Can automated copy number variation pathogenicity prediction tools replace manual review under the ACMG framework? A clinical validation

Abstract

Background: Copy Number Variation (CNV) is associated with numerous complex diseases, yet their pathogenicity assessment remains heavily dependent on manual expert review, a process that is cumbersome and time-consuming. Although a variety of automated tools for CNV pathogenicity prediction have been developed, systematic assessments of their performance and clinical generalizability under the five-category classification framework defined by ACMG standards remain scarce. Methods: This study constructed an expert-reviewed CNV benchmark dataset. Under the ACMG five-category system, it systematically compared the classification prediction capabilities of five prediction tools across both deletion and duplication CNV samples. Furthermore, it assessed their applicability in real clinical samples and evaluated potential biases towards specific variations. Results: The study indicates that different tools exhibit performance variations and specialised strengths across different classification tasks, yet overall capability in distinguishing clinical risk boundaries across the five categories remains inadequate. Among these, AnnotSV demonstrated outstanding performance in predicting pathogenic samples, achieving accuracy rates of 95.16% and 86.39% on the deletion and duplication benchmark datasets respectively. ClassifyCNV demonstrated optimal performance in variant of uncertain significance classification with an accuracy of 95.33%. Furthermore, all models exhibited significantly superior five-category discrimination capabilities for deletion samples compared with duplication samples, revealing pronounced performance biases across variant types. Conclusion: Current automated CNV prediction tools demonstrate higher discriminatory capacity for deletion variants than for duplications; however, they remain insufficient to fully replace manual expert review within the ACMG five-category framework. Further optimisation using multicentre, consistently curated clinical gold-standard datasets is required to improve their reliability and clinical applicability.

Keywords

Automated tools / Copy number variation / Machine learning / Performance analysis / Variant interpretation

CRediT authorship contribution statement

Y Zhang: data curation, formal analysis, methodology, visualization, and writing—review and editing. X Zhang: conceptualization, supervision, funding acquisition. D Chen: conceptualization, supervision, project administration, and writing—original draft, review, and editing.

Ethics approval and consent to participate

The study protocol was reviewed and approved by the Institutional Review Board of the Third Affiliated Hospital of Shenzhen University (Approval No.: 2026-LHQRMYY-KYLL-001).

Consent for publication

Not applicable

Funding

This work was supported by the Shenzhen Medical Key Discipline Construction (SZXK054).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

Not applicable

References

[1] A.J. Marian, Clinical Interpretation and Management of Genetic Variants, JACC Basic Transl. Sci. 5 (2020) 1029-1042, https://doi.org/10.1016/j.jacbts.2020.05.013

[2] M. Pariyar, A. Johns, R.F. Thorne, R.J. Scott, K.A. Avery—Kiejda, Copy number variation in triple negative breast cancer samples associated with lymph node metastasis, Neoplasia 23 (2021) 743-753, https://doi.org/10.1016/j.neo.2021.05.016

[3] I. Kushima, B. Aleksic, M. Nakatochi, T. Shimamura, T. Okada, Y. Uno, et al., Comparative Analyses of Copy—Number Variation in Autism Spectrum Disorder and Schizophrenia Reveal Etiological Overlap and Biological Insights, Cell Rep. 24 (2018) 2838-2856, https://doi.org/10.1016/j.celrep.2018.08.022

[4] D. Ganesamoorthy, D.L. Bruno, G. McGillivray, F. Norris, S.M. White, S. Adroub, et al., Meeting the challenge of interpreting high—resolution single nucleotide polymorphism array data in prenatal diagnosis: does increased diagnostic power outweigh the dilemma of rare variants? BJOG 120 (2013) 594-606, https://doi.org/10.1111/1471—0528.12150

[5] X. Dong, B. Liu, L. Yang, H. Wang, B. Wu, R. Liu, et al., Clinical exome sequencing as the first—tier test for diagnosing developmental disorders covering both CNV and SNV: a Chinese cohort, J. Med Genet 57 (2020) 558-566, https://doi.org/10.1136/jmedgenet—2019—106377

[6] E.R. Riggs, E.F. Andersen, A.M. Cherry, S. Kantarci, H. Kearney, A. Patel, et al., Technical standards for the interpretation and reporting of constitutional copy number variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics (ACMG) and the Clinical Genome Resource (ClinGen), Genet Med 22 (2020) 245-257, https://doi.org/10.1038/s41436—019—0686—8

[7] M. Costa, A. García S, A. León, O. Pastor, The promises and pitfalls of automated variant interpretation: a comprehensive review, Brief. Bioinforma. 26 (2025) bbaf545, https://doi.org/10.1093/bib/bbaf545

