-
Agentic End-to-End De Novo Protein Design for Tailored Dynamics Using a Language Diffusion Model
Authors:
Bo Ni,
Markus J. Buehler
Abstract:
Proteins are dynamic molecular machines whose biological functions, spanning enzymatic catalysis, signal transduction, and structural adaptation, are intrinsically linked to their motions. Designing proteins with targeted dynamic properties, however, remains a challenge due to the complex, degenerate relationships between sequence, structure, and molecular motion. Here, we introduce VibeGen, a gen…
▽ More
Proteins are dynamic molecular machines whose biological functions, spanning enzymatic catalysis, signal transduction, and structural adaptation, are intrinsically linked to their motions. Designing proteins with targeted dynamic properties, however, remains a challenge due to the complex, degenerate relationships between sequence, structure, and molecular motion. Here, we introduce VibeGen, a generative AI framework that enables end-to-end de novo protein design conditioned on normal mode vibrations. VibeGen employs an agentic dual-model architecture, comprising a protein designer that generates sequence candidates based on specified vibrational modes and a protein predictor that evaluates their dynamic accuracy. This approach synergizes diversity, accuracy, and novelty during the design process. Via full-atom molecular simulations as direct validation, we demonstrate that the designed proteins accurately reproduce the prescribed normal mode amplitudes across the backbone while adopting various stable, functionally relevant structures. Notably, generated sequences are de novo, exhibiting no significant similarity to natural proteins, thereby expanding the accessible protein space beyond evolutionary constraints. Our work integrates protein dynamics into generative protein design, and establishes a direct, bidirectional link between sequence and vibrational behavior, unlocking new pathways for engineering biomolecules with tailored dynamical and functional properties. This framework holds broad implications for the rational design of flexible enzymes, dynamic scaffolds, and biomaterials, paving the way toward dynamics-informed AI-driven protein engineering.
△ Less
Submitted 14 February, 2025;
originally announced February 2025.
-
MechAgents: Large language model multi-agent collaborations can solve mechanics problems, generate new data, and integrate knowledge
Authors:
Bo Ni,
Markus J. Buehler
Abstract:
Solving mechanics problems using numerical methods requires comprehensive intelligent capability of retrieving relevant knowledge and theory, constructing and executing codes, analyzing the results, a task that has thus far mainly been reserved for humans. While emerging AI methods can provide effective approaches to solve end-to-end problems, for instance via the use of deep surrogate models or v…
▽ More
Solving mechanics problems using numerical methods requires comprehensive intelligent capability of retrieving relevant knowledge and theory, constructing and executing codes, analyzing the results, a task that has thus far mainly been reserved for humans. While emerging AI methods can provide effective approaches to solve end-to-end problems, for instance via the use of deep surrogate models or various data analytics strategies, they often lack physical intuition since knowledge is baked into the parametric complement through training, offering less flexibility when it comes to incorporating mathematical or physical insights. By leveraging diverse capabilities of multiple dynamically interacting large language models (LLMs), we can overcome the limitations of conventional approaches and develop a new class of physics-inspired generative machine learning platform, here referred to as MechAgents. A set of AI agents can solve mechanics tasks, here demonstrated for elasticity problems, via autonomous collaborations. A two-agent team can effectively write, execute and self-correct code, in order to apply finite element methods to solve classical elasticity problems in various flavors (different boundary conditions, domain geometries, meshes, small/finite deformation and linear/hyper-elastic constitutive laws, and others). For more complex tasks, we construct a larger group of agents with enhanced division of labor among planning, formulating, coding, executing and criticizing the process and results. The agents mutually correct each other to improve the overall team-work performance in understanding, formulating and validating the solution. Our framework shows the potential of synergizing the intelligence of language models, the reliability of physics-based modeling, and the dynamic collaborations among diverse agents, opening novel avenues for automation of solving engineering problems.
△ Less
Submitted 14 November, 2023;
originally announced November 2023.
-
ForceGen: End-to-end de novo protein generation based on nonlinear mechanical unfolding responses using a protein language diffusion model
Authors:
Bo Ni,
David L. Kaplan,
Markus J. Buehler
Abstract:
Through evolution, nature has presented a set of remarkable protein materials, including elastins, silks, keratins and collagens with superior mechanical performances that play crucial roles in mechanobiology. However, going beyond natural designs to discover proteins that meet specified mechanical properties remains challenging. Here we report a generative model that predicts protein designs to m…
▽ More
Through evolution, nature has presented a set of remarkable protein materials, including elastins, silks, keratins and collagens with superior mechanical performances that play crucial roles in mechanobiology. However, going beyond natural designs to discover proteins that meet specified mechanical properties remains challenging. Here we report a generative model that predicts protein designs to meet complex nonlinear mechanical property-design objectives. Our model leverages deep knowledge on protein sequences from a pre-trained protein language model and maps mechanical unfolding responses to create novel proteins. Via full-atom molecular simulations for direct validation, we demonstrate that the designed proteins are novel, and fulfill the targeted mechanical properties, including unfolding energy and mechanical strength, as well as the detailed unfolding force-separation curves. Our model offers rapid pathways to explore the enormous mechanobiological protein sequence space unconstrained by biological synthesis, using mechanical features as target to enable the discovery of protein materials with superior mechanical properties.
