{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T19:20:22Z","timestamp":1776280822604,"version":"3.50.1"},"posted":{"date-parts":[[2026,4,9]]},"group-title":"Bioinformatics","reference-count":48,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T00:00:00Z","timestamp":1775692800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,4,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>Existing RNA 3D structure quality assessment (QA) methods rely on local geometric descriptors or statistical potentials that evaluate atomic-level contacts but are blind to global topological coherence. This creates a critical failure mode\u2014structures that are \u201clocally correct but globally wrong\u201d\u2014where well-formed local helices mask misplaced domains and incorrect overall packing.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>\n                    We introduce SpecRNA-QA, a lightweight RNA QA method based on multi-scale graph-Laplacian features of inter-nucleotide contact networks. In CASP16 leave-one-out cross-validation, it achieves median per-target Spearman\n                    <jats:italic>\u03c1<\/jats:italic>\n                    = 0.69 (target-clustered bootstrap 95% CI [0.64, 0.73]) versus 0.47 for an internal geometry baseline\u2014a +0.22 gap that is significant at\n                    <jats:italic>p<\/jats:italic>\n                    = 1.2 \u00d7 10\n                    <jats:sup>\u221210<\/jats:sup>\n                    (one-sided Wilcoxon signed-rank) and reflects a per-target win rate of 93%. The gain is concentrated on large, multi-domain RNAs, where global coherence is poorly captured by local descriptors. In a contextual comparison with established statistical potentials, local energy-based scores remain strongest on compact RNAs, while SpecRNA-QA yields the strongest signal we observed on targets longer than 200 nt; within the single-threaded runtime budget used here, the strongest local-energy comparator, rsRNASP, timed out on 22 of 26 large targets, and we report an explicit paired head-to-head on the four commonly scored targets in Section 4.2. A training-free heuristic variant further shows that the spectral prior carries intrinsic quality information even in the absence of labeled QA data.\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Availability<\/jats:title>\n                  <jats:p>\n                    SpecRNA-QA is available as a Python package at\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" ext-link-type=\"uri\" xlink:href=\"https:\/\/github.com\/yudabitrends\/specrnaq\">https:\/\/github.com\/yudabitrends\/specrnaq<\/jats:ext-link>\n                    .\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Contact<\/jats:title>\n                  <jats:p>\n                    <jats:email>ybi3@gsu.edu<\/jats:email>\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Supplementary information<\/jats:title>\n                  <jats:p>Supplementary data are available online.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Key Points<\/jats:title>\n                  <jats:list list-type=\"bullet\">\n                    <jats:list-item>\n                      <jats:p>SpecRNA-QA uses multi-scale graph-Laplacian spectra to score global RNA fold coherence that local geometric descriptors and local statistical potentials can miss.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:p>The method uncovers a size-dependent division of labor: on compact RNAs that can be scored exhaustively, atom-level statistical potentials such as rsRNASP remain strongest, whereas on &gt;200 nt RNAs\u2014where the strongest local comparator times out on most targets under the single-threaded runtime budget used here\u2014SpecRNA-QA provides the strongest signal we observed.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:p>Heat-kernel traces at intermediate diffusion times emerge as the most discriminative spectral features and form an interpretable bridge between local packing and long-range tertiary organization.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:p>A training-free heuristic variant of SpecRNA-QA retains informative spectral signal without any labeled QA data, supporting the interpretation of the learned model as amplifying a real structural signal rather than overfitting one.<\/jats:p>\n                    <\/jats:list-item>\n                  <\/jats:list>\n                <\/jats:sec>","DOI":"10.64898\/2026.04.06.716854","type":"posted-content","created":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T18:10:20Z","timestamp":1775758220000},"source":"Crossref","is-referenced-by-count":0,"title":["Spectral Graph Features for Reference-free RNA 3D Quality Assessment"],"prefix":"10.64898","author":[{"given":"Ying","family":"Zhu","sequence":"first","affiliation":[{"name":"Wenzhou Institute, University of Chinese Academy of Sciences, Wenzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huaiwen","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xidian University, Xi\u2019an 710126, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vince