README.md

---
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)
```