Beyond in silico prediction: multi-omics to identify a pathogenic deep intronic HNRNPK variant in Au-Kline syndromeDownload PDF Download PDF Brief CommunicationOpen accessPublished: 04 September 2026Z. Kowalzyk1,J. Porrmann1,P.Y.B. Au2,S. Choufani3,R. Weksberg ORCID: orcid.org/0000-0002-6501-41503,4,K. Hackmann1,A. Kübler1,A. Kögler5,N. S. Lewis1,A. Köhler1,J. Wagner1,M. Bermudez1,C. Hübner6,J. Schallner7,R. Jauss8,C. Vogelberg9,B. Mayer10,N. Di Donato1,E. Schröck1,D. Le Duc1 &…A. Jahn1 Journal of Human Genetics (2026) Cite this articleSave articleView saved researchAbstractPathogenic variants in HNRNPK are associated with autosomal dominant Au-Kline syndrome (AKS, Au-Kline-Okamoto syndrome, OMIM #616580). This syndrome is characterized by developmental delay and intellectual disability, hypotonia, and distinctive facial features. Despite the use of whole-genome sequencing (WGS) as a powerful diagnostic tool, we nearly dismissed a novel intronic variant (NM_031263.4(HNRNPK):c.214-55 T > A) affecting HNRNPK splicing and function. Although commonly used bioinformatic splice prediction tools, including SpliceAI and PDIVAS, yielded inconclusive results, Face2Gene analysis indicated a high phenotypic similarity to AKS. Characteristic facial features described by Choufani et al. [1] supported the clinical diagnosis of AKS. Subsequent functional studies demonstrated aberrant splicing with intron retention, and DNA methylation profiling revealed a positive HNRNPK-specific episignature. These insights and the de novo status support an evaluation as likely pathogenic. This case report supports the relevance of facial analysis and comprehensive variant validation strategies, particularly for deep intronic variants with ambiguous in silico splicing predictions.IntroductionHeterogeneous nuclear ribonucleoprotein K (hnRNPK) is an RNA- and DNA-binding protein that plays a central role in gene regulation, cellular signaling, and chromatin remodeling. Its three K-homologous (KH) domains are highly conserved across other RNA-binding proteins and are essential for both transcriptional regulation and mRNA splicing. Phosphorylation of hnRNPK regulates its function and can indirectly impact gene expression [2]. Heterozygous pathogenic variants in HNRNPK cause autosomal dominant Au-Kline Syndrome (AKS, Au-Kline-Okamoto syndrome), a neurodevelopmental disorder characterized by developmental delay and intellectual disability, hypotonia, distinctive facial features, and variable autonomic dysfunction. Multiple congenital anomalies like hydronephrosis, palate abnormalities, congenital heart disease, and oligodontia were described (OMIM #616580). The disease mechanism is suspected to be loss-of-function of HNRNPK [1].Clinical reportHere, we report a female proband who was referred shortly after birth due to omphalocele, valvular aortic stenosis, and reflux nephropathy. At the latest follow-up at age eight years, she presented with developmental delay (IQ: 51 KI-95% = 48–64), hypotonia, short stature (–2.85 SD), and a suspicion of an interstitial lung disease (ILD) (on imaging: non-progressive diffuse changes in lung structure with ground-glass opacities and increased pulmonary hyperinflation), with recurrent respiratory infections and oxygen requirement. The renal and pulmonary symptoms improved over time. Facial features included long palpebral fissures, almond-shaped eyes, shallow orbits, a broad nose with a wide nasal bridge and thick alae nasi, and a downturned mouth. After extensive diagnostic investigations without a definitive result, Face2Gene analysis suggested AKS with the highest gestalt match (Fig. 1A). The proband achieved a clinical severity score of 7/14 according to Choufani et al. [1], which falls within the range reported