@article{Sen2022,

title = {Characterizing and explaining the impact of disease-associated mutations in proteins without known structures or structural homologs},

author = {Sen, Neeladri

and Anishchenko, Ivan

and Bordin N

and Sillitoe, Ian

and Velankar, Sameer

and Baker, David

and Orengo, Christine},

url = {https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9294430/},

doi = {10.1093/bib/bbac187},

year  = {2022},

date = {2022-07-18},

journal = {Briefings in Bioinformatics},

abstract = {Mutations in human proteins lead to diseases. The structure of these proteins can help understand the mechanism of such diseases and develop therapeutics against them. With improved deep learning techniques, such as RoseTTAFold and AlphaFold, we can predict the structure of proteins even in the absence of structural homologs. We modeled and extracted the domains from 553 disease-associated human proteins without known protein structures or close homologs in the Protein Databank. We noticed that the model quality was higher and the Root mean square deviation (RMSD) lower between AlphaFold and RoseTTAFold models for domains that could be assigned to CATH families as compared to those which could only be assigned to Pfam families of unknown structure or could not be assigned to either. We predicted ligand-binding sites, protein-protein interfaces and conserved residues in these predicted structures. We then explored whether the disease-associated missense mutations were in the proximity of these predicted functional sites, whether they destabilized the protein structure based on ddG calculations or whether they were predicted to be pathogenic. We could explain 80% of these disease-associated mutations based on proximity to functional sites, structural destabilization or pathogenicity. When compared to polymorphisms, a larger percentage of disease-associated missense mutations were buried, closer to predicted functional sites, predicted as destabilizing and pathogenic. Usage of models from the two state-of-the-art techniques provide better confidence in our predictions, and we explain 93 additional mutations based on RoseTTAFold models which could not be explained based solely on AlphaFold models.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Lovelock2022,

title = {The road to fully programmable protein catalysis},

author = {Sarah L. Lovelock and Rebecca Crawshaw and Sophie Basler and Colin Levy and David Baker and Donald Hilvert and Anthony P. Green 

},

url = {https://www.nature.com/articles/s41586-022-04456-z, Nature

https://www.bakerlab.org/wp-content/uploads/2022/06/s41586-022-04456-z.pdf, Download PDF},

doi = {10.1038/s41586-022-04456-z},

year  = {2022},

date = {2022-06-01},

journal = {Nature},

abstract = {The ability to design efficient enzymes from scratch would have a profound effect on chemistry, biotechnology and medicine. Rapid progress in protein engineering over the past decade makes us optimistic that this ambition is within reach. The development of artificial enzymes containing metal cofactors and noncanonical organocatalytic groups shows how protein structure can be optimized to harness the reactivity of nonproteinogenic elements. In parallel, computational methods have been used to design protein catalysts for diverse reactions on the basis of fundamental principles of transition state stabilization. Although the activities of designed catalysts have been quite low, extensive laboratory evolution has been used to generate efficient enzymes. Structural analysis of these systems has revealed the high degree of precision that will be needed to design catalysts with greater activity. To this end, emerging protein design methods, including deep learning, hold particular promise for improving model accuracy. Here we take stock of key developments in the field and highlight new opportunities for innovation that should allow us to transition beyond the current state of the art and enable the robust design of biocatalysts to address societal needs.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Linder2022,

title = {Interpreting neural networks for biological sequences by learning stochastic masks},

author = {Linder, Johannes and La Fleur, Alyssa and Chen, Zibo and Ljubetič, Ajasja and Baker, David and Kannan, Sreeram and Seelig, Georg},

url = {https://www.nature.com/articles/s42256-021-00428-6, Nature Machine Intelligence},

doi = {10.1038/s42256-021-00428-6},

year  = {2022},

date = {2022-01-25},

urldate = {2022-01-25},

journal = {Nature Machine Intelligence},

abstract = {Sequence-based neural networks can learn to make accurate predictions from large biological datasets, but model interpretation remains challenging. Many existing feature attribution methods are optimized for continuous rather than discrete input patterns and assess individual feature importance in isolation, making them ill-suited for interpreting nonlinear interactions in molecular sequences. Here, building on work in computer vision and natural language processing, we developed an approach based on deep learning—scrambler networks—wherein the most important sequence positions are identified with learned input masks. Scramblers learn to predict position-specific scoring matrices where unimportant nucleotides or residues are scrambled by raising their entropy. We apply scramblers to interpret the effects of genetic variants, uncover nonlinear interactions between cis-regulatory elements, explain binding specificity for protein–protein interactions, and identify structural determinants of de novo-designed proteins. We show that scramblers enable efficient attribution across large datasets and result in high-quality explanations, often outperforming state-of-the-art methods.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Baek2022,

