• Modeling protein-small molecule conformational ensembles with PLACER
  Methods Machine Learning Small Molecules Enzymes
  Anishchenko I, Kipnis Y, Kalvet I, Zhou G, Krishna R, Pellock SJ, Lauko A, Lee GR, An L, Dauparas J, DiMaio F, Baker D.
  Proc Natl Acad Sci U S A, 2025 | doi:10.1073/pnas.2427161122
  
  Abstract

  Modeling the conformational heterogeneity of protein-small molecule interactions is important for understanding natural systems and evaluating designed systems but remains an outstanding challenge. We reasoned that while residue-level descriptions of biomolecules are efficient for de novo structure prediction, for probing heterogeneity of interactions with small molecules in the folded state, an entirely atomic-level description could have advantages in speed and generality. We developed a graph neural network called PLACER (protein-ligand atomistic conformational ensemble resolver) trained to recapitulate correct atomic positions from partially corrupted input structures from the Cambridge Structural Database and the Protein Data Bank; the nodes of the graph are the atoms in the system. PLACER accurately generates structures of diverse organic small molecules given knowledge of their atom composition and bonding. When given a description of the larger protein context, it builds up structures of small molecules and protein side chains for protein-small molecule docking. Because PLACER is rapid and stochastic, ensembles of predictions can be readily generated to map conformational heterogeneity. In enzyme design efforts described here and elsewhere, we find that using PLACER to assess the accuracy and preorganization of the designed active sites results in higher success rates and higher activities; we obtain a preorganized retroaldolase with a / of 11,000 Mmin, considerably higher than any pre-deep learning design for this reaction. We anticipate that PLACER will be widely useful for rapidly generating conformational ensembles of small molecule and small molecule-protein systems and for designing higher activity preorganized enzymes.

  PDF
• Parametrically guided design of beta barrels and transmembrane nanopores using deep learning
  Methods
  David Kim, Joseph Watson, David Juergens, Sagardip Majumder, Ria Sonigra, Stacey Gerben, Alex Kang, Asim Bera, Xinting Li, David Baker. Proceedings of the National Academy of Sciences of the United States of America, 2025 | doi:https://doi.org/10.1073/pnas.2425459122
  
  Abstract

  Francis Crick’s global parameterization of coiled coil geometry has been widely useful for guiding design of new protein structures and functions. However, design guided by similar global parameterization of beta barrel structures has been less successful, likely due to the deviations from ideal barrel geometry required to maintain inter-strand hydrogen bonding without introducing backbone strain. Instead, beta barrels have been designed using 2D structural blueprints; while this approach has successfully generated new fluorescent proteins, transmembrane nanopores, and other structures, it requires expert knowledge and provides only indirect control over the global shape. Here we show that the simplicity and control over shape and structure provided by parametric representations can be generalized beyond coiled coils by taking advantage of the rich sequence-structure relationships implicit in RoseTTAFold based design methods. Starting from parametrically generated barrel backbones, both RFjoint inpainting and RFdiffusion readily incorporate backbone irregularities necessary for proper folding with minimal deviation from the idealized barrel geometries. We show that for beta barrels across a broad range of beta sheet parameterizations, these methods achieve high in silico and experimental success rates, with atomic accuracy confirmed by an X-ray crystal structure of a rare barrel topology, and de novo designed transmembrane nanopores with conductances ranging from 200 to 500 pS. By combining the simplicity and control of parametric generation with the high success rates of deep learning based protein design methods, our approach makes the design of proteins where global shape confers function, such as beta barrel nanopores, more precisely specifiable and accessible.

  PDF
• Atomically accurate de novo design of antibodies with RFdiffusion
  Methods Drug Discovery Machine Learning
  Bennett NR, Watson JL, Ragotte RJ, Borst AJ, See DL, Weidle C, Biswas R, Yu Y, Shrock EL, Ault R, Leung PJY, Huang B, Goreshnik I, Tam J, Carr KD, Singer B, Criswell C, Wicky BIM, Vafeados D, Sanchez MG, Kim HM, Torres SV, Chan S, Sun SM, Spear T, Sun Y, O’Reilly K, Maris JM, Sgourakis NG, Melnyk RA, Liu CC, Baker D.
  bioRxiv, 2025 | doi:10.1038/s41586-025-09721-5
  
