Before and after AlphaFold2: An overview of protein structure prediction

Abstract

Three-dimensional protein structure is directly correlated with its function and its determination is critical to understanding biological processes and addressing human health and life science problems in general. Although new protein structures are experimentally obtained over time, there is still a large difference between the number of protein sequences placed in Uniprot and those with resolved tertiary structure. In this context, studies have emerged to predict protein structures by methods based on a template or free modeling. In the last years, different methods have been combined to overcome their individual limitations, until the emergence of AlphaFold2, which demonstrated that predicting protein structure with high accuracy at unprecedented scale is possible. Despite its current impact in the field, AlphaFold2 has limitations. Recently, new methods based on protein language models have promised to revolutionize the protein structural biology allowing the discovery of protein structure and function only from evolutionary patterns present on protein sequence. Even though these methods do not reach AlphaFold2 accuracy, they already covered some of its limitations, being able to predict with high accuracy more than 200 million proteins from metagenomic databases. In this mini-review, we provide an overview of the breakthroughs in protein structure prediction before and after AlphaFold2 emergence.

Recent Advances and Challenges in Protein Structure Prediction

“…The motivation behind protein language models is to treat amino acids and protein sequences as words/symbols and sentences in NLP, respectively. Based on the assumption that similar semantics come from amino acids appearing in similar contexts, protein language models can learn evolutionary and functional patterns solely from amino acid sequences . ESMFold, developed by Meta Fundamental AI Research Protein Team (FAIR), is an efficient end-to-end sequence-to-structure method based on a language model.…”

“…For the MSA-free end-to-end methods, the pretrained deep learning models are usually protein language models which drive from natural language processing (NLP). 33 The motivation behind protein language models is to treat amino acids and protein sequences as words/symbols and sentences in NLP, respectively. Based on the assumption that similar semantics come from amino acids appearing in similar contexts, protein language models can learn evolutionary and functional patterns solely from amino acid sequences.…”

“…Based on the assumption that similar semantics come from amino acids appearing in similar contexts, protein language models can learn evolutionary and functional patterns solely from amino acid sequences. 33 ESMFold, 11 developed by Meta Fundamental AI Research Protein Team (FAIR), is an efficient end-to-end sequence-tostructure method based on a language model. It directly infers protein 3D structures at atomic resolution from sequences without using any MSA or template information.…”

Using AlphaFold Predictions in Viral Research

“…According to some studies, it appears that AF2 is unable to predict defects in protein structures caused by mutations [ 79 ]. One investigation showed that differences between mutated and wild-type structures predicted by AlphaFold were extremely small [ 80 ].…”

Rosetta Energy Analysis of AlphaFold2 models: Point Mutations and Conformational Ensembles

“…While these studies are promising, the dogma remains that AlphaFold2 is not able to predict the effect of point mutations. 12–15…”