Application of Sequence Embedding in Protein Sequence-Based Predictions

TL;DR

This paper reviews recent advances in applying sequence embedding techniques from NLP to protein sequences, highlighting their use in contact prediction, secondary structure, and function prediction.

Contribution

It provides a comprehensive overview of various protein sequence embedding methods and their applications in bioinformatics.

Findings

Sequence embeddings improve prediction accuracy

Embedding methods enable new insights into protein structure

Various embedding approaches are reviewed and compared

Abstract

In sequence-based predictions, conventionally an input sequence is represented by a multiple sequence alignment (MSA) or a representation derived from MSA, such as a position-specific scoring matrix. Recently, inspired by the development in natural language processing, several applications of sequence embedding have been observed. Here, we review different approaches of protein sequence embeddings and their applications including protein contact prediction, secondary structure, prediction, and function prediction.