for individuals with a positive HNRNPK DNAm episignature (3 − 11/14).Fig. 1Full size imageHNRNPK Overview and Results. A Face2Gene Composite photo of Au-Kline Syndrome. B Gene model with (likely) pathogenic variants in HNRNPK (NM_031263.4, hg38) from ClinVar [15] and Choufani et al. [1]. Different variant types and in silico splice prediction scores are indicated (red square = reported variant NM_031263.4(HNRNPK):c.214-55 T > A). C Sashimi plot (shows exons 3-8 from 15 in total, hg38) of RNA-seq data from patients' derived fibroblasts, samples untreated and treated with Cycloheximide (CHX) for nonsense-mediated mRNA decay (NMD) inhibition. The untreated sample suggests an aberrant splicing effect in intron 6. NMD inhibition proved an activation of a cryptic splice site 55 bp upstream of exon 6 within intron 5 (red arrow), resulting in exon 6 elongation. D HNRNPK-specific DNA methylation (DNAm) episignature performed by Choufani et al. [1] on the patient’s DNA (sample HNRNPK_EX1673, red arrow) as well as positive and negative controlsSubsequent re-analysis of sequencing data with segregation analysis identified a heterozygous intronic de novo variant in HNRNPK (NM_031263.4:c.214-55 T > A), which had been missed by commonly used bioinformatic splice prediction tools, including SpliceAI [3] and PDIVAS [4].Methods and resultsWe obtained written informed consent for genetic analysis and publication of this case report from the patient’s parents. Consent for publication did not include facial photographs. Cytogenetic analysis using G-banding revealed a normal 46,XX karyotype. Microarray-based comparative genomic hybridization (aCGH, Human CHG Microarray Kit 2x400k, Agilent), methylation analysis by multiplex ligation-dependent probe amplification (MLPA), and targeted testing for Beckwith-Wiedemann syndrome (ME030-C3, MCR-Holland) and Kagami-Ogata syndrome (ME032-A1, MCR-Holland) showed no pathogenic copy number variations or abnormal methylation patterns.Face2Gene suggested AKS with high confidence [5]. Trio whole-exome sequencing (WES, Illumina-IDT Exome Enrichment Kit, NextSeq 500, Illumina) was performed on genomic DNA from the patient and both parents. Validated variant calling included exonic regions and flanking intronic sequences ( ± 50 bp), followed by prioritization using Human Phenotype Ontology (HPO) terms. ILD panel evaluation (ABCA3, CSF2RB, FLNA, FOXF1, NKX2-1, SFTPA1, SFTPA2, SFTPB, SFTPC, and TBX4) and an open-exome approach revealed no candidate variants consistent with the phenotype.Subsequently, whole-genome sequencing (WGS, Illumina PCR-free DNA library preparation Kit, NovaSeq 6000, Illumina) identified a heterozygous deep intronic variant in HNRNPK (NM_031263.4:c.214-55 T > A, our entry ClinVar Variation ID:2576028, Fig. 1B). However, bioinformatic splice prediction tools showed low and non-informative scores (SpliceAI AG Δ =0.1 at –2 bp, Pangolin AG Δ =0.09 at –98 bp, PDIVAS = 0.026; Supplementary Table 1) [6], while other in-silico prediction tools showed increased scores (CADD PHRED = 17.61). Segregation analysis confirmed the variant to be de novo, consistent with the known molecular mechanism underlying AKS.To further investigate a potential disease-causing transcriptional effect, skin-derived fibroblasts from the patient were analysed by short-read RNA sequencing (Illumina TruSeq Stranded mRNA Prep Kit with poly-A enrichment; NextSeq 500, Illumina). RNA-seq revealed low-level aberrant splicing (percent spliced in, PSI = 0.75%), creating a new splice acceptor site 55 bp upstream of exon 6 within intron 5 (Fig. 1C, untreated CHX). Based on these findings, nonsense-mediated mRNA decay (NMD) inhibition using cycloheximide (CHX; Sigma-Aldrich) confirmed the enrichment of an aberrant transcript (PSI = 24.9%, Fig. 1C, treated CHX [7]). These in vitro results suggest NMD of the aberrant transcript leading to a premature termination codon (p.