Merge branch 'main' of hf.co:Oxer11/ESM-S
Browse files- README.md +37 -0
- asset/predictor.png +0 -0
- asset/retriever.png +0 -0
- asset/training.png +0 -0
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- Oxer11/Protein-Function-Annotation
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
tags:
|
| 8 |
+
- Protein Langauge Model
|
| 9 |
+
- AI for Drug Discovery
|
| 10 |
+
- AI for Science
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# ESM-S
|
| 14 |
+
|
| 15 |
+
ESM-S (https://arxiv.org/abs/2402.05856) is a series of structure-informed protein language models, which are trained on remote homology detection tasks for distilling structural information.
|
| 16 |
+
The corresponding datasets can be downloaded at https://huggingface.co/datasets/Oxer11/Protein-Function-Annotation.
|
| 17 |
+
The codebase can be found at https://github.com/DeepGraphLearning/esm-s.
|
| 18 |
+
|
| 19 |
+

|
| 20 |
+
|
| 21 |
+
# Evaluation Performance
|
| 22 |
+
|
| 23 |
+
Freezing model weights and train a 2-layer MLP on downstream function prediction tasks.
|
| 24 |
+

|
| 25 |
+
|
| 26 |
+
Using ESM-S representations to retrieve similar proteins for function annotation.
|
| 27 |
+

|
| 28 |
+
|
| 29 |
+
# BibTeX
|
| 30 |
+
```
|
| 31 |
+
@article{zhang2024structureplm,
|
| 32 |
+
title={Structure-Informed Protein Language Model},
|
| 33 |
+
author={Zhang, Zuobai and Lu, Jiarui and Chenthamarakshan, Vijil and Lozano, Aurelie and Das, Payel and Tang, Jian},
|
| 34 |
+
journal={arXiv preprint arXiv:2402.05856},
|
| 35 |
+
year={2024}
|
| 36 |
+
}
|
| 37 |
+
```
|
asset/predictor.png
ADDED
|
asset/retriever.png
ADDED
|
asset/training.png
ADDED
|