Improved protein structure prediction using potentials from deep learning

By Andrew W. Senior, Richard Evans, John M. Jumper, James Kirkpatrick, Laurent Sifre, Tim Green, Chongli Qin, Augustin Žídek, A. Nelson, Alex Bridgland and 9 more. 2020

2,936 citations recorded by the source corpus.

Subfields: Materials Chemistry, Molecular Biology, Computational Biology & Bioinformatics

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

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Accurate prediction of protein structures and interactions using a 3-track neural network (2021, positive): “The biannual Critical Assessment of Structure Prediction (CASP) meetings have demonstrated that deep-learning methods such as AlphaFold ( 1 , 2 ) and trRosetta ( 3 ), which extract information fromthe largedata-baseofknownproteinstructuresintheProteinDataBank(PDB),outperformmoretraditional approache”
Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences (2019, background): “State-of-the-art neural networks for tertiary structure and contact prediction use deep residual architectures with two-dimensional convolutions over pairwise feature maps to output a contact prediction or distance potential for each pair of residues (54, 56, 59).”
AlphaFold2 and its applications in the fields of biology and medicine (2023, methodology): “47 A detailed description for the architecture of AF2 can be found in reference.”
Highly accurate protein structure prediction with AlphaFold (2021, methodology): “Despite the long history of applying neural networks to structure prediction 14,42,43 , they have only recently come to improve structure prediction 10,11,44,45 .”
De novo protein design by deep network hallucination (2020, background): “(6) -- DKL(P||Q) -- is the same as in Eq.(2): it measures the quality of the generated distance and orientation maps with no regard to the 3D structure.”
Learning inverse folding from millions of predicted structures (2022, methodology): “We rank the sequences by distogram LDDT scores (Senior et al., 2020) as a proxy for the quality of the predictions.”
Continual World: A Robotic Benchmark For Continual Reinforcement Learning (2021, background): “assumption, as for example in image processing [27, 46, 9, 38], language modelling [49, 10] or biological applications [44].”
GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation (2022, background): “Most recently, AlphaFold Senior et al. (2020); Jumper et al. (2021) has signiﬁcantly improved the performance of protein structure generation.”
Multi-Scale Representation Learning on Proteins (2022, methodology): “This limitation could however be partly remedied, (i.) by the recent advancements in protein structure prediction [Senior et al., 2020, Jumper et al., 2021, AlphaFold] [Baek et al., 2021, RoseTTAFold] and protein structure determination methods such as cryo-electron microscopy [Callaway, 2020], and…”
Transformer protein language models are unsupervised structure learners (2020, methodology): “Supervised contact prediction Recently, supervised methods using deep learning have resulted in breakthrough results in supervised contact prediction (Wang et al., 2017; Jones & Kandathil, 2018; Yang et al., 2019; Senior et al., 2020; Adhikari & Elofsson, 2020).”
Artificial intelligence to deep learning: machine intelligence approach for drug discovery (2021, methodology): “…of proteins in two steps: (i) firstly, using a CNN it transforms an amino acid sequence of a protein to distance matrix as well as a torsion angle matrix, (ii) secondly, using a gradient optimization technique it translates these two matrices into the three-dimensional structure of a protein [75].”