ProtTrans: Towards Cracking the Language of Life's Code Through Self-Supervised Deep Learning and High Performance Computing

Abstract

Computational biology and bioinformatics provide vast data gold-mines from protein sequences, ideal for Language Models taken from NLP. These LMs reach for new prediction frontiers at low inference costs. Here, we trained two auto-regressive models (Transformer-XL, XLNet) and four auto-encoder models (BERT, Albert, Electra, T5) on data from UniRef and BFD containing up to 393 billion amino acids. The LMs were trained on the Summit supercomputer using 5616 GPUs and TPU Pod up-to 1024 cores. Dimensionality reduction revealed that the raw protein LM-embeddings from unlabeled data captured some biophysical features of protein sequences. We validated the advantage of using the embeddings as exclusive input for several subsequent tasks. The first was a per-residue prediction of protein secondary structure (3-state accuracy Q3=81%-87%); the second were per-protein predictions of protein sub-cellular localization (ten-state accuracy: Q10=81%) and membrane vs. water-soluble (2-state accuracy Q2=91%). For the per-residue predictions the transfer of the most informative embeddings (ProtT5) for the first time outperformed the state-of-the-art without using evolutionary information thereby bypassing expensive database searches. Taken together, the results implied that protein LMs learned some of the grammar of the language of life. To facilitate future work, we released our models at https://github.com/agemagician/ProtTrans.

arXiv comment17 pages, 9 figures, 4 tables

References in corpus (24)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Language Models are Few-Shot Learners
• An Overview of Multi-Task Learning in Deep Neural Networks
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• SQuAD: 100,000+ Questions for Machine Comprehension of Text
• mT5: A massively multilingual pre-trained text-to-text transformer
• Horovod: fast and easy distributed deep learning in TensorFlow
• Generating Long Sequences with Sparse Transformers
• Big Bird: Transformers for Longer Sequences
• CodeBERT: A Pre-Trained Model for Programming and Natural Languages
• Multi-Task Deep Neural Networks for Natural Language Understanding
• Learning protein sequence embeddings using information from structure
• Evaluating Protein Transfer Learning with TAPE
• StaQC: A Systematically Mined Question-Code Dataset from Stack Overflow
• ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
• Attention Interpretability Across NLP Tasks
• A Closer Look at Deep Learning Heuristics: Learning rate restarts, Warmup and Distillation
• Unsupervised Translation of Programming Languages
• TFLMS: Large Model Support in TensorFlow by Graph Rewriting
• Pre-Training of Deep Bidirectional Protein Sequence Representations with Structural Information
• Neural Program Search: Solving Programming Tasks from Description and Examples
• Optimal Gradient Checkpoint Search for Arbitrary Computation Graphs

Cited by in corpus (9)

• Exploring the Limits of Out-of-Distribution Detection
• Pre-trained Language Models in Biomedical Domain: A Systematic Survey
• Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures
• Profile Prediction: An Alignment-Based Pre-Training Task for Protein Sequence Models
• White paper: The Helix Pathogenicity Prediction Platform
• Evolution Is All You Need: Phylogenetic Augmentation for Contrastive Learning
• Pre-training Protein Language Models with Label-Agnostic Binding Pairs Enhances Performance in Downstream Tasks
• A Brief Review of Machine Learning Techniques for Protein Phosphorylation Sites Prediction
• Learning to Extend Program Graphs to Work-in-Progress Code