Introduction
The RDDC RNA Splicing Prediction Model (RNA Splicer) AI tool plays a critical role in resolving the genetic basis of complex disorders like 46,XY Disorders of Sex Development (DSD), which often present with similar clinical phenotypes. In a study involving 21 patients with 46,XY DSD without a uterus, the tool successfully predicted the functional consequences of a novel variant at a canonical splice site in the *LHCGR* gene. RDDC's prediction (exon skipping/intron retention) was subsequently precisely validated by a minigene assay. This not only clarified the variant's pathogenicity (classifying it as LP) but also highlighted the significant value of the "WES + RDDC prediction + functional validation" paradigm in enhancing diagnostic accuracy for rare diseases and guiding genetic counseling and PGT.
Research Challenge: WES Identifies a Splice Site VUS
46,XY DSD patients without a uterus share overlapping clinical phenotypes (e.g., primary amenorrhea, external genitalia abnormalities), making precise diagnosis difficult. This study utilized Whole Exome Sequencing (WES) to analyze the molecular etiology in 21 such patients, aiming to discover novel variants. In this process, the research team identified a novel intronic variant in the **LHCGR gene** in one patient: `c.680+1G>T`. This variant, located at the canonical splice donor site of intron 8, was absent from public databases and classified as a "Variant of Uncertain Significance" (VUS). Determining its specific impact on mRNA splicing was essential to confirm if it was the cause of the patient's phenotype.
RDDC's Precise Prediction: Unveiling Loss-of-Function Splicing Aberrations
To assess the potential pathogenic mechanism of the `c.680+1G>T` VUS, the researchers employed the RDDC RNA Splicing Prediction Model bioinformatics AI tool. RDDC's prediction provided a clear and specific molecular hypothesis, indicating that the variant would disrupt the splice donor site and likely activate cryptic splice sites. This would lead to two primary aberrant splicing consequences:
- Intron retention
- Exon skipping
Both predicted outcomes strongly suggested that the variant would severely disrupt the normal processing of mRNA, leading to a truncated or non-functional LHCGR protein, thereby causing the disease.
Experimental Validation and Clinical Value: From Prediction to Precise Diagnosis
The research team promptly validated RDDC's predictions using an *in vitro* minigene assay. The experimental results robustly confirmed RDDC's forecast: the vector carrying the `c.680+1G>T` mutation indeed produced aberrant transcripts when transfected into cells, consistent with the splicing anomaly patterns predicted by RDDC. The perfect concordance between the *in silico* prediction and the *in vitro* validation provided the critical pathogenic evidence (fulfilling ACMG PVS1+PS3 criteria) needed to upgrade the VUS to "Likely Pathogenic" (LP).
This case powerfully demonstrates that RDDC RNA Splicer is a robust tool for interpreting the pathogenicity of splice site VUS. It provides rapid, reliable functional predictions for novel, rare variants discovered by WES, effectively assisting clinical pathogenicity assessment. This "WES + RDDC prediction + functional validation" pipeline not only successfully aided the diagnosis of the *LHCGR* variant but also provided a methodological reference for analyzing other genes in the study (like *NR5A1*), significantly improving the diagnostic accuracy for DSD and providing a solid molecular basis for genetic counseling and reproductive interventions like PGT.
Content Source and Disclaimer
This article is a compilation and interpretation of the scientific study cited below, intended to highlight the application of RDDC bioinformatics tools. All research data and conclusions belong to the original authors and publication.
Original Article:
Ding L, Luo M, Deng S, et al. The clinical diversity and molecular etiology in 46, XY disorders of sex development patients without uterus. Orphanet Journal of Rare Diseases. 2025.
Article Link:
https://pmc.ncbi.nlm.nih.gov/articles/PMC12007265/
