---
title: uni-fold-monomer
canonical_url: "https://www.modelscope.cn/models/DPTech/uni-fold-monomer"
md_url: "https://www.modelscope.cn/models/DPTech/uni-fold-monomer.md"
repository: DPTech/uni-fold-monomer
chinese_name: "Uni-Fold-Monomer 开源的蛋白质单体结构预测模型"
last_updated: 2023-01-05
license: "Apache License 2.0"
pipeline_tag: protein-structure
tasks:
  - protein-structure
model_type:
  - unifold
library_name:
  - pytorch
frameworks:
  - pytorch
domain:
  - science
downloads: 1289
stars: 48
tags:
  - evoformer
  - DPTechnology
  - unifold
  - "蛋白质结构预测"
---

# uni-fold-monomer

> uni-fold-monomer - DPTech 在 ModelScope 开源的模型。开源的蛋白质单体结构预测模型，输入蛋白质单体的一级结构（1D序列），预测蛋白质的三级结构（3D位置），同时给出预测结果的置信度。本模型主要用于蛋白质单体的预测。

DPTech/uni-fold-monomer 是 ModelScope 魔搭社区上的protein-structure模型，采用 Apache License 2.0 许可。

- **Repository**: DPTech/uni-fold-monomer
- **License**: Apache License 2.0
- **Tasks**: protein-structure
- **Tags**: evoformer, DPTechnology, unifold, 蛋白质结构预测
- **Downloads**: 1289
- **Stars**: 48
- **Last updated**: 2023-01-05

Source: https://www.modelscope.cn/models/DPTech/uni-fold-monomer

---

# Uni-Fold-Monomer 开源的蛋白质单体结构预测模型

## 模型描述
输入蛋白质单体的一级结构（1D序列），预测蛋白质的三级结构（3D位置），同时给出预测结果的置信度。

## 期望模型使用方式以及适用范围
本模型主要用于蛋白质单体的预测。

### 如何使用
目前仅支持GPU运行，需要额外安装[Uni-Core](https://github.com/dptech-corp/Uni-Core/releases)。

#### 代码范例
```python
from modelscope.pipelines import pipeline

pipeline_ins = pipeline(
    task='protein-structure',
    model='DPTech/uni-fold-monomer')
protein = 'LILNLRGGAFVSNTQITMADKQKKFINEIQEGDLVRSYSITDETFQQNAVTSIVKHEADQLCQINFGKQHVVC'
outputs = pipeline_ins(protein)
```

### 模型局限性以及可能的偏差
只适用于单体蛋白质、多聚体蛋白请使用 [Uni-Fold-Multimer](https://modelscope.cn/models/DPTech/uni-fold-multimer/summary)。

模型性能依赖于同源序列的数量，如果同源序列数量少，模型可能无法保证预测结果。

对于超长序列，预测时间需要较久，显存也可能不够。


## 训练代码及训练数据
模型根据AlphaFold的论文描述训练，训练代码开源在[GitHub](https://github.com/dptech-corp/Uni-Fold)。

训练使用的数据 [Uni-Fold-Data](https://modelscope.cn/datasets/DPTech/Uni-Fold-Data/summary) 也在 ModelScope 上开放。


### 相关论文以及引用信息
```bibtex
@article {uni-fold,
	author = {Li, Ziyao and Liu, Xuyang and Chen, Weijie and Shen, Fan and Bi, Hangrui and Ke, Guolin and Zhang, Linfeng},
	title = {Uni-Fold: An Open-Source Platform for Developing Protein Folding Models beyond AlphaFold},
	year = {2022},
	doi = {10.1101/2022.08.04.502811},
	URL = {https://www.biorxiv.org/content/10.1101/2022.08.04.502811v3},
	eprint = {https://www.biorxiv.org/content/10.1101/2022.08.04.502811v3.full.pdf},
	journal = {bioRxiv}
}
```