(Tyr72Ilefs*35)) and possibly leaky splicing. However, incomplete NMD inhibition or nuclear retention of the aberrant transcript cannot be ruled out.Additionally, genome-wide DNAm analysis (Illumina human Infinium MethylationEPIC BeadChip) was performed on patient blood-derived DNA. A comparative analysis including a reference pathogenic HNRNPK-cohort (n = 26) and unaffected controls (n = 46) classified the variant as pathogenic based on HNRNPK-specific DNAm profile (Fig. 1D[1]).DiscussionInitial standard clinical genetic testing, including karyotyping, trio WES, and methylation analysis (ICR1/2 and MEG3), was inconclusive, but given suggestive Face2Gene analysis, WGS identified a deep intronic de novo variant in HNRNPK (NM_031263.4:c.214-55 T > A). Additional RNA analysis and DNAm profiling supported the clinical diagnosis of AKS [1]. The variant was classified as “likely pathogenic” according to ACMG/AMP criteria [8] and ClinGen specifications [9] based on the following criteria: PVS1_moderate (RNA), PS2 (de novo), PM2_supporting (absent in GnomAD v.4.1), and PP4 (DNAm). The suspected interstitial lung disease might extend the phenotypic spectrum of AKS.Several other non-canonical and intronic splice variants in HNRNPK have been reported in the literature, but bioinformatic splice prediction scores may be inconclusive (Fig. 1B and Supplementary Table 1). Individuals carrying such variants exhibited intermediate or positive AKS DNAm episignatures [1], which may be explained by incomplete reduction of functional hnRNPK from this allele due to leaky splicing. While other pathogenic variants have been reported in intron 6 (Fig. 1), different from our variant, an intron retention was observed for c.214-35 A > G [10], which does not support a common mechanism. Variant pathogenicity can be supported through trio analysis and HNRNPK-specific DNAm episignatures, emphasizing the value of epigenetic approaches for the diagnosis of rare disorders such as AKS [1, 11].Here, the distinctive facial features were crucial for prioritizing AKS. However, in patients with non-specific developmental delay and complex phenotypes, this phenotype-driven approach may not be feasible. Deep intronic variants may be missed by bioinformatic pipelines and yield inconclusive in silico predictions, which underscores the need for further refinement and careful interpretation. In this context, trio-analysis, as well as multi-omics approaches, including transcriptomic and epigenetic analyses, can improve the diagnostic yield in individuals with suspected genetic disease [12,13,14]. However, their implementation in routine care remains challenging, as they require further evaluation.ReferencesChoufani S, McNiven V, Cytrynbaum C, Jangjoo M, Adam MP, Bjornsson HT. An HNRNPK-specific DNA methylation signature makes sense of missense variants and expands the phenotypic spectrum of Au-Kline syndrome. Am J Hum Genet. 2022;109:1867–84. https://doi.org/10.1016/j.ajhg.2022.08.014.Article CAS PubMed PubMed Central Google Scholar Wang Z, Qiu H, He J, Liu L, Xue W, Fox A. The emerging roles of hnRNPK. J Cell Physiol. 2020;235:1995–2008. https://doi.org/10.1002/jcp.29186.Article CAS PubMed Google Scholar Jaganathan K, Kyriazopoulou Panagiotopoulou S, McRae JF, Darbandi SF, Knowles D, Li YI. Predicting splicing from primary sequence with deep learning. Cell. 2019;176:535–48.e24. https://doi.org/10.1016/j.cell.2018.12.015.Article CAS PubMed Google Scholar Kurosawa R, Iida K, Ajiro M, Awaya T, Yamada M, Kosaki K. PDIVAS: pathogenicity predictor for Deep-Intronic Variants causing aberrant splicing. BMC Genomics. 