title = {Deep learning and protein structure modeling},

author = {Minkyung Baek and David Baker},

url = {https://www.nature.com/articles/s41592-021-01360-8, Nature Methods

https://www.bakerlab.org/wp-content/uploads/2022/01/Baek_Baker_NatureMethods2022_Deep_Learning_and_Protein_Structure_Modeling.pdf, Download PDF

},

doi = {10.1038/s41592-021-01360-8},

year  = {2022},

date = {2022-01-22},

urldate = {2022-01-22},

journal = {Nature Methods},

abstract = {Deep learning has transformed protein structure modeling. Here we relate AlphaFold and RoseTTAFold to classical physically based approaches to protein structure prediction, and discuss the many areas of structural biology that are likely to be affected by further advances in deep learning.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Du2021,

title = {The trRosetta server for fast and accurate protein structure prediction},

author = {Du, Zongyang

and Su, Hong

and Wang, Wenkai

and Ye, Lisha

and Wei, Hong

and Peng, Zhenling

and Anishchenko, Ivan

and Baker, David

and Yang, Jianyi},

url = {https://www.nature.com/articles/s41596-021-00628-9

https://www.bakerlab.org/wp-content/uploads/2022/01/Du_etal_NatProt2021_trRosetta_server.pdf},

doi = {10.1038/s41596-021-00628-9},

year  = {2021},

date = {2021-12-01},

urldate = {2021-12-01},

journal = {Nature Protocols},

abstract = {The trRosetta (transform-restrained Rosetta) server is a web-based platform for fast and accurate protein structure prediction, powered by deep learning and Rosetta. With the input of a protein’s amino acid sequence, a deep neural network is first used to predict the inter-residue geometries, including distance and orientations. The predicted geometries are then transformed as restraints to guide the structure prediction on the basis of direct energy minimization, which is implemented under the framework of Rosetta. The trRosetta server distinguishes itself from other similar structure prediction servers in terms of rapid and accurate de novo structure prediction. As an illustration, trRosetta was applied to two Pfam families with unknown structures, for which the predicted de novo models were estimated to have high accuracy. Nevertheless, to take advantage of homology modeling, homologous templates are used as additional inputs to the network automatically. In general, it takes ~1 h to predict the final structure for a typical protein with ~300 amino acids, using a maximum of 10 CPU cores in parallel in our cluster system. To enable large-scale structure modeling, a downloadable package of trRosetta with open-source codes is available as well. A detailed guidance for using the package is also available in this protocol. The server and the package are available at https://yanglab.nankai.edu.cn/trRosetta/ and https://yanglab.nankai.edu.cn/trRosetta/download/, respectively.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Baek2021b,

title = {Protein oligomer modeling guided by predicted inter-chain contacts in CASP14},

author = {Minkyung Baek and Ivan Anishchenko and Hahnbeom Park and Ian R. Humphreys and David Baker},

url = {https://onlinelibrary.wiley.com/doi/abs/10.1002/prot.26197, Proteins},

doi = {10.1002/prot.26197},

year  = {2021},

date = {2021-07-29},

urldate = {2021-07-29},

journal = {Proteins},

abstract = {For CASP14, we developed deep learning-based methods for predicting homo-oligomeric and hetero-oligomeric contacts and used them for oligomer modeling. To build structure models, we developed an oligomer structure generation method that utilizes predicted inter-chain contacts to guide iterative restrained minimization from random backbone structures. We supplemented this gradient-based fold-and-dock method with template-based and ab initio docking approaches using deep learning-based subunit predictions on 29 assembly targets. These methods produced oligomer models with summed Z-scores 5.5 units higher than the next best group, with the fold-and-dock method having the best relative performance. Over the eight targets for which this method was used, the best of the five submitted models had average oligomer TM-score of 0.71 (average oligomer TM-score of the next best group: 0.64), and explicit modeling of inter-subunit interactions improved modeling of six out of 40 individual domains (ΔGDT-TS > 2.0).



