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.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
The official GitHub repository: https://github.com/agemagician/ProtTrans
Major changes: 1) Added new results that outperform SOT methods on several tasks. 2) Added two new language models, “Electra and T5,” with a total of 5 variations were added. 3) Compared min-, max-, mean-pooling for subcellular Prediction 4) Added a new test set for secondary structure evaluation and moved two of the old sets to SOM. 5) Compared logistic regression, FNN, LSTM & CNN for secondary structure Prediction 6) Added SCOPe t-SNE plots to compare vectors from a) random LM, b) AA frequency, and c) pre-trained LM 7) Added table on dataset statistics, e.g. how many samples from which class in which set.
6. https://github.com/google-research/text-to-text-transfer-transformer