2023;24:601. https://doi.org/10.1186/s12864-023-09645-2.Article CAS PubMed PubMed Central Google Scholar Hsieh TC, Bar-Haim A, Moosa S, Ehmke N, Gripp KW, Pantel JT. GestaltMatcher facilitates rare disease matching using facial phenotype descriptors. Nat Genet. 2022;54:349–57. https://doi.org/10.1038/s41588-021-01010-x.Article CAS PubMed PubMed Central Google Scholar Walker LC, Hoya Mdela, Wiggins GAR, Lindy A, Vincent LM, Parsons MT. Using the ACMG/AMP framework to capture evidence related to predicted and observed impact on splicing: Recommendations from the ClinGen SVI Splicing Subgroup. Am J Hum Genet. 2023;110:1046–67. https://doi.org/10.1016/j.ajhg.2023.06.002.Article CAS PubMed PubMed Central Google Scholar Garrido-Martín D, Palumbo E, Guigó R, Breschi A. ggsashimi: Sashimi plot revised for browser- and annotation-independent splicing visualization. PLoS Comput Biol. 2018;14:e1006360. https://doi.org/10.1371/journal.pcbi.1006360.Article CAS PubMed PubMed Central Google Scholar Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med Off J Am Coll Med Genet. 2015;17:405–24. https://doi.org/10.1038/gim.2015.30.Article Google Scholar Rehm HL, Berg JS, Brooks LD, Bustamante CD, Evans JP, Landrum MJ. ClinGen—the clinical genome resource. N Engl J Med. 2015;372:2235–42. https://doi.org/10.1056/NEJMsr1406261.Article CAS PubMed PubMed Central Google Scholar Murdock DR, Dai H, Burrage LC, Rosenfeld JA, Ketkar S, Müller MF. Transcriptome-directed analysis for Mendelian disease diagnosis overcomes limitations of conventional genomic testing. J Clin Invest. 2021;131:e141500. https://doi.org/10.1172/JCI141500.Article CAS PubMed PubMed Central Google Scholar Mingoia M, Meloni A, Sedda S, Choufani S, Asunis I, Gemma G. A novel intronic variant in the KH3 domain of HNRNPK leads to a mild form of Au-Kline syndrome. Clin Genet. 2025;108:576–81. https://doi.org/10.1111/cge.14763.Article CAS PubMed PubMed Central Google Scholar Kernohan KD, Boycott KM. The expanding diagnostic toolbox for rare genetic diseases. Nat Rev Genet. 2024;25:401–15. https://doi.org/10.1038/s41576-023-00683-w.Article CAS PubMed Google Scholar Zhao S, Macakova K, Sinson JC, Dai H, Rosenfeld J, Zapata GE. Clinical validation of RNA sequencing for Mendelian disorder diagnostics. Am J Hum Genet. 2025;112:779–92. https://doi.org/10.1016/j.ajhg.2025.02.006.Article CAS PubMed PubMed Central Google Scholar Bournazos AM, Riley LG, Bommireddipalli S, Ades L, Akesson LS, Al-Shinnag M. Standardized practices for RNA diagnostics using clinically accessible specimens reclassifies 75% of putative splicing variants. Genet Med Off J Am Coll Med Genet. 2022;24:130–45. https://doi.org/10.1016/j.gim.2021.09.001.Article CAS Google Scholar Landrum MJ, Lee JM, Riley GR, Jang W, Rubinstein WS, Church DM. ClinVar: public archive of relationships among sequence variation and human phenotype. Nucleic Acids Res. 2014;42:D980–985. https://doi.org/10.1093/nar/gkt1113.Article CAS PubMed Google Scholar Download referencesAcknowledgementsThe authors would like to thank the family for their generous participation in this study and the healthcare providers involved in their care.FundingOpen Access funding enabled and organized by Projekt DEAL.Author informationAuthors and AffiliationsInstitute for Clinical Genetics, University Hospital Carl Gustav Carus at TUD Dresden University of Technology and Faculty of Medicine of TUD Dresden University of Technology, Dresden, GermanyZ. Kowalzyk, J. Porrmann, K. Hackmann, A. Kübler, N. S. Lewis, A. Köhler, J. Wagner, M. Bermudez, N. Di Donato, E. Schröck, D. Le Duc & A. JahnDepartment of Medical Genetics, Alberta Children’s Hospital Research Institute, Cumming School of Medicine, University of Calgary, Calgary, AB, CanadaP.Y.B. AuGenetics and Genome