ProteinBERT: a universal deep-learning model of protein sequence and function

Keyphrases

• Annotation Prediction 50%
• Architectural Elements 50%
• Art Performance 50%
• Biological Sequences 50%
• Biophysical Characteristics 50%
• Deep Language 50%
• Deep Language Models 50%
• Deep Learning 100%
• Efficient Framework 50%
• End Processing 50%
• Fast Model 50%
• Gene Ontology Annotation 50%
• Global Representation 50%
• High Efficiency 50%
• Highly Flexible 50%
• Language Modeling 100%
• Language Task 50%
• Limited Labeled Data 50%
• Local Representation 50%
• Long Sequence 50%
• Natural Language 50%
• Post-translational Modification 50%
• Protein Function 100%
• Protein Properties 50%
• Protein Sequence 100%
• Protein Structure 50%
• Repurposing 50%
• Small Animal Model 50%
• Text Analysis 50%

Biochemistry, Genetics and Molecular Biology

• Gene Ontology 100%
• Peptide Sequence 100%
• Posttranslational Modification 100%
• Protein Property 100%
• Protein Structure 100%

Neuroscience

• Amino Acid Sequence 100%
• Posttranslational Modification 100%
• Protein Function 100%
• Protein Structure 100%