D.","family":"Calhoun","sequence":"additional","affiliation":[{"name":"School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuda","family":"Bi","sequence":"additional","affiliation":[{"name":"TReNDS Center, Georgia State University, Atlanta, GA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"issue":"1","key":"2026041511350711000_2026.04.06.716854v2.1","doi-asserted-by":"crossref","first-page":"192","DOI":"10.1002\/prot.70072","article-title":"Assessment of nucleic acid structure prediction in CASP16","volume":"94","year":"2026","journal-title":"Proteins"},{"key":"2026041511350711000_2026.04.06.716854v2.2","doi-asserted-by":"publisher","DOI":"10.1002\/prot.26602"},{"issue":"12","key":"2026041511350711000_2026.04.06.716854v2.3","doi-asserted-by":"crossref","first-page":"2287","DOI":"10.1038\/s41592-024-02487-0","article-title":"Accurate RNA 3D structure prediction using a language model-based deep learning approach","volume":"21","year":"2024","journal-title":"Nature Methods"},{"key":"2026041511350711000_2026.04.06.716854v2.4","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0703836104"},{"key":"2026041511350711000_2026.04.06.716854v2.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.str.2020.05.011"},{"key":"2026041511350711000_2026.04.06.716854v2.6","doi-asserted-by":"publisher","DOI":"10.1038\/nature06684"},{"key":"2026041511350711000_2026.04.06.716854v2.7","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkv1479"},{"key":"2026041511350711000_2026.04.06.716854v2.8","doi-asserted-by":"publisher","DOI":"10.1126\/sci-ence.adl2528"},{"key":"2026041511350711000_2026.04.06.716854v2.9","doi-asserted-by":"publisher","DOI":"10.1101\/2024.10.10.615955"},{"key":"2026041511350711000_2026.04.06.716854v2.10","doi-asserted-by":"publisher","DOI":"10.1101\/2024.11.19.624167"},{"key":"2026041511350711000_2026.04.06.716854v2.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.bpj.2021.11.016"},{"key":"2026041511350711000_2026.04.06.716854v2.12","doi-asserted-by":"publisher","DOI":"10.1089\/cmb.2019.0251"},{"key":"2026041511350711000_2026.04.06.716854v2.13","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btr093"},{"issue":"1","key":"2026041511350711000_2026.04.06.716854v2.14","doi-asserted-by":"crossref","first-page":"lqad016","DOI":"10.1093\/nargab\/lqad016","article-title":"cgRNASP: coarse-grained statistical potentials with residue separation for RNA structure evaluation","volume":"5","year":"2023","journal-title":"NAR Genomics and Bioinformatics"},{"key":"2026041511350711000_2026.04.06.716854v2.15","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkv141"},{"issue":"1","key":"2026041511350711000_2026.04.06.716854v2.16","doi-asserted-by":"crossref","first-page":"2777","DOI":"10.1038\/s41467-021-23100-4","article-title":"Pairing a high-resolution statistical potential with a nucleobase-centric sampling algorithm for improving RNA model refinement","volume":"12","year":"2021","journal-title":"Nature Communications"},{"key":"2026041511350711000_2026.04.06.716854v2.17","doi-asserted-by":"publisher","DOI":"10.1126\/science.abe5650"},{"key":"2026041511350711000_2026.04.06.716854v2.18","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbae064"},{"key":"2026041511350711000_2026.04.06.716854v2.19","doi-asserted-by":"publisher","DOI":"10.1261\/rna.075341.120"},{"key":"2026041511350711000_2026.04.06.716854v2.20","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btq662"},{"key":"2026041511350711000_2026.04.06.716854v2.21","doi-asserted-by":"crossref","unstructured":"Renaud, N. et al. DeepRank: a deep learning framework for data mining 3D protein\u2013protein interfaces. Nature Communications, 12:7068, 2021.","DOI":"10.1038\/s41467-021-27396-0"},{"key":"2026041511350711000_2026.04.06.716854v2.22","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btaa714"},{"issue":"16","key":"2026041511350711000_2026.04.06.716854v2.23","doi-asserted-by":"crossref","first-page":"2332","DOI":"10.1093\/bioinformatics\/btab118","article-title":"VoroCNN: deep convolutional neural network built on 3D Voronoi tessellation of protein structures","volume":"37","year":"2021","journal-title":"Bioinformatics"},{"key":"2026041511350711000_2026.04.06.716854v2.24","unstructured":"Chung, F.R.K. Spectral Graph Theory. American Mathematical Society, 1997."},{"issue":"8","key":"2026041511350711000_2026.04.06.716854v2.25","doi-asserted-by":"crossref","first-page":"814","DOI":"10.1002\/cnm.2655","article-title":"Persistent homology analysis of protein structure, flexibility, and folding","volume":"30","year":"2014","journal-title":"International Journal for Numerical Methods in Biomedical Engineering"},{"issue":"20","key":"2026041511350711000_2026.04.06.716854v2.26","doi-asserted-by":"crossref","first-page":"1502","DOI":"10.1002\/jcc.23953","article-title":"Multidimensional persistence in biomolecular data","volume":"36","year":"2015","journal-title":"Journal of Computational Chemistry"},{"issue":"7","key":"2026041511350711000_2026.04.06.716854v2.27","doi-asserted-by":"crossref","first-page":"e1005690","DOI":"10.1371\/journal.pcbi.1005690","article-title":"TopologyNet: topology based deep convolutional and multi-task neural networks for biomolecular property predictions","volume":"13","year":"2017","journal-title":"PLoS Computational Biology"},{"issue":"9","key":"2026041511350711000_2026.04.06.716854v2.28","doi-asserted-by":"crossref","first-page":"e3376","DOI":"10.1002\/cnm.3376","article-title":"Persistent spectral graph","volume":"36","year":"2020","journal-title":"International Journal for Numerical Methods in Biomedical Engineering"},{"key":"2026041511350711000_2026.04.06.716854v2.29","doi-asserted-by":"crossref","unstructured":"Chen, T. and Guestrin, C. XGBoost: a scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD, pages 785\u2013794, 2016.","DOI":"10.1145\/2939672.2939785"},{"key":"2026041511350711000_2026.04.06.716854v2.30","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btt473"},{"issue":"4","key":"2026041511350711000_2026.04.06.716854v2.31","doi-asserted-by":"crossref","first-page":"702","DOI":"10.1002\/prot.20264","article-title":"Scoring function for automated assessment of protein structure template quality","volume":"57","year":"2004","journal-title":"Proteins: Structure, Function, and Bioinformatics"},{"key":"2026041511350711000_2026.04.06.716854v2.32","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gki524"},{"key":"2026041511350711000_2026.04.06.716854v2.33","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkg571"},{"issue":"1","key":"2026041511350711000_2026.04.06.716854v2.34","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1038\/s41592-023-02086-5","article-title":"Accurate prediction of protein\u2013nucleic acid complexes using RoseTTAFoldNA","volume":"21","year":"2024","journal-title":"Nature Methods"},{"key":"2026041511350711000_2026.04.06.716854v2.35","doi-asserted-by":"crossref","unstructured":"Wang, W. et al. trRosettaRNA: automated prediction of RNA 3D structure with transformer network. Nature Communications, 14:7266, 2023.","DOI":"10.1038\/s41467-023-42528-4"},{"key":"2026041511350711000_2026.04.06.716854v2.36","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkz1108"},{"key":"2026041511350711000_2026.04.06.716854v2.37","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-024-07487-W"},{"issue":"4","key":"2026041511350711000_2026.04.06.716854v2.38","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1007\/s11222-007-9033-z","article-title":"A tutorial on spectral clustering","volume":"17","year":"2007","journal-title":"Statistics and Computing"},{"key":"2026041511350711000_2026.04.06.716854v2.39","doi-asserted-by":"publisher","DOI":"10.1002\/pro.3330"},{"issue":"3","key":"2026041511350711000_2026.04.06.716854v2.40","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1007\/s10100-013-0296-5","article-title":"MCQ4Structures to compute similarity of molecule structures","volume":"22","year":"2014","journal-title":"Central European Journal of Operations Research"},{"issue":"5","key":"2026041511350711000_2026.04.06.716854v2.41","doi-asserted-by":"crossref","first-page":"1383","DOI":"10.1111\/j.1467-8659.2009.01515.x","article-title":"A concise and provably informative multi-scale signature based on heat diffusion","volume":"28","year":"2009","journal-title":"Computer Graphics Forum"},{"key":"2026041511350711000_2026.04.06.716854v2.42","doi-asserted-by":"crossref","unstructured":"Bronstein, M.M. and Kokkinos, I. Scale-invariant heat kernel signatures for non-rigid shape recognition. In 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 1704\u20131711, 2010.","DOI":"10.1109\/CVPR.2010.5539838"},{"key":"2026041511350711000_2026.04.06.716854v2.43","doi-asserted-by":"crossref","unstructured":"Aubry, M. , Schlickewei, U. , and Cremers, D. The wave kernel signature: a quantum mechanical approach to shape analysis. In 2011 IEEE International Conference on Computer Vision Workshops, pages 1626\u20131633, 2011.","DOI":"10.1109\/ICCVW.2011.6130444"},{"key":"2026041511350711000_2026.04.06.716854v2.44","doi-asserted-by":"publisher","DOI":"10.1007\/s00454-002-2885-2"},{"key":"2026041511350711000_2026.04.06.716854v2.45","doi-asserted-by":"publisher","DOI":"10.1090\/S0273-0979-09-01249-X"},{"key":"2026041511350711000_2026.04.06.716854v2.46","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbac024"},{"key":"2026041511350711000_2026.04.06.716854v2.47","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gku972"},{"issue":"3","key":"2026041511350711000_2026.04.06.716854v2.48","doi-asserted-by":"crossref","first-page":"239","DOI":"10.1023\/A:1024068626366","article-title":"Inference for the generalization error","volume":"52","year":"2003","journal-title":"Machine Learning"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.04.06.716854","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T18:35:14Z","timestamp":1776278114000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.04.06.716854"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,9]]},"references-count":48,"URL":"https:\/\/doi.org\/10.64898\/2026.04.06.716854","relation":{},"subject":[],"published":{"date-parts":[[2026,4,9]]},"subtype":"preprint"}}