[8] F. He, B. Lin, K. Mou, L. Jin, J. Liu, A machine learning model for the prediction of down syndrome in second trimester antenatal screening, Clin. Chim. Acta 521 (2021) 206-211, https://doi.org/10.1016/j.cca.2021.07.015

[9] G. Vandeweyer, E. Reyniers, W. Wuyts, L. Rooms, R.F. Kooy, CNV—WebStore: Online CNV Analysis, Storage and Interpretation, BMC Bioinforma. 12 (2011) 4, https://doi.org/10.1186/1471—2105—12—4

[10] M. Macnee, E. Pérez—Palma, T. Brünger, C. Klöckner, K. Platzer, A. Stefanski, et al., CNV—ClinViewer: enhancing the clinical interpretation of large copy—number variants online, Bioinformatics 39 (2023) btad290, https://doi.org/10.1093/bioinformatics/btad290

[11] G.A. Erikson, N. Deshpande, B.G. Kesavan, A. Torkamani, SG—ADVISER CNV: copy—number variant annotation and interpretation, Genet Med 17 (2015) 714-718, https://doi.org/10.1038/gim.2014.180

[12] J. Fino, B. Marques, Z. Dong, D. David, SVInterpreter: A Comprehensive Topologically Associated Domain—Based Clinical Outcome Prediction Tool for Balanced and Unbalanced Structural Variants, Front Genet 12 (2021) 757170, https://doi.org/10.3389/fgene.2021.757170

[13] S. Chen, C. Liu, X. Luan, Y. Wang, Y. Xu, Y. Li, et al., Application and clinical utility assessment of natural language processing—based software for copy—number variants interpretation, J. Transl. Med 23 (2025) 1052, https://doi.org/10.1186/s12967—025—07063—4

[14] M. Liu, Y. Zhong, H. Liu, D. Liang, E. Liu, Y. Zhang, et al., REDBot: Natural language process methods for clinical copy number variation reporting in prenatal and products of conception diagnosis, Molec Gen. & Gen. Med 8 (2020) e1488, https://doi.org/10.1002/mgg3.1488

[15] F. Requena, H.H. Abdallah, A. García, P. Nitschké, S. Romana, V. Malan, et al., CNVxplorer: a web tool to assist clinical interpretation of CNVs in rare disease patients, Nucleic Acids Res. 49 (2021) W93-W103, https://doi.org/10.1093/nar/gkab347

[16] C. Fan, Z. Wang, Y. Sun, J. Sun, X. Liu, L. Kang, et al., AutoCNV: a semiautomatic CNV interpretation system based on the 2019 ACMG/ClinGen Technical Standards for CNVs, Bmc Genom. 22 (2021) 721, https://doi.org/10.1186/s12864—021—08011—4

[17] M. Gažiová, T. Sládeček, O. Pös, M. Števko, W. Krampl, Z. Pös, et al., Automated prediction of the clinical impact of structural copy number variations, Sci. Rep. 12 (2022) 555, https://doi.org/10.1038/s41598—021—04505—z

[18] Q. Ding, C. Somerville, R. Manshaei, B. Trost, M.S. Reuter, K. Kalbfleisch, et al., SCIP: software for efficient clinical interpretation of copy number variants detected by whole—genome sequencing, Hum. Genet 142 (2023) 201-216, https://doi.org/10.1007/s00439—022—02494—1

[19] Sharo A.G., Hu Z., Sunyaev S.R., Brenner S.E. StrVCTVRE: A supervised learning method to predict the pathogenicity of human genome structural variants n.d.

[20] B. Poszewiecka, V.M. Pienkowski, K. Nowosad, J.D. Robin, K. Gogolewski, A. Gambin, TADeus2: a web server facilitating the clinical diagnosis by pathogenicity assessment of structural variations disarranging 3D chromatin structure, Nucleic Acids Res 50 (2022) W744-W752, https://doi.org/10.1093/nar/gkac318

[21] R.J. Schuetz, D. Ceyhan, A.A. Antoniou, B.P. Chaudhari, P. White, CNVoyant a machine learning framework for accurate and explainable copy number variant classification, Sci. Rep. 14 (2024) 22411, https://doi.org/10.1038/s41598—024—72470—4