△ Less
Submitted 15 December, 2023; v1 submitted 16 October, 2023;
originally announced October 2023.
-
Topological design of graphene
Authors:
Bo Ni,
Teng Zhang,
Jiaoyan Li,
Xiaoyan Li,
Huajian Gao
Abstract:
Topological defects (e.g. pentagons, heptagons and pentagon-heptagon pairs) have been widely observed in large scale graphene and have been recognized to play important roles in tailoring the mechanical and physical properties of two-dimensional materials in general. Thanks to intensive studies over the past few years, optimizing properties of graphene through topological design has become a new a…
▽ More
Topological defects (e.g. pentagons, heptagons and pentagon-heptagon pairs) have been widely observed in large scale graphene and have been recognized to play important roles in tailoring the mechanical and physical properties of two-dimensional materials in general. Thanks to intensive studies over the past few years, optimizing properties of graphene through topological design has become a new and promising direction of research. In this chapter, we review some of the recent advances in experimental, computational and theoretical studies on the effects of topological defects on mechanical and physical properties of graphene and applications of topologically designed graphene. The discussions cover out-of-plane effects, inverse problems of designing distributions of topological defects that make a graphene sheet conform to a targeted three-dimensional surface, grain boundary engineering for graphene strength, curved graphene for toughness enhancement and applications in engineering energy materials, multifunctional materials and interactions with biological systems. Despite the rapid developments in experiments and simulations, our understanding on the relations between topological defects and mechanical and physical properties of graphene and other 2D materials is still in its infancy. The intention here is to draw the attention of the research community to some of the open questions in this field.
△ Less
Submitted 1 March, 2018;
originally announced March 2018.
-
Low-temperature heat transport of Nd_{2-x}Ce_{x}CuO_{4} single crystals
Authors:
X. Zhao,
Z. Y. Zhao,
B. Ni,
J. C. Wu,
F. B. Zhang,
J. D. Song,
S. J. Li,
X. F. Sun,
X. G. Li
Abstract:
We report a study of the Ce doping effect on the thermal conductivity (κ) of Nd_{2-x}Ce_{x}CuO_{4} (NCCO) at low temperatures down to 0.3 K and in magnetic fields up to 14 T. It is found that with Ce doping, the electronic thermal conductivity increases; at the same time, the a-axis field induced changes in κ(H), associated with the spin flop and spin polarization of Nd^{3+} sublattice, and the sp…
▽ More
We report a study of the Ce doping effect on the thermal conductivity (κ) of Nd_{2-x}Ce_{x}CuO_{4} (NCCO) at low temperatures down to 0.3 K and in magnetic fields up to 14 T. It is found that with Ce doping, the electronic thermal conductivity increases; at the same time, the a-axis field induced changes in κ(H), associated with the spin flop and spin polarization of Nd^{3+} sublattice, and the spin flop of Cu^{2+} sublattice, gradually disappear. These are clearly due to the electron doping and the destruction of the antiferromagnetic orders. In the superconducting NCCO with x = 0.14 and 0.18, although the electronic thermal conductivity shows sizable field dependencies with H // c, the paramagnetic scattering of phonons is still playing the dominant role in the heat transport, which is different from many other cuprates. In the lightly doped samples (x = 0.03 and 0.06), the low-T κ(H) isotherms with H // c show a step-like anomaly and is likely related to the spin/charge stripes.
△ Less
Submitted 28 July, 2014;
originally announced July 2014.
-
Single crystal growth of the pyrochlores $R_2$Ti$_2$O$_7$ ($R$ = rare earth) by the optical floating-zone method
Authors:
Q. J. Li,
L. M. Xu,
C. Fan,
F. B. Zhang,
Y. Y. Lv,
B. Ni,
Z. Y. Zhao,
X. F. Sun
Abstract:
We report a systematic study on the crystal growth of the rare-earth titanates $R_2$Ti$_2$O$_7$ ($R$ = Gd, Tb, Dy, Ho, Y, Er, Yb and Lu) and Y-doped Tb$_{2-x}$Y$_x$Ti$_2$O$_7$ ($x$ = 0.2 and 1) using an optical floating-zone method. High-quality single crystals were successfully obtained and the growth conditions were carefully optimized. The oxygen pressure was found to be the most important para…
▽ More
We report a systematic study on the crystal growth of the rare-earth titanates $R_2$Ti$_2$O$_7$ ($R$ = Gd, Tb, Dy, Ho, Y, Er, Yb and Lu) and Y-doped Tb$_{2-x}$Y$_x$Ti$_2$O$_7$ ($x$ = 0.2 and 1) using an optical floating-zone method. High-quality single crystals were successfully obtained and the growth conditions were carefully optimized. The oxygen pressure was found to be the most important parameter and the appropriate ones are 0.1--0.4 MPa, depending on the radius of rare-earth ions. The growth rate is another parameter and was found to be 2.5--4 mm/h for different rare-earth ions. X-ray diffraction data demonstrated the good crystallinity of these crystals. The basic physical properties of these crystals were characterized by the magnetic susceptibility and specific heat measurements.
