for analyzing transcriptomes of millions of single cells
Cirrocumulus: An interactive visualization tool for large-scale single-cell genomics data
Harmony-PyTorch: An efficient PyTorch implementation of Harmony algorithm, used for data integration and batch correction
NMF-Torch: An efficient PyTorch implementation of Non-negative Matrix Factorization (NMF) and integrative NMF (iNMF) algorithms, used for gene program analysis and batch correction
TorchDR: PyTorch implementation of dimensionality reduction methods including UMAP, TSNE and many others.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4
