Chapter 10: Whole-Genome Clinical Diagnostics: Deep Intronic & Regulatory Variants
Chapter 10: Whole-Genome Clinical Diagnostics: Deep Intronic & Regulatory Variants
While Whole-Exome Sequencing (WES) queries only the 1.5% of the genome encoding protein exons, clinical Whole-Genome Sequencing (cWGS) captures the entire 3.2 billion base pair genome. cWGS discovers pathogenic variations buried deep within non-coding introns, promoter regions, and distant enhancers that escape exome capture.
Deep intronic point mutations can activate cryptic splice donor or acceptor sites, causing pseudo-exon inclusion that disrupts open reading frames. Illumina's SpliceAI (a 32-layer residual neural network) evaluates 10,000 bp sequence contexts to predict splicing alterations with exceptional accuracy.
1. Annotating Variants with SpliceAI
# Run SpliceAI on clinical VCF
spliceai \\
-I candidate_variants.vcf \\
-O spliceai_annotated.vcf \\
-R ref_grch38.fa \\
-A grch38_canonical.gtf \\
-D 4999 # Scan up to 5,000 bp flanking distance
# SpliceAI output delta scores (0.0 to 1.0):
# DS_AG (Acceptor Gain), DS_AL (Acceptor Loss), DS_DG (Donor Gain), DS_DL (Donor Loss)
# Delta score > 0.5: High probability of pathogenic aberrant splicing2. Pathogenic Non-Coding Promoters: The TERT Paradigm
In cancer, the two most common non-coding mutations are TERT promoter C228T (\(-124 \text{ C>T}\)) and C250T (\(-146 \text{ C>T}\)). These single-base mutations create de novo 11 bp binding motifs (CCGGAA) for ETS transcription factors (GABP), continuously reactivating telomerase expression to confer cellular immortality.
Common Pitfalls & Troubleshooting Checklist
- Java Heap Exhaustion (GATK / Picard): If processes terminate with
java.lang.OutOfMemoryError, set your environment variable explicitly:export _JAVA_OPTIONS="-Xmx16g -Djava.io.tmpdir=/tmp". - Index Timestamp Mismatch: If tools complain that a BAM index is missing or out of date, regenerate with
samtools index -@ 4 sample.bam. - Conda Channel Conflicts: Avoid ABI incompatibilities between bioconda and conda-forge by setting strict priority:
conda config --set channel_priority strict. - Chromosome Naming Contamination: Always verify whether reference FASTA headers use
chr1or1; divergent naming causes silent alignment drops.