SwissProt-EC-leaf

# Dataset

Swissprot is a high quality manually annotated protein database. The dataset contains annotations with the functional properties of the proteins. Here we extract proteins with Enzyme Commission labels.

The dataset is ported from Protinfer: https://github.com/google-research/proteinfer.

The leaf level EC-labels are extracted and indexed, the mapping is provided in `idx_mapping.json`. Proteins without leaf-level-EC tags are removed.

## Example
The protein Q87BZ2 have the following EC tags.

EC:2.-.-.- (Transferases)
EC:2.7.-.- (Transferring phosphorus-containing groups)
EC:2.7.1.- (Phosphotransferases with an alcohol group as acceptor)
EC:2.7.1.30 (Glycerol kinase)

We only extract the leaf level labels, here EC:2.7.1.30, corresponding to glycerol kinase.

# 数据集
Swiss-Prot（Swissprot）是一款经人工注释的高质量蛋白质数据库，其收录了蛋白质功能属性相关的注释信息。本数据集提取了带有酶学委员会（Enzyme Commission，EC）标签的蛋白质条目。

该数据集源自Protinfer项目，原始来源地址为：https://github.com/google-research/proteinfer。

我们对叶层级EC标签进行了提取与索引，标签与索引的映射关系存储于`idx_mapping.json`文件中，所有不包含叶层级EC标签的蛋白质均已被剔除。

## 示例
蛋白质Q87BZ2包含如下EC标签：

EC:2.-.-.- (转移酶类)
EC:2.7.-.- (转移含磷基团类)
EC:2.7.1.- (以醇基团为受体的磷酸转移酶)
EC:2.7.1.30 (甘油激酶)

我们仅保留叶层级标签，本例中即对应甘油激酶的EC:2.7.1.30。