Module 6: Read-Level Verification & Visualizing Somatic Hotspots in IGV
1. The Clinical Imperative of Manual Read Inspection
Automated variant callers are powerful, but no clinical oncology molecular report should be issued without manual visual inspection of raw aligned reads in the Integrative Genomics Viewer (IGV). Studies show that up to 15% of low-VAF candidate somatic calls generated by automated pipelines in degraded FFPE or liquid biopsy samples represent technical artifacts (such as strand bias, mapping ambiguities, or alignment errors near indels) that human visual inspection immediately detects.
2. Common IGV Red Flags: Distinguishing True Mutations from Artifacts
| Visual Pattern in IGV | Technical Mechanism | Classification | Clinical Action |
|---|---|---|---|
| Extreme Strand Bias | Mutant base call is present exclusively on forward reads (pink) and completely absent on reverse reads (blue), or vice versa. | Artifact: typically PCR deamination or optical sequencing error. | Reject call as false positive. |
| End-of-Read Clumping | Mutant base occurs solely within the first or last 5–8 base pairs of aligned reads. | Artifact: terminal read quality drop or adapter ligation artifact. | Reject call; true biological mutations occur uniformly distributed across read bodies. |
| Low Mapping Quality ($MQ < 20$) | Reads supporting the variant exhibit white translucent shading in IGV ($MQ=0$). | Artifact: multi-mapping reads aligning to pseudogenes or segmental duplications. | Reject call; verified driver mutations reside in uniquely mapped reads ($MQ=60$). |
| Homopolymer Run Trailing | Indel or base change occurs immediately adjacent to a run of $\ge 6$ identical nucleotides (e.g. AAAAAAA). | Artifact: polymerase slippage during thermal cycling. | Apply stringent homopolymer penalty filter. |
| Bilateral Strand Distribution & Balanced Insert Sizes | Variant is supported by both forward and reverse reads with varying start/end positions and high mapping qualities. | True Somatic Mutation: authentic biological biological event. | Confirm variant and proceed to clinical curation. |
3. Step-by-Step IGV Inspection: The Three Case Study Hotspots
- Case 1 (Lung): EGFR Exon 21 L858R (chr7:55,259,515 T>G)
Navigate to
chr7:55,259,515in GRCh38. Observe the missense substitution converting codonCTG(Leucine) toCGG(Arginine). Verify that mutant reads show equal representation across forward and reverse strands, with an observed VAF of ~35% confirming a clonal driver mutation. - Case 2 (Colon): KRAS Exon 2 G13D (chr12:25,398,281 C>T)
Navigate to
chr12:25,398,281. Confirm the codon 13 substitutionGGC > GAC(Glycine to Aspartate). Notice the high VAF (~62%), indicating copy-neutral Loss of Heterozygosity (LOH) where the remaining wild-type allele was deleted. - Case 3 (Breast): TP53 Exon 5 R175H (chr17:7,577,120 G>A)
Navigate to
chr17:7,577,120. Observe the dominant hotspot mutation in the DNA-binding core domain disrupting zinc coordination. Verify that the paired normal blood BAM shows zero mutant reads, confirming true somatic origin.
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.