Residency · Residency · Medical Genetics Genomics
RNA Sequencing as a Diagnostic Tool
Introduction
RNA sequencing (RNA-seq) is emerging as a powerful complementary diagnostic tool in clinical genetics, particularly for cases unsolved by DNA-based testing. By directly interrogating the transcriptome, RNA-seq can detect aberrant splicing, allele-specific expression, and expression outliers, as well as the functional consequences of variants of uncertain significance, thereby increasing diagnostic yield by an estimated 7-17% beyond exome or genome sequencing alone.
Biological Rationale
Why RNA Matters in Diagnostics
Many disease-causing variants exert their effects at the RNA level but are invisible or uninterpretable at the DNA level. Deep intronic variants creating cryptic splice sites are missed by exome sequencing and often classified as VUS by genome sequencing. Synonymous variants and missense variants near splice junctions may disrupt splicing without obvious DNA-level predictions. Regulatory variants affecting gene expression (in promoter, enhancer, or UTR regions) are detectable through expression changes. Nonsense-mediated decay (NMD) reduces transcript levels from loss-of-function alleles, creating detectable monoallelic expression patterns.
Technical Approach
Sample Types and Tissue Considerations
Whole blood collected in PAXgene tubes is the most accessible sample type and is suitable for approximately 60% of known disease genes. Fibroblasts from skin biopsy express a broader range of genes and represent the gold standard for musculoskeletal and connective tissue disorders. Skeletal muscle biopsy is essential for myopathies and mitochondrial disorders, expressing genes not active in blood or fibroblasts. Tissue-specific expression represents a major limitation, as some disease-relevant genes are only expressed in brain, liver, or other inaccessible tissues. Induced pluripotent stem cells (iPSCs) can theoretically provide any cell type but are impractical for routine diagnostics.
Library Preparation
Poly-A selection captures mature mRNA and is standard for clinical applications. Ribosomal RNA depletion captures a broader RNA population including non-polyadenylated transcripts and pre-mRNA, which is useful for detecting intron retention. Adequate depth typically requires 50-100 million paired-end reads per sample. Strand-specific library preparation is preferred for accurate transcript quantification.
Analytical Methods
Aberrant Splicing Detection
Split-read analysis identifies novel splice junctions not present in reference annotations. Intron retention detects failure to remove intronic sequences. Exon skipping identifies loss of exons from mature transcripts. Cryptic exon inclusion reveals deep intronic variants activating pseudo-exons. Alternative splice site usage identifies shifted donor or acceptor sites. Tools used include FRASER, LeafCutter, rMATS, and SpliceAI (DNA-level prediction validated by RNA-seq).
Expression Outlier Analysis
This approach compares a patient's gene expression against a reference cohort of similarly processed samples, identifying genes with significantly reduced or elevated expression (typically more than 2 standard deviations from the cohort mean). Monoallelic expression (MAE) represents loss of expression from one allele, suggesting NMD of a loss-of-function allele or a regulatory variant. Tools include OUTRIDER, ANEVA-DOT, and the DROP pipeline.
Allele-Specific Expression
This method quantifies the relative expression of each allele using heterozygous SNPs as markers. Significant allelic imbalance suggests regulatory variants, NMD, or epigenetic silencing. It requires heterozygous coding variants in the gene of interest for informative assessment.
Diagnostic Yield and Clinical Evidence
Key Studies
Cummings et al. (2017) applied RNA-seq of muscle biopsies in 50 undiagnosed myopathy patients and achieved a 35% diagnostic rate with additional diagnoses beyond WES. Fresard et al. (2019) used blood RNA-seq in 94 undiagnosed rare disease patients and provided diagnoses or strong candidates in 7.5%. Yepez et al. (2022) applied the DROP (Detection of RNA Outliers Pipeline) to over 1,000 samples, demonstrating robust aberrant expression and splicing detection. Overall, RNA-seq adds 7-17% incremental diagnostic yield when applied after negative or inconclusive WES/WGS.
| RNA-seq Analytical Method | What It Detects | Key Tools | Clinical Example |
|---|---|---|---|
| Aberrant splicing | Exon skipping, cryptic exons, intron retention, alternative splice sites | FRASER, LeafCutter, rMATS | Deep intronic variant creating cryptic splice site in rare disease gene |
| Expression outlier | Significantly reduced or elevated gene expression vs. cohort | OUTRIDER, DROP pipeline | Monoallelic expression from NMD of loss-of-function allele |
| Allele-specific expression | Imbalanced expression between two alleles | ANEVA-DOT | Regulatory variant silencing one allele; NMD detection |
Types of Diagnoses Made
Diagnoses include cryptic splice variants in known disease genes confirmed by aberrant junction reads, deep intronic variants validated by demonstrating cryptic exon inclusion, VUS reclassification to likely pathogenic based on demonstrated functional impact, and expression outliers pointing to previously unsuspected deletions or regulatory variants.
Limitations and Challenges
Tissue-specific expression means many disease genes are not expressed in accessible tissues. Robust outlier detection requires a sufficiently large reference cohort processed identically. RNA degradation from pre-analytical variables significantly affects data quality, necessitating strict sample handling protocols. Some truncating variants escape NMD, and some NMD transcripts are partially detectable, limiting sensitivity. RNA-seq is not yet widely available as a clinical test, with most experience in research or specialized reference laboratories. Normal RNA-seq results do not exclude a functional impact occurring only in unsampled tissues.
Clinical Integration
When to Order RNA-seq
RNA-seq is indicated after negative or inconclusive WES/WGS with strong clinical suspicion of a genetic disorder, to evaluate a VUS near a splice site that cannot be resolved by in-silico tools alone, when a candidate deep intronic variant identified by WGS requires functional validation, in neuromuscular disorders (particularly high yield from muscle biopsy RNA-seq), and for suspected disorders of gene regulation where expression level is expected to be altered.
Reporting Framework
Reports should describe aberrant splicing events with reference to the DNA variant when identified, quantify expression outliers with statistical significance relative to the reference cohort, and integrate RNA-seq findings with DNA-level variant interpretation for ACMG classification (functional data as PS3/BS3 evidence).
Clinical Pearls
RNA-seq is most powerful as a complementary test after negative WES/WGS, not as a standalone diagnostic tool. Tissue selection is critical: blood RNA-seq will miss variants in genes not expressed in leukocytes, while muscle biopsy RNA-seq has the highest yield for myopathies. RNA-seq can reclassify VUS by providing functional evidence of aberrant splicing or expression, directly supporting ACMG PS3 criteria. Building and maintaining a well-matched control cohort is essential for reliable outlier detection.
References
- Cummings BB, Marshall JL, Tukiainen T, et al. Improving genetic diagnosis in Mendelian disease with transcriptome sequencing. Science Translational Medicine. 2017;9(386):eaal5209.
- Fresard L, Smail C, Ferraro NM, et al. Identification of rare-disease genes using blood transcriptome sequencing and large control cohorts. Nature Medicine. 2019;25(6):911-919.
- Yepez VA, Gusic M, Kopajtich R, et al. Clinical implementation of RNA sequencing for Mendelian disease diagnostics. Genome Medicine. 2022;14:38.
- Lee H, Huang AY, Wang LK, et al. Diagnostic utility of transcriptome sequencing for rare diseases. Genetics in Medicine. 2020;22(3):490-499.