  Abstract

  Despite the central role that antibodies play in modern medicine, there is currently no method to design novel antibodies that bind a specific epitope entirely . Instead, antibody discovery currently relies on animal immunization or random library screening approaches. Here, we demonstrate that combining computational protein design using a fine-tuned RFdiffusion network alongside yeast display screening enables the generation of antibody variable heavy chains (VHHs) and single chain variable fragments (scFvs) that bind user-specified epitopes with atomic-level precision. To verify this, we experimentally characterized VHH binders to four disease-relevant epitopes using multiple orthogonal biophysical methods, including cryo-EM, which confirmed the proper Ig fold and binding pose of designed VHHs targeting influenza hemagglutinin and toxin B (TcdB). For the influenza-targeting VHH, high-resolution structural data further confirmed the accuracy of CDR loop conformations. While initial computational designs exhibit modest affinity, affinity maturation using OrthoRep enables production of single-digit nanomolar binders that maintain the intended epitope selectivity. We further demonstrate the de novo design of single-chain variable fragments (scFvs), creating binders to TcdB and a Phox2b peptide-MHC complex by combining designed heavy and light chain CDRs. Cryo-EM structural data confirmed the proper Ig fold and binding pose for two distinct TcdB scFvs, with high-resolution data for one design additionally verifying the atomically accurate conformations of all six CDR loops. Our approach establishes a framework for the rational computational design, screening, isolation, and characterization of fully de novo antibodies with atomic-level precision in both structure and epitope targeting.

  PDF

• Multistate and functional protein design using RoseTTAFold sequence space diffusion
  Methods Machine Learning
  Lisanza SL, Gershon JM, Tipps SWK, Sims JN, Arnoldt L, Hendel SJ, Simma MK, Liu G, Yase M, Wu H, Tharp CD, Li X, Kang A, Brackenbrough E, Bera AK, Gerben S, Wittmann BJ, McShan AC, Baker D.
  Nat Biotechnol, 2024 | doi:10.1038/s41587-024-02395-w
  
  Abstract

  Protein denoising diffusion probabilistic models are used for the de novo generation of protein backbones but are limited in their ability to guide generation of proteins with sequence-specific attributes and functional properties. To overcome this limitation, we developed ProteinGenerator (PG), a sequence space diffusion model based on RoseTTAFold that simultaneously generates protein sequences and structures. Beginning from a noised sequence representation, PG generates sequence and structure pairs by iterative denoising, guided by desired sequence and structural protein attributes. We designed thermostable proteins with varying amino acid compositions and internal sequence repeats and cage bioactive peptides, such as melittin. By averaging sequence logits between diffusion trajectories with distinct structural constraints, we designed multistate parent-child protein triples in which the same sequence folds to different supersecondary structures when intact in the parent versus split into two child domains. PG design trajectories can be guided by experimental sequence-activity data, providing a general approach for integrated computational and experimental optimization of protein function.

  PDF
• Generalized biomolecular modeling and design with RoseTTAFold All-Atom
  Methods
  Krishna R, Wang J, Ahern W, Sturmfels P, Venkatesh P, Kalvet I, Lee GR, Morey-Burrows FS, Anishchenko I, Humphreys IR, McHugh R, Vafeados D, Li X, Sutherland GA, Hitchcock A, Hunter CN, Kang A, Brackenbrough E, Bera AK, Baek M, DiMaio F, Baker D.
  Science, 2024 | doi:10.1126/science.adl2528
  
  Abstract

  Deep-learning methods have revolutionized protein structure prediction and design but are presently limited to protein-only systems. We describe RoseTTAFold All-Atom (RFAA), which combines a residue-based representation of amino acids and DNA bases with an atomic representation of all other groups to model assemblies that contain proteins, nucleic acids, small molecules, metals, and covalent modifications, given their sequences and chemical structures. By fine-tuning on denoising tasks, we developed RFdiffusion All-Atom (RFdiffusionAA), which builds protein structures around small molecules. Starting from random distributions of amino acid residues surrounding target small molecules, we designed and experimentally validated, through crystallography and binding measurements, proteins that bind the cardiac disease therapeutic digoxigenin, the enzymatic cofactor heme, and the light-harvesting molecule bilin.

