Deep Semantic Protein Representation for Annotation, Discovery, and Engineering

As, Schwartz; Gj, Hannum; Zr, Dwiel; Me, Smoot; Ar, Grant; Jm, Knight; Sa, Becker; Eads,; LaFave, Matthew C.; Eavani, Harini; Y, Liu; Ak, Banerjee; Th, Richardson

doi:10.1101/365965

Cited by 20 publications

(17 citation statements)

References 38 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These sequences contain information about the sequence motifs and patterns that result in a functional protein, and the structural and functional annotations provide clues as to how structure and function arise from sequence. These annotations can be learned from sequences, 74 and embeddings trained on these annotations may be able to transfer knowledge from UniProt to specific problems of interest. 82 These large quantities of unlabeled or partially-labeled sequence data may also enable machine-learning models to generate artificial protein diversity leading to novel protein functions.…”

Section: Discussionmentioning

confidence: 99%

Machine-learning-guided directed evolution for protein engineering

2019

View full text Add to dashboard Cite

Machine learning (ML)-guided directed evolution is a new paradigm for biological design that enables optimization of complex functions. ML methods use data to predict how sequence maps to function without requiring a detailed model of the underlying physics or biological pathways. To demonstrate ML-guided directed evolution, we introduce the steps required to build ML sequence-function models and use them to guide engineering, making recommendations at each stage. This review covers basic concepts relevant to using ML for protein engineering as well as the current literature and applications of this new engineering paradigm. ML methods accelerate directed evolution by learning from information contained in all measured variants and using that information to select sequences that are likely to be improved. We then provide two case studies that demonstrate the ML-guided directed evolution process. We also look to future opportunities where ML will enable discovery of new protein functions and uncover the relationship between protein sequence and function.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine-learning-guided directed evolution for protein engineering

2019

View full text Add to dashboard Cite

show abstract

“…If the embedding TABLE II ACCURACY ON TEST DATA WHEN TRAINING AND TESTING OUR NETWORK WITH DIFFERENT AMOUNTS OF CLASSES. WE ALSO COMPARE TO OTHER PREVIOUSLY PUBLISHED NETWORK ARCHITECTURE DSPACE [21].…”

Section: Embedding Analysismentioning

confidence: 99%

Unaligned Sequence Similarity Search Using Deep Learning

Senter¹,

Royalty

Steen

et al. 2019

2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

View full text Add to dashboard Cite

Gene annotation has traditionally required direct comparison of DNA sequences between an unknown gene and a database of known ones using string comparison methods. However, these methods do not provide useful information when a gene does not have a close match in the database. In addition, each comparison can be costly when the database is large since it requires alignments and a series of string comparisons. In this work we propose a novel approach: using recurrent neural networks to embed DNA or amino-acid sequences in a lowdimensional space in which distances correlate with functional similarity. This embedding space overcomes both shortcomings of the method of aligning sequences and comparing homology. First, it allows us to obtain information about genes which do not have exact matches by measuring their similarity to other ones in the database. If our database is labeled this can provide labels for a query gene as is done in traditional methods. However, even if the database is unlabeled it allows us to find clusters and infer some characteristics of the gene population. In addition, each comparison is much faster than traditional methods since the distance metric is reduced to the Euclidean distance, and thus efficient approximate nearest neighbor algorithms can be used to find the best match. We present results showing the advantage of our algorithm. More specifically we show how our embedding can be useful for both classification tasks when our labels are known, and clustering tasks where our sequences belong to classes which have not been seen before.

show abstract

“…Deep learning approaches achieving state-of-the-art performance in different machine learning applications are able to uncover representations in an end-to-end manner replacing classical feature engineering. Protein structure prediction (AlQuraishi 2019), , function prediction (X. Liu 2017; Asgari and Mofrad 2015;Zhou et al 2019) and semantic search (Schwartz et al 2018) are promising state-of-the-art applications of deep networks in protein informatics. All of these approaches are, nonetheless, domain-specific and few studies have proposed a universal representation for proteins (Alley et al 2019;Asgari and Mofrad 2015) as a challenging task.…”

Section: Introductionmentioning

confidence: 99%

TripletProt: Deep Representation Learning of Proteins based on Siamese Networks

Nourani

Asgari²,

McHardy

et al. 2020

Preprint

View full text Add to dashboard Cite

We introduce TripletProt, a new approach for protein representation learning based on the Siamese neural networks. We evaluate TripletProt comprehensively in protein functional annotation tasks including sub-cellular localization (14 categories) and gene ontology prediction (more than 2000 classes), which are both challenging multi-class multi-label classification machine learning problems. We compare the performance of TripletProt with the state-of-the-art approaches including recurrent language model-based approach (i.e., UniRep), as well as proteinprotein interaction (PPI) network and sequence-based method (i.e., DeepGO). Our TripletProt showed an overall improvement of F1 score in the above mentioned comprehensive functional annotation tasks, solely relying on the PPI network. TripletProt and in general Siamese Network offer great potentials for the protein informatics tasks and can be widely applied to similar tasks.

show abstract

Deep Semantic Protein Representation for Annotation, Discovery, and Engineering

Cited by 20 publications

References 38 publications

Machine-learning-guided directed evolution for protein engineering

Machine-learning-guided directed evolution for protein engineering

Unaligned Sequence Similarity Search Using Deep Learning

TripletProt: Deep Representation Learning of Proteins based on Siamese Networks

Contact Info

Product

Resources

About