△ Less
Submitted 12 November, 2013;
originally announced November 2013.
-
Low-temperature heat transport of Nd_2CuO_4: Roles of Nd magnons and spin-structure transitions
Authors:
Z. Y. Zhao,
X. M. Wang,
B. Ni,
Q. J. Li,
C. Fan,
W. P. Ke,
W. Tao,
L. M. Chen,
X. Zhao,
X. F. Sun
Abstract:
We report the magnetic-field dependence of thermal conductivity (κ) of an insulating cuprate Nd_2CuO_4 at very low temperatures down to 0.3 K. It is found that apart from the paramagnetic moments scattering on phonons, the Nd^{3+} magnons can act as either heat carriers or phonon scatterers, which strongly depends on the long-range antiferromagnetic transition and the field-induced transitions of…
▽ More
We report the magnetic-field dependence of thermal conductivity (κ) of an insulating cuprate Nd_2CuO_4 at very low temperatures down to 0.3 K. It is found that apart from the paramagnetic moments scattering on phonons, the Nd^{3+} magnons can act as either heat carriers or phonon scatterers, which strongly depends on the long-range antiferromagnetic transition and the field-induced transitions of spin structure. In particular, the Nd^{3+} magnons can effectively transport heat in the spin-flopped state of the Nd^{3+} sublattice. However, both the magnon transport and the magnetic scattering are quenched at very high fields. The spin re-orientations under the in-plane field can be conjectured from the detailed field dependence of κ.
△ Less
Submitted 8 April, 2011;
originally announced April 2011.
-
Global and local critical current density in superconducting SmFeAsO$_{1-x}$F$_x$ measured by two methods
Authors:
E. S. Otabe,
M. Kiuchi,
S. Kawai,
Y. Morita,
J. Ge,
B. Ni,
Z. Gao,
L. Wang,
Y. Qi,
X. Zhang,
Y. Ma
Abstract:
The critical current densities of polycrystalline bulk SmFeAsO$_{1-x}$F$_x$ prepared by the powder-in-tube (PIT) method and by a conventional solid-state reaction were investigated using the remnant magnetic moment method and Campbell's method. Two types of shielding current, corresponding to global and local critical current densities $J_{\rm c}$ were observed using both measurement methods. Th…
▽ More
The critical current densities of polycrystalline bulk SmFeAsO$_{1-x}$F$_x$ prepared by the powder-in-tube (PIT) method and by a conventional solid-state reaction were investigated using the remnant magnetic moment method and Campbell's method. Two types of shielding current, corresponding to global and local critical current densities $J_{\rm c}$ were observed using both measurement methods. The global and local $J_{\rm c}$ were on the order of $10^7$ A/m$^2$ and $10^{10}$ A/m$^2$ at 5 K, respectively. The local $J_{\rm c}$ decreased slightly with increasing magnetic field. The global $J_{\rm c}$ was independent of the preparation method, while the local $J_{\rm c}$ was larger for samples prepared by PIT than for those prepared by solid-state reaction.
△ Less
Submitted 2 July, 2009; v1 submitted 6 February, 2009;
originally announced February 2009.
-
Transition from Mott insulator to superconductor in GaNb$_{4}$Se$_{8}$ and GaTa$_{4}$Se$_{8}$ under high pressure
Authors:
M. M. Abd-Elmeguid,
B. Ni,
D. I. Khomskii,
R. Pocha,
D. Johrendt,
X. Wang,
K. Syassen
Abstract:
Electronic conduction in GaM$_{4}$Se$_{8}$ (M=Nb;Ta) compounds with the fcc GaMo$_{4}$S$_{8}$-type structure originates from hopping of localized unpaired electrons (S=1/2) among widely separated tetrahedral M$_{4}$ metal clusters. We show that under pressure these systems transform from Mott insulators to a metallic and superconducting state with T$_{C}$=2.9 and 5.8K at 13 and 11.5GPa for GaNb…
▽ More
Electronic conduction in GaM$_{4}$Se$_{8}$ (M=Nb;Ta) compounds with the fcc GaMo$_{4}$S$_{8}$-type structure originates from hopping of localized unpaired electrons (S=1/2) among widely separated tetrahedral M$_{4}$ metal clusters. We show that under pressure these systems transform from Mott insulators to a metallic and superconducting state with T$_{C}$=2.9 and 5.8K at 13 and 11.5GPa for GaNb$_{4}$Se$_{8}$ and GaTa$_{4}$Se$_{8}$, respectively. The occurrence of superconductivity is shown to be connected with a pressure-induced decrease of the MSe$_{6}$ octahedral distortion and simultaneous softening of the phonon associated with MSe-bonds.
△ Less
Submitted 12 April, 2005;
originally announced April 2005.
