Prot Bert Bfd

🚀 ProtBert-BFD model

A pre - trained model on protein sequences using a masked language modeling (MLM) objective, which captures important biophysical properties of proteins.

🚀 Quick Start

The ProtBert - BFD model can be directly used with a pipeline for masked language modeling or for protein feature extraction. You can also fine - tune it on downstream tasks.

✨ Features

• Pretrained on large - scale data: Trained on BFD, a dataset with 2.1 billion protein sequences.
• Self - supervised learning: Pretrained in a self - supervised fashion on raw protein sequences.
• Captures biophysical properties: The LM - embeddings from unlabeled data capture important biophysical properties governing protein shape.

💻 Usage Examples

Basic Usage

You can use this model directly with a pipeline for masked language modeling:

>>> from transformers import BertForMaskedLM, BertTokenizer, pipeline
>>> tokenizer = BertTokenizer.from_pretrained('Rostlab/prot_bert_bfd', do_lower_case=False )
>>> model = BertForMaskedLM.from_pretrained("Rostlab/prot_bert_bfd")
>>> unmasker = pipeline('fill-mask', model=model, tokenizer=tokenizer)
>>> unmasker('D L I P T S S K L V V [MASK] D T S L Q V K K A F F A L V T')

[{'score': 0.1165614128112793,
  'sequence': '[CLS] D L I P T S S K L V V L D T S L Q V K K A F F A L V T [SEP]',
  'token': 5,
  'token_str': 'L'},
 {'score': 0.08976086974143982,
  'sequence': '[CLS] D L I P T S S K L V V V D T S L Q V K K A F F A L V T [SEP]',
  'token': 8,
  'token_str': 'V'},
 {'score': 0.08864385634660721,
  'sequence': '[CLS] D L I P T S S K L V V S D T S L Q V K K A F F A L V T [SEP]',
  'token': 10,
  'token_str': 'S'},
 {'score': 0.06227643042802811,
  'sequence': '[CLS] D L I P T S S K L V V A D T S L Q V K K A F F A L V T [SEP]',
  'token': 6,
  'token_str': 'A'},
 {'score': 0.06194969266653061,
  'sequence': '[CLS] D L I P T S S K L V V T D T S L Q V K K A F F A L V T [SEP]',
  'token': 15,
  'token_str': 'T'}]

Advanced Usage

Here is how to use this model to get the features of a given protein sequence in PyTorch:

from transformers import BertModel, BertTokenizer
import re
tokenizer = BertTokenizer.from_pretrained('Rostlab/prot_bert_bfd', do_lower_case=False )
model = BertModel.from_pretrained("Rostlab/prot_bert_bfd")
sequence_Example = "A E T C Z A O"
sequence_Example = re.sub(r"[UZOB]", "X", sequence_Example)
encoded_input = tokenizer(sequence_Example, return_tensors='pt')
output = model(**encoded_input)

📚 Documentation

Model Description

ProtBert - BFD is based on the Bert model, which was pretrained on a large corpus of protein sequences in a self - supervised fashion. Each protein sequence is treated as a separate document, and the Next sentence prediction is not used. The masking follows the original Bert training, randomly masking 15% of the amino acids in the input. The feature extracted from this model shows that the LM - embeddings from unlabeled data capture important biophysical properties governing protein shape.

Intended Uses & Limitations

The model can be used for protein feature extraction or fine - tuned on downstream tasks. In some tasks, fine - tuning the model can lead to higher accuracy than using it as a feature extractor.

Training Data

The ProtBert - BFD model was pretrained on BFD, a dataset consisting of 2.1 billion protein sequences.

Training Procedure

Preprocessing

The protein sequences are uppercased and tokenized using a single space and a vocabulary size of 21. The inputs of the model are of the form [CLS] Protein Sequence A [SEP] Protein Sequence B [SEP]. The preprocessing step was performed twice, once for a combined length (2 sequences) of less than 512 amino acids, and another time for a combined length (2 sequences) of less than 2048 amino acids. The masking procedure for each sequence follows the original Bert model: 15% of the amino acids are masked, with different replacement rules.

Pretraining

The model was trained on a single TPU Pod V3 - 1024 for one million steps in total. 800k steps were done using a sequence length of 512 (batch size 32k), and 200K steps using a sequence length of 2048 (batch size 6k). The optimizer used is Lamb with a learning rate of 0.002, a weight decay of 0.01, learning rate warm - up for 140k steps, and linear decay of the learning rate after.

Evaluation Results

When fine - tuned on downstream tasks, this model achieves the following results:

🔧 Technical Details

The model is based on the Bert architecture. The main differences from the original Bert version include the way of dealing with sequences as separate documents and the specific masking procedure for protein sequences. The training process is carefully designed, including preprocessing steps and the choice of optimizer and training parameters.

📄 License

Not provided in the original document.

BibTeX entry and citation info

@article {Elnaggar2020.07.12.199554,
	author = {Elnaggar, Ahmed and Heinzinger, Michael and Dallago, Christian and Rehawi, Ghalia and Wang, Yu and Jones, Llion and Gibbs, Tom and Feher, Tamas and Angerer, Christoph and Steinegger, Martin and BHOWMIK, DEBSINDHU and Rost, Burkhard},
	title = {ProtTrans: Towards Cracking the Language of Life{\textquoteright}s Code Through Self-Supervised Deep Learning and High Performance Computing},
	elocation-id = {2020.07.12.199554},
	year = {2020},
	doi = {10.1101/2020.07.12.199554},
	publisher = {Cold Spring Harbor Laboratory},
	abstract = {Computational biology and bioinformatics provide vast data gold-mines from protein sequences, ideal for Language Models (LMs) taken from Natural Language Processing (NLP). These LMs reach for new prediction frontiers at low inference costs. Here, we trained two auto-regressive language models (Transformer-XL, XLNet) and two auto-encoder models (Bert, Albert) on data from UniRef and BFD containing up to 393 billion amino acids (words) from 2.1 billion protein sequences (22- and 112 times the entire English Wikipedia). The LMs were trained on the Summit supercomputer at Oak Ridge National Laboratory (ORNL), using 936 nodes (total 5616 GPUs) and one TPU Pod (V3-512 or V3-1024). We validated the advantage of up-scaling LMs to larger models supported by bigger data by predicting secondary structure (3-states: Q3=76-84, 8 states: Q8=65-73), sub-cellular localization for 10 cellular compartments (Q10=74) and whether a protein is membrane-bound or water-soluble (Q2=89). Dimensionality reduction revealed that the LM-embeddings from unlabeled data (only protein sequences) captured important biophysical properties governing protein shape. This implied learning some of the grammar of the language of life realized in protein sequences. The successful up-scaling of protein LMs through HPC to larger data sets slightly reduced the gap between models trained on evolutionary information and LMs. Availability ProtTrans: \<a href="https://github.com/agemagician/ProtTrans"\>https://github.com/agemagician/ProtTrans\</a\>Competing Interest StatementThe authors have declared no competing interest.},
	URL = {https://www.biorxiv.org/content/early/2020/07/21/2020.07.12.199554},
	eprint = {https://www.biorxiv.org/content/early/2020/07/21/2020.07.12.199554.full.pdf},
	journal = {bioRxiv}
}

Created by Ahmed Elnaggar/@Elnaggar_AI | LinkedIn

📦 Installation

Not provided in the original document.

💡 Usage Tip In some tasks, you could gain more accuracy by fine - tuning the model rather than using it as a feature extractor.