},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Norn2021,

title = {Protein sequence design by conformational landscape optimization},

author = {Norn, Christoffer and Wicky, Basile I. M. and Juergens, David and Liu, Sirui and Kim, David and Tischer, Doug and Koepnick, Brian and Anishchenko, Ivan and Baker, David and Ovchinnikov, Sergey},

url = {https://www.pnas.org/content/118/11/e2017228118, PNAS

https://www.bakerlab.org/wp-content/uploads/2021/03/Norn_etal_PNAS2021_LandscapeOptimization.pdf, Download PDF},

doi = {10.1073/pnas.2017228118},

year  = {2021},

date = {2021-03-16},

urldate = {2021-03-16},

journal = {Proceedings of the National Academy of Sciences},

volume = {118},

number = {11},

abstract = {Almost all proteins fold to their lowest free energy state, which is determined by their amino acid sequence. Computational protein design has primarily focused on finding sequences that have very low energy in the target designed structure. However, what is most relevant during folding is not the absolute energy of the folded state but the energy difference between the folded state and the lowest-lying alternative states. We describe a deep learning approach that captures aspects of the folding landscape, in particular the presence of structures in alternative energy minima, and show that it can enhance current protein design methods.The protein design problem is to identify an amino acid sequence that folds to a desired structure. Given Anfinsen{textquoteright}s thermodynamic hypothesis of folding, this can be recast as finding an amino acid sequence for which the desired structure is the lowest energy state. As this calculation involves not only all possible amino acid sequences but also, all possible structures, most current approaches focus instead on the more tractable problem of finding the lowest-energy amino acid sequence for the desired structure, often checking by protein structure prediction in a second step that the desired structure is indeed the lowest-energy conformation for the designed sequence, and typically discarding a large fraction of designed sequences for which this is not the case. Here, we show that by backpropagating gradients through the transform-restrained Rosetta (trRosetta) structure prediction network from the desired structure to the input amino acid sequence, we can directly optimize over all possible amino acid sequences and all possible structures in a single calculation. We find that trRosetta calculations, which consider the full conformational landscape, can be more effective than Rosetta single-point energy estimations in predicting folding and stability of de novo designed proteins. We compare sequence design by conformational landscape optimization with the standard energy-based sequence design methodology in Rosetta and show that the former can result in energy landscapes with fewer alternative energy minima. We show further that more funneled energy landscapes can be designed by combining the strengths of the two approaches: the low-resolution trRosetta model serves to disfavor alternative states, and the high-resolution Rosetta model serves to create a deep energy minimum at the design target structure.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Hiranuma2021,

title = {Improved protein structure refinement guided by deep learning based accuracy estimation},

author = {Naozumi Hiranuma and Hahnbeom Park and Minkyung Baek and Ivan Anishchenko and Justas Dauparas and David Baker 

},

url = {https://www.nature.com/articles/s41467-021-21511-x, Nature Communications

https://www.bakerlab.org/wp-content/uploads/2021/02/Hiranuma_etal_NatureComms2021_DeepLearningStructureRefinement.pdf, Download PDF},

doi = {10.1038/s41467-021-21511-x},

year  = {2021},

date = {2021-02-26},

urldate = {2021-02-26},

journal = {Nature Communications},

volume = {12},

number = {1340},

abstract = {We develop a deep learning framework (DeepAccNet) that estimates per-residue accuracy and residue-residue distance signed error in protein models and uses these predictions to guide Rosetta protein structure refinement. The network uses 3D convolutions to evaluate local atomic environments followed by 2D convolutions to provide their global contexts and outperforms other methods that similarly predict the accuracy of protein structure models. Overall accuracy predictions for X-ray and cryoEM structures in the PDB correlate with their resolution, and the network should be broadly useful for assessing the accuracy of both predicted structure models and experimentally determined structures and identifying specific regions likely to be in error. Incorporation of the accuracy predictions at multiple stages in the Rosetta refinement protocol considerably increased the accuracy of the resulting protein structure models, illustrating how deep learning can improve search for global energy minima of biomolecules.



