arXiv subjects

Han Tang

Publications and source records attributed to Han Tang.

6 recordsLinked to original sources

arXiv2025

UNAAGI: Atom-Level Diffusion for Generating Non-Canonical Amino Acid Substitutions

Proposing beneficial amino acid substitutions, whether for mutational effect prediction or protein engineering, remains a central challenge in structural biology. Recent inverse folding models, trained to reconstruct sequences from structure, have had considerable impact in identifying functional mutations. However, current approaches are constrained to designing sequences composed exclusively of natural amino acids (NAAs). The larger set of non-canonical amino acids (NCAAs), which offer greater chemical diversity, and are frequently used in in-vivo protein engineering, remain largely inaccessible for current variant effect prediction methods. To address this gap, we introduce \textbf{UNAAGI}, a diffusion-based generative model that reconstructs residue identities from atomic-level structure using an E(3)-equivariant framework. By modeling side chains in full atomic detail rather than as discrete tokens, UNAAGI enables the exploration of both canonical and non-canonical amino acid substitutions within a unified generative paradigm. We evaluate our method on experimentally benchmarked mutation effect datasets and demonstrate that it achieves substantially improved performance on NCAA substitutions compared to the current state-of-the-art. Furthermore, our results suggest a shared methodological foundation between protein engineering and structure-based drug design, opening the door for a unified training framework across these domains.

q-bio.BM↗

arXiv2025

Enhancing Oxygen Reduction Reaction on Pt-Based Electrocatalysts through Surface Decoration for Improved OH Reduction Equilibrium and Reduced H2O Adsorption

Electrochemical energy and substance conversion devices involve complex electrode processes, characterized by multiple charge transfer steps, competing pathways, and various intermediates. Such complexity makes it challenging to enhance electrocatalytic activity. The prevailing strategy typically focuses on optimizing the geometric and electronic structures of the electrocatalysts to align the adsorption energies of reaction intermediates with the peak of the activity Volcano curve. In this study, we demonstrate that surface decoration can effectively shape the micro reaction environment for the model system of oxygen reduction reaction (ORR) on Pt electrodes. By applying a partial hydrophobic I* adlayer on the Pt surface, we can shift the equilibrium of OH* reduction and weaken H2O* adsorption, which significantly enhances ORR kinetics. With in situ scan tunneling microscopy (STM) and theoretical calculations, our study reveals the formation of isolated Pt2 surface units situated in a hydrophobic valley surrounded by adsorbed iodine atoms. This minimalist Pt2 active unit exhibits significantly greater activity for ORR compared to an extended Pt surface. This strategy could pave the way for developing highly efficient catalysts with potential applications in fuel cell technology and metal air batteries and extension to other electrochemical conversion reactions such as ammonia synthesis and CO2 reduction.

physics.chem-ph↗

arXiv2025

Dynamics of chiral phase transition in a $N_f=2+1$ soft-wall AdS/QCD model

We investigate the real-time dynamics of the chiral phase transition in a soft-wall AdS/QCD model, of which the mass plane phase diagram from equilibrium calculation is qualitatively consistent with the so-called Columbia plot. By directly solving the non-equilibrium evolution of the order parameter of the chiral phase transition, i.e. the chiral condensate, we study the thermalization of the QCD matter in different regions of the quark mass plane. It is shown that, when the system is close to the transition region, the thermalization process will show non-trivial behavior in the intermediate time region.

hep-ph↗

arXiv2024

Contextual Molecule Representation Learning from Chemical Reaction Knowledge

In recent years, self-supervised learning has emerged as a powerful tool to harness abundant unlabelled data for representation learning and has been broadly adopted in diverse areas. However, when applied to molecular representation learning (MRL), prevailing techniques such as masked sub-unit reconstruction often fall short, due to the high degree of freedom in the possible combinations of atoms within molecules, which brings insurmountable complexity to the masking-reconstruction paradigm. To tackle this challenge, we introduce REMO, a self-supervised learning framework that takes advantage of well-defined atom-combination rules in common chemistry. Specifically, REMO pre-trains graph/Transformer encoders on 1.7 million known chemical reactions in the literature. We propose two pre-training objectives: Masked Reaction Centre Reconstruction (MRCR) and Reaction Centre Identification (RCI). REMO offers a novel solution to MRL by exploiting the underlying shared patterns in chemical reactions as \textit{context} for pre-training, which effectively infers meaningful representations of common chemistry knowledge. Such contextual representations can then be utilized to support diverse downstream molecular tasks with minimum finetuning, such as affinity prediction and drug-drug interaction prediction. Extensive experimental results on MoleculeACE, ACNet, drug-drug interaction (DDI), and reaction type classification show that across all tested downstream tasks, REMO outperforms the standard baseline of single-molecule masked modeling used in current MRL. Remarkably, REMO is the pioneering deep learning model surpassing fingerprint-based methods in activity cliff benchmarks.

cs.LG↗

arXiv2023

Masses of the conjectured H-dibaryon at different temperatures

We present a lattice QCD determination of masses of the conjectured H-dibaryon $m_H$ at nine different temperatures $T/T_c =0.24, 0.63, 0.76, 0.84, 0.95, 1.09, 1.27, 1.52, 1.90$. In the meantime, the masses of baryon $N$, $\Sigma$, $\Xi$ and $\Lambda$ at different temperatures are also computed. The simulation is performed on anisotropic lattice with $N_f=2+1$ flavours of clover fermion at quark mass which corresponds to $m_\pi=384(4) {\rm MeV} $. The thermal ensembles were provided by the FASTSUM collaboration and the zero temperature ensembles by the Hadspec collaboration. We also calculate the spectral density of the correlation function of those particles. The spectral density distributions show rich peak structure at the lowest temperature, while at intermediate temperatures, the mass values of those particles obtained by extrapolation method reflect a two-peak structure. While the spectral density for octet baryon becomes smooth at $T/T_c = 1.27, 1.52, 1.90$, the spectral density for H-dibaryon becoms smooth at $T/T_c = 1.90$. At $T/T_c =0.24 $, the mass difference of H-dibaryon and $\Lambda$ pair $\Delta m = m_H - 2\,m_{\Lambda} $ is estimated to be $\Delta m = -14.6(6.2) {\rm MeV}$ which suggests there exists a bound H-dibaryon state.

hep-lat↗

arXiv2022

Sparse Mixture Once-for-all Adversarial Training for Efficient In-Situ Trade-Off Between Accuracy and Robustness of DNNs

Existing deep neural networks (DNNs) that achieve state-of-the-art (SOTA) performance on both clean and adversarially-perturbed images rely on either activation or weight conditioned convolution operations. However, such conditional learning costs additional multiply-accumulate (MAC) or addition operations, increasing inference memory and compute costs. To that end, we present a sparse mixture once for all adversarial training (SMART), that allows a model to train once and then in-situ trade-off between accuracy and robustness, that too at a reduced compute and parameter overhead. In particular, SMART develops two expert paths, for clean and adversarial images, respectively, that are then conditionally trained via respective dedicated sets of binary sparsity masks. Extensive evaluations on multiple image classification datasets across different models show SMART to have up to 2.72x fewer non-zero parameters costing proportional reduction in compute overhead, while yielding SOTA accuracy-robustness trade-off. Additionally, we present insightful observations in designing sparse masks to successfully condition on both clean and perturbed images.

cs.CV↗