Resultado de búsqueda
DeepLoc 2.0 is a multi-label predictor, which means that is able to predict one or more localizations for any given protein. It can differentiate between 10 different localizations: Nucleus, Cytoplasm, Extracellular, Mitochondrion, Cell membrane, Endoplasmic reticulum, Chloroplast, Golgi apparatus, Lysosome/Vacuole and Peroxisome.
DeepLoc-1.0 predicts the subcellular localization of eukaryotic proteins. It can differentiate between 10 different localizations: Nucleus, Cytoplasm, Extracellular, Mitochondrion, Cell membrane, Endoplasmic reticulum, Chloroplast, Golgi apparatus, Lysosome/Vacuole and Peroxisome.
- Overview
- Data
- Models
- Setup
- Training Workflow
- Citation
Multi-label subcellular localization and sorting signal prediction based on protein foundation models (https://github.com/agemagician/ProtTrans, https://github.com/facebookresearch/esm).
Prediction webserver is available at https://services.healthtech.dtu.dk/services/DeepLoc-2.0/
The 'data_files' folder contains the data for training
1.multisub_5_partitions_unique.csv: Annotated SwissProt Sequences, labels, and partitions for subcellular localization
2.multisub_ninesignals.pkl, sorting_signals.csv: Annotated SwissProt Sequences and sorting signal annotations
3.Processed FASTA files for generating embeddings
Two models dubbed Fast (ESM1b) and Accurate (ProtT5) are used. refers to one of these.
It is recommened to setup a conda environment using
conda env create -f environment.yml
Training is divided into two stages:
python train_sl.py --model
1.Generate and store embeddings for faster training. Note: h5 files of ~30-40 GB are stored in "data_files/embeddings".
2.Train subcellular localization and interpretable attention.
3.Generate predictions and intermediate representations for sorting signal prediction.
4.Compute metrics on the SwissProt CV dataset.
If you found this useful please consider citing
30 de abr. de 2022 · DeepLoc 2.0 uses a transformer-based protein language model to predict multi-label subcellular localization and provides interpretability via the attention and sorting signal prediction.
5 de jul. de 2022 · Protein Sorting Signals. The prediction of protein subcellular localization is of great relevance for proteomics research. Here, we propose an update to the popular tool DeepLoc with multi-localization prediction and improvements in both performance and interpretability.
8 de abr. de 2024 · DeepLoc 2.0 is a popular web server for the prediction of protein subcellular localization and sorting signals. Here, we introduce DeepLoc 2.1, which additionally classifies the input proteins into the membrane protein types Transmembrane, Peripheral, Lipid-anchored and Soluble.
DeepLoc2 on Biowulf. DeepLoc2 uses deep learning to predict subcellular localization of eukaryotic proteins. DeepLoc 2.0 predicts the subcellular localization (s) of eukaryotic proteins. DeepLoc 2.0 is a multi-label predictor, which means that is able to predict one or more localizations for any given protein.