[22] C.C.Y. Au—Yeung, Y.—T. Cheung, J.Y.T. Cheng, K.W.H. Ip, S.—D. Lee, V.Y.T. Yang, et al., UniVar: A variant interpretation platform enhancing rare disease diagnosis through robust filtering and unified analysis of SNV, INDEL, CNV and SV, Comput. Biol. Med 185 (2025) 109560, https://doi.org/10.1016/j.compbiomed.2024.109560

[23] V. Geoffroy, Y. Herenger, A. Kress, C. Stoetzel, A. Piton, H. Dollfus, et al., AnnotSV: an integrated tool for structural variations annotation, Bioinformatics 34 (2018) 3572-3574, https://doi.org/10.1093/bioinformatics/bty304

[24] K. Lv, D. Chen, D. Xiong, H. Tang, T. Ou, L. Kan, et al., dbCNV: deleteriousness—based model to predict pathogenicity of copy number variations, BMC Genom. 24 (2023) 131, https://doi.org/10.1186/s12864—023—09225—4

[25] T. Sládeček, M. Gažiová, M. Kucharík, A. Zaťková, Z. Pös, O. Pös, et al., Combination of expert guidelines—based and machine learning—based approaches leads to superior accuracy of automated prediction of clinical effect of copy number variations, Sci. Rep. 13 (2023) 10531, https://doi.org/10.1038/s41598—023—37352—1

[26] T.A. Gurbich, V.V. Ilinsky, ClassifyCNV: a tool for clinical annotation of copy—number variants, Sci. Rep. 10 (2020) 20375, https://doi.org/10.1038/s41598—020—76425—3

[27] L. Zhang, J. Shi, J. Ouyang, R. Zhang, Y. Tao, D. Yuan, et al., X—CNV: genome—wide prediction of the pathogenicity of copy number variations, Genome Med 13 (2021) 132, https://doi.org/10.1186/s13073—021—00945—4

[28] de Godoy V.C.S.M., Bellucco F.T., Colovati M., de Oliveira—Junior H.R., Moysés—Oliveira M., Melaragno M.I. . Copy number variation (CNV) identification, interpretation, and database from Brazilian patients. Genet Mol Biol n.d.; 43: e20190218. https://doi.org/10.1590/1678—4685—GMB—2019—0218.

[29] G.V. Drozdov, A.A. Kashevarova, I.N. Lebedev, Copy number variations in spontaneous abortions: a meta—analysis, J. Assist Reprod. Genet 42 (2025) 1039-1052, https://doi.org/10.1007/s10815—025—03420—w

[30] P. Shi, H. Liang, Y. Hou, D. Chen, H. Ren, C. Wang, et al., The uncertainty of copy number variants: pregnancy decisions and clinical follow—up, Am. J. Obstet. Gynecol. 229 (2023) 170.e1-170.e8, https://doi.org/10.1016/j.ajog.2023.01.022

[31] J. Wang, L. Chen, C. Zhou, L. Wang, H. Xie, Y. Xiao, et al., Prospective chromosome analysis of 3429 amniocentesis samples in China using copy number variation sequencing, Am. J. Obstet. Gynecol. 219 (2018) 287.e1-287.e18, https://doi.org/10.1016/j.ajog.2018.05.030

[32] D. Zou, L. Wang, J. Liao, H. Xiao, J. Duan, T. Zhang, et al., Genome sequencing of 320 Chinese children with epilepsy: a clinical and molecular study, Brain 144 (2021) 3623-3634, https://doi.org/10.1093/brain/awab233

[33] R.L. Collins, J.T. Glessner, E. Porcu, M. Lepamets, R. Brandon, C. Lauricella, et al., A cross—disorder dosage sensitivity map of the human genome, Cell 185 (2022) 3041-3055.e25, https://doi.org/10.1016/j.cell.2022.06.036

[34] N. Bastida—Lertxundi, E. López—López, M.A. Piñán, A. Puiggros, A. Navajas, F. Solé, et al., Errors in the interpretation of copy number variations due to the use of public databases as a reference, Cancer Genet 207 (2014) 164-167, https://doi.org/10.1016/j.cancergen.2014.03.001

[35] J.—M. Ravel, M. Renaud, J. Muller, A. Becker, É. Renard, T. Remen, et al., Clinical utility of periodic reinterpretation of CNVs of uncertain significance: an 8—year retrospective study, Genome Med 15 (2023) 39, https://doi.org/10.1186/s13073—023—01191—6

[36] P. Tan, D. Li, L. Chang, J. Shi, Y. Han, R. Zhang, et al., Evaluation of noninvasive prenatal screening for copy number variations among screening laboratories, Clin. Biochem 118 (2023) 110617, https://doi.org/10.1016/j.clinbiochem.2023.110617