Model description
We introduce PINNACLE, a flexible geometric deep-learning approach that is trained on contextualized protein interaction networks to generate context-PINNACLE protein representations. Leveraging a human multi-organ single-cell transcriptomic atlas, PINNACLE provides 394,760 protein representations split across 156 cell type contexts from 24 tissues and organs. To load the pre-trained model, use the Files and Versions tab files.
Abstract
Understanding protein function and developing molecular therapies require deciphering the cell types in which proteins act as well as the interactions between proteins. However, modeling protein interactions across biological contexts remains challenging for existing algorithms. Here we introduce PINNACLE, a geometric deep learning approach that generates context-aware protein representations. Leveraging a multiorgan single-cell atlas, PINNACLE learns on contextualized protein interaction networks to produce 394,760 protein representations from 156 cell type contexts across 24 tissues. PINNACLE’s embedding space reflects cellular and tissue organization, enabling zero-shot retrieval of the tissue hierarchy. Pretrained protein representations can be adapted for downstream tasks: enhancing 3D structure-based representations for resolving immuno-oncological protein interactions, and investigating drugs’ effects across cell types. PINNACLE outperforms state-of-the-art models in nominating therapeutic targets for rheumatoid arthritis and inflammatory bowel diseases and pinpoints cell type contexts with higher predictive capability than context-free models. PINNACLE’s ability to adjust its outputs on the basis of the context in which it operates paves the way for large-scale context-specific predictions in biology.
Code
We support retrieval of PINNACLE embeddings.
from tdc.resource.pinnacle import PINNACLE
pinnacle = PINNACLE()
embeds = pinnacle.get_embeds()
We also share here the code used to benchmark PINNACLE on the TDC.scDTI task. https://colab.research.google.com/drive/14WeNbDNyDi_45hFayBMtt2MBOvhyRRvy?usp=sharing
TDC Citation
@inproceedings{
velez-arce2024signals,
title={Signals in the Cells: Multimodal and Contextualized Machine Learning Foundations for Therapeutics},
author={Alejandro Velez-Arce and Kexin Huang and Michelle M Li and Xiang Lin and Wenhao Gao and Bradley Pentelute and Tianfan Fu and Manolis Kellis and Marinka Zitnik},
booktitle={NeurIPS 2024 Workshop on AI for New Drug Modalities},
year={2024},
url={https://openreview.net/forum?id=kL8dlYp6IM}
}
Additional Citations
- Li, Michelle, et al. “Contextual AI models for single-cell protein biology” Nature Methods (2024)
Model Homepage
https://huggingface.co/spaces/michellemli/PINNACLE
Model Github
https://github.com/mims-harvard/PINNACLE
Dataset
Dataset entry in Therapeutics Data Commons for single-cell drug-target nomination https://tdcommons.ai/multi_pred_tasks/scdti/