bloyal/deeploc
收藏资源简介:
--- license: cc-by-4.0 --- # DeepLoc-2.0 Training Data Dataset from https://services.healthtech.dtu.dk/services/DeepLoc-2.0/ used to train the DeepLoc-2.0 model. ## Data preparation Data downloaded and processed using the following Python script: ```python import pandas as pd df = pd.read_csv('https://services.healthtech.dtu.dk/services/DeepLoc-2.0/data/Swissprot_Train_Validation_dataset.csv').drop(['Unnamed: 0', 'Partition'], axis=1) df['labels'] = df[['Cell membrane', 'Cytoplasm','Endoplasmic reticulum', 'Extracellular', 'Golgi apparatus', 'Lysosome/Vacuole', 'Mitochondrion', 'Nucleus', 'Peroxisome', 'Plastid']].astype('float32').values.tolist() df['Membrane'] = df['Membrane'].astype('float32') df = df[['Kingdom', 'ACC', 'Sequence','Membrane','labels']] train = df.sample(frac=0.8) df = df.drop(train.index) val = df.sample(frac=0.5) test = df.drop(val.index) train = train.reset_index(drop=True) val = val.reset_index(drop=True) test = test.reset_index(drop=True) train.to_parquet('deeploc-train.parquet', index=False) val.to_parquet('deploc-val.parquet', index=False) test.to_parquet('deeploc-test.parquet', index=False) ``` ## Labels {'Cell membrane': 0, 'Cytoplasm': 1, 'Endoplasmic reticulum': 2, 'Extracellular': 3, 'Golgi apparatus': 4, 'Lysosome/Vacuole': 5, 'Mitochondrion': 6, 'Nucleus': 7, 'Peroxisome': 8, 'Plastid': 9} ## Citation **DeepLoc-2.0:** ``` Vineet Thumuluri and others, DeepLoc 2.0: multi-label subcellular localization prediction using protein language models, Nucleic Acids Research, Volume 50, Issue W1, 5 July 2022, Pages W228–W234, https://doi.org/10.1093/nar/gkac278 ``` The DeepLoc data is a derivative of the UniProt dataset: **UniProt** ``` The UniProt Consortium UniProt: the Universal Protein Knowledgebase in 2023 Nucleic Acids Res. 51:D523–D531 (2023) ```
--- 许可证:知识共享署名4.0(CC BY 4.0) --- # DeepLoc-2.0 训练数据集 本数据集源自https://services.healthtech.dtu.dk/services/DeepLoc-2.0/,用于训练DeepLoc-2.0模型。 ## 数据准备 本数据集通过以下Python脚本完成下载与处理: python import pandas as pd df = pd.read_csv('https://services.healthtech.dtu.dk/services/DeepLoc-2.0/data/Swissprot_Train_Validation_dataset.csv').drop(['Unnamed: 0', 'Partition'], axis=1) df['labels'] = df[['Cell membrane', 'Cytoplasm','Endoplasmic reticulum', 'Extracellular', 'Golgi apparatus', 'Lysosome/Vacuole', 'Mitochondrion', 'Nucleus', 'Peroxisome', 'Plastid']].astype('float32').values.tolist() df['Membrane'] = df['Membrane'].astype('float32') df = df[['Kingdom', 'ACC', 'Sequence','Membrane','labels']] train = df.sample(frac=0.8) df = df.drop(train.index) val = df.sample(frac=0.5) test = df.drop(val.index) train = train.reset_index(drop=True) val = val.reset_index(drop=True) test = test.reset_index(drop=True) train.to_parquet('deeploc-train.parquet', index=False) val.to_parquet('deploc-val.parquet', index=False) test.to_parquet('deeploc-test.parquet', index=False) ## 标签 { "细胞膜(Cell membrane)": 0, "细胞质(Cytoplasm)": 1, "内质网(Endoplasmic reticulum)": 2, "细胞外区域(Extracellular)": 3, "高尔基体(Golgi apparatus)": 4, "溶酶体/液泡(Lysosome/Vacuole)": 5, "线粒体(Mitochondrion)": 6, "细胞核(Nucleus)": 7, "过氧化物酶体(Peroxisome)": 8, "质体(Plastid)": 9 } ## 引用信息 **DeepLoc-2.0:** Vineet Thumuluri 等. DeepLoc 2.0:基于蛋白质语言模型的多标签亚细胞定位预测[J]. 核酸研究(Nucleic Acids Research), 2022, 第50卷, 第W1期, 2022年7月5日, 第W228-W234页. https://doi.org/10.1093/nar/gkac278 DeepLoc数据集衍生自通用蛋白质知识库(UniProt)数据集: **通用蛋白质知识库(UniProt):** The UniProt Consortium UniProt: the Universal Protein Knowledgebase in 2023 Nucleic Acids Res. 51:D523–D531 (2023)
DeepLoc-2.0 Training Data
数据来源
- 数据集来源于DeepLoc-2.0,用于训练DeepLoc-2.0模型。
数据准备
- 数据通过Python脚本下载并处理,原始数据为CSV格式,文件名为
Swissprot_Train_Validation_dataset.csv。 - 数据处理包括删除不必要的列、转换标签类型,并重新组织数据结构。
- 最终数据分为训练集、验证集和测试集,分别保存为
deeploc-train.parquet、deploc-val.parquet和deeploc-test.parquet。
标签定义
- 标签包括10个类别,每个类别对应一个整数值,如下:
Cell membrane: 0Cytoplasm: 1Endoplasmic reticulum: 2Extracellular: 3Golgi apparatus: 4Lysosome/Vacuole: 5Mitochondrion: 6Nucleus: 7Peroxisome: 8Plastid: 9
引用信息
- 数据集引用自DeepLoc-2.0的研究,详细信息可参考Nucleic Acids Research, Volume 50, Issue W1, 5 July 2022。
- 数据集是基于UniProt数据集的衍生,UniProt的引用信息可参考Nucleic Acids Res. 51:D523–D531 (2023)。




