Deep learning techniques have significantly impacted protein structure prediction and protein design.

By Robin Pearce, Yang Zhang. 2021

112 citations recorded by the source corpus.

Protein structure prediction and design can be regarded as two inverse processes governed by the same folding principle. Although progress remained stagnant over the past two decades, the recent application of deep neural networks to spatial constraint prediction and end-to-end model training has significantly improved the accuracy of protein structure prediction, largely solving the problem at the fold level for single-domain proteins. The field of protein design has also witnessed dramatic improvement, where noticeable examples have shown that information stored in neural-network models can be used to advance functional protein design. Thus, incorporation of deep learning techniques into different steps of protein folding and design approaches represents an exciting future direction and should continue to have a transformative impact on both fields.

Subfields: Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

1. Robin Pearce
2. Yang Zhang
3. Chengxin Zhang
4. Yang Li
5. Wei Zheng
6. Eric W. Bell
7. Xiaogen Zhou
8. Guijun Zhang

In Molecular Biology

1. Robin Pearce
2. Yang Zhang
3. Chengxin Zhang
4. Eric W. Bell
5. Yang Li
6. Wei Zheng
7. Xiaogen Zhou
8. Guijun Zhang

Citations in the corpus

• Impact of protein conformational diversity on AlphaFold predictions (2021, methodology): “In the last year, the computational tool AlphaFold2 [2] developed by DeepMind, reached an impressive performance in predicting protein structures with an accuracy similar to experimental techniques [3,4].”
• I-TASSER-MTD: a deep-learning-based platform for multi-domain protein structure and function prediction (2022, background): “11 ) and RoseTTAFold 82 ), which are largely a block box to both developer and users 83 , I-TASSER-MTD has several advantages due to the fact that its simulation process is accessible and interpretable.”
• De Novo RNA Tertiary Structure Prediction at Atomic Resolution Using Geometric Potentials from Deep Learning (2022, background): “It is practically encouraging that medium to higher resolution structures could be created for complex folds with few homologous RNA sequences, which has been one of the most challenging problems for deep learning-based protein structure modeling methods 15,17,26 .”