},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Ziatdinov2021,

title = {Quantifying the Dynamics of Protein Self-Organization Using Deep Learning Analysis of Atomic Force Microscopy Data},

author = {Ziatdinov, Maxim

and Zhang, Shuai

and Dollar, Orion

and Pfaendtner, Jim

and Mundy, Christopher J.

and Li, Xin

and Pyles, Harley

and Baker, David

and De Yoreo, James J.

and Kalinin, Sergei V.},

url = {https://pubs.acs.org/doi/10.1021/acs.nanolett.0c03447},

doi = {10.1021/acs.nanolett.0c03447},

year  = {2021},

date = {2021-01-13},

journal = {Nano Letters},

abstract = {The dynamics of protein self-assembly on the inorganic surface and the resultant geometric patterns are visualized using high-speed atomic force microscopy. The time dynamics of the classical macroscopic descriptors such as 2D fast Fourier transforms, correlation, and pair distribution functions are explored using the unsupervised linear unmixing, demonstrating the presence of static ordered and dynamic disordered phases and establishing their time dynamics. The deep learning (DL)-based workflow is developed to analyze detailed particle dynamics and explore the evolution of local geometries. Finally, we use a combination of DL feature extraction and mixture modeling to define particle neighborhoods free of physics constraints, allowing for a separation of possible classes of particle behavior and identification of the associated transitions. Overall, this work establishes the workflow for the analysis of the self-organization processes in complex systems from observational data and provides insight into the fundamental mechanisms.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Yang2020,

title = {Improved protein structure prediction using predicted interresidue orientations},

author = {Yang, Jianyi and Anishchenko, Ivan and Park, Hahnbeom and Peng, Zhenling and Ovchinnikov, Sergey and Baker, David},

url = {https://www.pnas.org/content/early/2020/01/01/1914677117

https://www.bakerlab.org/wp-content/uploads/2020/01/Yang2020_ImprovedStructurePredictionInterresidueOrientations.pdf

},

doi = {10.1073/pnas.1914677117},

isbn = {0027-8424},

year  = {2020},

date = {2020-01-02},

journal = {Proceedings of the National Academy of Sciences},

abstract = {Protein structure prediction is a longstanding challenge in computational biology. Through extension of deep learning-based prediction to interresidue orientations in addition to distances, and the development of a constrained optimization by Rosetta, we show that more accurate models can be generated. Results on a set of 18 de novo-designed proteins suggests the proposed method should be directly applicable to current challenges in de novo protein design.The prediction of interresidue contacts and distances from coevolutionary data using deep learning has considerably advanced protein structure prediction. Here, we build on these advances by developing a deep residual network for predicting interresidue orientations, in addition to distances, and a Rosetta-constrained energy-minimization protocol for rapidly and accurately generating structure models guided by these restraints. In benchmark tests on 13th Community-Wide Experiment on the Critical Assessment of Techniques for Protein Structure Prediction (CASP13)- and Continuous Automated Model Evaluation (CAMEO)-derived sets, the method outperforms all previously described structure-prediction methods. Although trained entirely on native proteins, the network consistently assigns higher probability to de novo-designed proteins, identifying the key fold-determining residues and providing an independent quantitative measure of the "ideality" of a protein structure. The method promises to be useful for a broad range of protein structure prediction and design problems.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{Wu2019,

title = {Protein contact prediction using metagenome sequence data and residual neural networks},

author = {Qi Wu and Zhenling Peng and Ivan Anishchenko and Qian Cong and David Baker and Jianyi Yang},

url = {https://academic.oup.com/bioinformatics/article/36/1/41/5512356},

doi = {10.1093/bioinformatics/btz477},

year  = {2019},

date = {2019-06-07},

journal = {Bioinformatics},

volume = {36},

number = {1},

abstract = {Motivation: Almost all protein residue contact prediction methods rely on the availability of deep multiple sequence alignments (MSAs). However, many proteins from the poorly populated families do not have sufficient number of homologs in the conventional UniProt database. Here we aim to solve this issue by exploring the rich sequence data from the metagenome sequencing projects. Results: Based on the improved MSA constructed from the metagenome sequence data, we developed MapPred, a new deep learning-based contact prediction method. MapPred consists of two component methods, DeepMSA and DeepMeta, both trained with the residual neural networks. DeepMSA was inspired by the recent method DeepCov, which was trained on 441 matrices of covariance features. By considering the symmetry of contact map, we reduced the number of matrices to 231, which makes the training more efficient in DeepMSA. Experiments show that DeepMSA outperforms DeepCov by 10–13% in precision. DeepMeta works by combining predicted contacts and other sequence profile features. Experiments on three benchmark datasets suggest that the contribution from the metagenome sequence data is significant with P-values less than 4.04E-17. MapPred is shown to be complementary and comparable the state-of-the-art methods. The success of MapPred is attributed to three factors: the deeper MSA from the metagenome sequence data, improved feature design in DeepMSA and optimized training by the residual neural networks.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}