  PDF
• Improving Protein Expression, Stability, and Function with ProteinMPNN
  Methods
  Sumida KH, Núñez-Franco R, Kalvet I, Pellock SJ, Wicky BIM, Milles LF, Dauparas J, Wang J, Kipnis Y, Jameson N, Kang A, De La Cruz J, Sankaran B, Bera AK, Jiménez-Osés G, Baker D.
  J Am Chem Soc, 2024 | doi:10.1021/jacs.3c10941
  
  Abstract

  Natural proteins are highly optimized for function but are often difficult to produce at a scale suitable for biotechnological applications due to poor expression in heterologous systems, limited solubility, and sensitivity to temperature. Thus, a general method that improves the physical properties of native proteins while maintaining function could have wide utility for protein-based technologies. Here, we show that the deep neural network ProteinMPNN, together with evolutionary and structural information, provides a route to increasing protein expression, stability, and function. For both myoglobin and tobacco etch virus (TEV) protease, we generated designs with improved expression, elevated melting temperatures, and improved function. For TEV protease, we identified multiple designs with improved catalytic activity as compared to the parent sequence and previously reported TEV variants. Our approach should be broadly useful for improving the expression, stability, and function of biotechnologically important proteins.

  PDF

Lab-Led

• De novo design of high-affinity binders of bioactive helical peptides
  Methods Peptides Machine Learning
  Vázquez Torres S, Leung PJY, Venkatesh P, Lutz ID, Hink F, Huynh HH, Becker J, Yeh AH, Juergens D, Bennett NR, Hoofnagle AN, Huang E, MacCoss MJ, Expòsit M, Lee GR, Bera AK, Kang A, De La Cruz J, Levine PM, Li X, Lamb M, Gerben SR, Murray A, Heine P, Korkmaz EN, Nivala J, Stewart L, Watson JL, Rogers JM, Baker D.
  Nature, 2024 | doi:10.1038/s41586-023-06953-1
  
  Abstract

  Many peptide hormones form an α-helix on binding their receptors, and sensitive methods for their detection could contribute to better clinical management of disease. De novo protein design can now generate binders with high affinity and specificity to structured proteins. However, the design of interactions between proteins and short peptides with helical propensity is an unmet challenge. Here we describe parametric generation and deep learning-based methods for designing proteins to address this challenge. We show that by extending RFdiffusion to enable binder design to flexible targets, and to refining input structure models by successive noising and denoising (partial diffusion), picomolar-affinity binders can be generated to helical peptide targets by either refining designs generated with other methods, or completely de novo starting from random noise distributions without any subsequent experimental optimization. The RFdiffusion designs enable the enrichment and subsequent detection of parathyroid hormone and glucagon by mass spectrometry, and the construction of bioluminescence-based protein biosensors. The ability to design binders to conformationally variable targets, and to optimize by partial diffusion both natural and designed proteins, should be broadly useful.

  PDF

Collaborator-Led

• Peptide-binding specificity prediction using fine-tuned protein structure prediction networks
  Methods
  Motmaen A, Dauparas J, Baek M, Abedi MH, Baker D, Bradley P.
  Proc Natl Acad Sci U S A, 2023 | doi:10.1073/pnas.2216697120
  
  Abstract

  Peptide-binding proteins play key roles in biology, and predicting their binding specificity is a long-standing challenge. While considerable protein structural information is available, the most successful current methods use sequence information alone, in part because it has been a challenge to model the subtle structural changes accompanying sequence substitutions. Protein structure prediction networks such as AlphaFold model sequence-structure relationships very accurately, and we reasoned that if it were possible to specifically train such networks on binding data, more generalizable models could be created. We show that placing a classifier on top of the AlphaFold network and fine-tuning the combined network parameters for both classification and structure prediction accuracy leads to a model with strong generalizable performance on a wide range of Class I and Class II peptide-MHC interactions that approaches the overall performance of the state-of-the-art NetMHCpan sequence-based method. The peptide-MHC optimized model shows excellent performance in distinguishing binding and non-binding peptides to SH3 and PDZ domains. This ability to generalize well beyond the training set far exceeds that of sequence-only models and should be particularly powerful for systems where less experimental data are available.

  PDF

• Robust deep learning-based protein sequence design using ProteinMPNN
  Methods Machine Learning MPNN
  Dauparas J, Anishchenko I, Bennett N, Bai H, Ragotte RJ, Milles LF, Wicky BIM, Courbet A, de Haas RJ, Bethel N, Leung PJY, Huddy TF, Pellock S, Tischer D, Chan F, Koepnick B, Nguyen H, Kang A, Sankaran B, Bera AK, King NP, Baker D.
  Science, 2022 | doi:10.1126/science.add2187
  
  Abstract

  Although deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here, we describe a deep learning-based protein sequence design method, ProteinMPNN, that has outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4% compared with 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using x-ray crystallography, cryo-electron microscopy, and functional studies by rescuing previously failed designs, which were made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target-binding proteins.

  PDF