Biology Program, Research Institute, The Hospital for Sick Children, Toronto, ON, CanadaS. Choufani & R. WeksbergDivision of Clinical and Metabolic Genetics, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, Toronto, ON, CanadaR. WeksbergCore Unit for Molecular Tumor Diagnostics (CMTD), National Center for Tumor Diseases (NCT), NCT/UCC Dresden, a partnership between German Cancer Research Center (DKFZ), Faculty of Medicine and University Hospital Carl Gustav Carus, TUD Dresden University of Technology and Helmholtz-Zentrum Dresden-Rossendorf (HZDR), Dresden, GermanyA. KöglerDepartment of Pediatrics, Faculty of Medicine of TUD Dresden University of Technology and University Hospital Carl Gustav Carus at TUD Dresden University of Technology, Dresden, GermanyC. HübnerDepartment of Neuropediatrics & iSPZ UKD, University Hospital Carl Gustav Carus at TUD Dresden University of Technology, Dresden, GermanyJ. SchallnerInstitute of Human Genetics, University of Leipzig Medical Center, Leipzig, GermanyR. JaussDepartment of Pediatric Pulmonology and Allergy, University Hospital Carl Gustav Carus, Technical University of Dresden, Dresden, GermanyC. VogelbergDepartment of Pediatric Nephrology, University Hospital Carl Gustav Carus, Technical University of Dresden, Dresden, GermanyB. MayerAuthorsZ. KowalzykView author publicationsSearch author on:PubMed Google ScholarJ. PorrmannView author publicationsSearch author on:PubMed Google ScholarP.Y.B. AuView author publicationsSearch author on:PubMed Google ScholarS. ChoufaniView author publicationsSearch author on:PubMed Google ScholarR. WeksbergView author publicationsSearch author on:PubMed Google ScholarK. HackmannView author publicationsSearch author on:PubMed Google ScholarA. KüblerView author publicationsSearch author on:PubMed Google ScholarA. KöglerView author publicationsSearch author on:PubMed Google ScholarN. S. LewisView author publicationsSearch author on:PubMed Google ScholarA. KöhlerView author publicationsSearch author on:PubMed Google ScholarJ. WagnerView author publicationsSearch author on:PubMed Google ScholarM. BermudezView author publicationsSearch author on:PubMed Google ScholarC. HübnerView author publicationsSearch author on:PubMed Google ScholarJ. SchallnerView author publicationsSearch author on:PubMed Google ScholarR. JaussView author publicationsSearch author on:PubMed Google ScholarC. VogelbergView author publicationsSearch author on:PubMed Google ScholarB. MayerView author publicationsSearch author on:PubMed Google ScholarN. Di DonatoView author publicationsSearch author on:PubMed Google ScholarE. SchröckView author publicationsSearch author on:PubMed Google ScholarD. Le DucView author publicationsSearch author on:PubMed Google ScholarA. JahnView author publicationsSearch author on:PubMed Google ScholarCorresponding authorCorrespondence to Z. Kowalzyk.Ethics declarationsCompeting interestsThe authors declare no competing interests.Ethics approvalWe obtained written informed consent for genetic analysis and publication of this case report from the patient’s parents. This study was conducted in accordance with the ethical standards of the Ethics Committee of the Technical University of Dresden, Germany (approval no. EK 273072018), and the principles of the Declaration of Helsinki.Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary informationRights and permissionsThis article is published under an open access license. Please check the 'Copyright Information' section either on this page or in the PDF for details of this license and what re-use is permitted. If your intended use exceeds what is permitted by the license or if you are unable to locate the licence and re-use information, please contact the Rights and Permissions team.About this article