An accurate nucleic acid-small molecule docking framework via geometric deep learning with large-scale pretraining

• URL: http://arxiv.org/abs/2606.05198v1
• Date: Fri, 22 May 2026 06:39:58 GMT
• Title: An accurate nucleic acid-small molecule docking framework via geometric deep learning with large-scale pretraining
• Abstract summary: NucleoDock is a deep learning framework for nucleic acid-small molecule docking.<n>It combines physics-guided large-scale pretraining on millions of docking-generated synthetic complexes with fine-tuning on curated experimental co-crystal structures.<n>On an external benchmark of 125 nucleic acid-ligand complexes, NucleoDock achieved a top-1 success rate of 56 percent.
• Score: 8.945556760958025
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Nucleic acids are increasingly recognized as therapeutic targets beyond conventional protein-centered drug discovery, yet accurate and efficient docking of small molecules to nucleic acid structures remains challenging. Physics-based docking methods often show limited accuracy and efficiency, whereas deep learning approaches are constrained by the scarcity of experimentally resolved nucleic acid-ligand complexes. Here, we present NucleoDock, a deep learning framework for nucleic acid-small molecule docking. To address data scarcity, NucleoDock combines physics-guided large-scale pretraining on millions of docking-generated synthetic complexes with fine-tuning on curated experimental co-crystal structures. It further integrates sequence- and structure-informed nucleotide representations with atomistic three-dimensional features to capture both biological context and binding-site geometry. A mixture density network-based geometric scoring head is used to model conditional interaction-distance distributions for pose ranking. On an external benchmark of 125 nucleic acid-ligand complexes, NucleoDock achieved a top-1 success rate of 56 percent at an RMSD cutoff of 2.0 Angstrom, outperforming rDock with 29 percent, while generating 100 poses in approximately 5 seconds per complex. Retrospective virtual screening on the ROBIN benchmark further showed improved early enrichment. NucleoDock represents a step toward bridging the methodological gap between protein- and nucleic acid-directed computational drug discovery.

Related papers

• A platform for nuclear symmetry-violation searches with laser-coolable molecules carrying spinful nuclei [0.0]
  We introduce a comprehensive experimental toolbox that integrates cooling, trapping, coherent state manipulation, and a complete precision-measurement protocol.<n>Using $137$Ba$19$F and nuclear-spin-dependent parity violation (NSD-PV) as representative species and benchmark application, our approach achieves a projected statistical sensitivity roughly two orders of magnitude beyond comparable molecular beams.
  arXiv  Detail & Related papers  (2026-08-06T15:09:11Z)
• SigmaDock: Untwisting Molecular Docking With Fragment-Based SE(3) Diffusion [22.586455704388843]
  Generative approaches promise faster, improved, and more diverse pose sampling than physics-based methods.<n>We introduce a novel fragmentation scheme, leveraging inductive biases from structural chemistry, to decompose into rigid-body fragments.<n>We present SigmaDock, an SE(3) diffusion model that generates poses by learning to reassemble these rigid bodies within the binding pocket.
  arXiv  Detail & Related papers  (2025-11-06T22:37:36Z)
• Zero-field identification and control of hydrogen-related electron-nuclear spin registers in diamond [73.17247851945764]
  We introduce an approach to identify the hyperfine components and nuclear spin species of spin defects through measurements on a nearby NV center.<n>Results provide a guide to resolving the defect structures using $textitab initio$ calculations.<n>Our characterization and control tools establish a framework to expand the defect landscape for hybrid electron-nuclear registers.
  arXiv  Detail & Related papers  (2025-10-22T13:50:54Z)
• SE(3)-Equivariant Ternary Complex Prediction Towards Target Protein Degradation [28.648225112411637]
  Targeted protein degradation (TPD) induced by small molecules has emerged as a rapidly evolving modality in drug discovery.<n>DeepTernary is a novel deep learning-based approach that directly predicts ternary structures in an end-to-end manner.
  arXiv  Detail & Related papers  (2025-02-26T06:33:24Z)
• Manifold-Constrained Nucleus-Level Denoising Diffusion Model for Structure-Based Drug Design [81.95343363178662]
  atoms must maintain a minimum pairwise distance to avoid separation violations. NucleusDiff models the interactions between atomic nuclei and their surrounding electron clouds by enforcing the distance constraint. It reduces violation rate by up to 1000% and enhances binding affinity by up to 22.16%, surpassing state-of-the-art models for structure-based drug design.
  arXiv  Detail & Related papers  (2024-09-16T08:42:46Z)
• NovoBench: Benchmarking Deep Learning-based De Novo Peptide Sequencing Methods in Proteomics [58.03989832372747]
  We present the first unified benchmark NovoBench for emphde novo peptide sequencing. It comprises diverse mass spectrum data, integrated models, and comprehensive evaluation metrics. Recent methods, including DeepNovo, PointNovo, Casanovo, InstaNovo, AdaNovo and $pi$-HelixNovo are integrated into our framework.
  arXiv  Detail & Related papers  (2024-06-16T08:23:21Z)
• Multi-scale Iterative Refinement towards Robust and Versatile Molecular Docking [17.28573902701018]
  Molecular docking is a key computational tool utilized to predict the binding conformations of small molecules to protein targets. We introduce DeltaDock, a robust and versatile framework designed for efficient molecular docking.
  arXiv  Detail & Related papers  (2023-11-30T14:09:20Z)
• Pre-Training on Large-Scale Generated Docking Conformations with HelixDock to Unlock the Potential of Protein-ligand Structure Prediction Models [42.16524616409125]
  In this work, we show that by pre-training on a large-scale docking conformation, we can obtain a protein-ligand structure prediction model with outstanding performance. The proposed model, HelixDock, aims to acquire the physical knowledge encapsulated by the physics-based docking tools during the pre-training phase.
  arXiv  Detail & Related papers  (2023-10-21T05:54:26Z)
• Structure Embedded Nucleus Classification for Histopathology Images [51.02953253067348]
  Most neural network based methods are affected by the local receptive field of convolutions. We propose a novel polygon-structure feature learning mechanism that transforms a nucleus contour into a sequence of points sampled in order. Next, we convert a histopathology image into a graph structure with nuclei as nodes, and build a graph neural network to embed the spatial distribution of nuclei into their representations.
  arXiv  Detail & Related papers  (2023-02-22T14:52:06Z)
• Protein-Ligand Complex Generator & Drug Screening via Tiered Tensor Transform [18.509174420141832]
  We develop an algorithm to generate diverse protein-ligand complex conformations for both pose and affinity estimation in drug screening. The 3T structure transformation is decoupled from the system physics, making future usage in other computational scientific domains possible.
  arXiv  Detail & Related papers  (2023-01-03T07:33:20Z)
• State-specific protein-ligand complex structure prediction with a multi-scale deep generative model [68.28309982199902]
  We present NeuralPLexer, a computational approach that can directly predict protein-ligand complex structures. Our study suggests that a data-driven approach can capture the structural cooperativity between proteins and small molecules, showing promise in accelerating the design of enzymes, drug molecules, and beyond.
  arXiv  Detail & Related papers  (2022-09-30T01:46:38Z)
• Explainable Deep Relational Networks for Predicting Compound-Protein Affinities and Contacts [80.69440684790925]
  DeepRelations is a physics-inspired deep relational network with intrinsically explainable architecture. It shows superior interpretability to the state-of-the-art. It boosts the AUPRC of contact prediction 9.5, 16.9, 19.3 and 5.7-fold for the test, compound-unique, protein-unique, and both-unique sets.
  arXiv  Detail & Related papers  (2019-12-29T00:14:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.