Clinical Molecular Genetics

• Most common short-limbed dwarfism; autosomal dominant; ≈80% de novo
• Gene FGFR3: normally limits bone growth → pathogenic variants make it constitutively active (gain of function)
• Most have p.Gly380Arg: 99% a G→A transition (purine for purine), 1% a G→C transversion
• De novo mutations arise exclusively on the paternal allele

1. PCR-amplify a 164 bp piece of exon 10
2. Digest: SfcI cuts c.1138G>A → 109 + 55 bp; MspI cuts c.1138G>C → 107 + 57 bp
3. Run on a gel: the variant creates the cut site, so mutant alleles give the smaller bands

Restriction digest gel

• Lane C (het control): 164 + 109 + 55; patients D–F uncut = negative

• Confirm a clinical diagnosis; prenatal diagnosis when short limbs are seen on ultrasound; prenatal diagnosis when both parents have achondroplasia
• Aa × Aa → 25% AA (homozygous: lethal), 50% Aa (achondroplasia), 25% aa (average stature)
• Incomplete dominance: the heterozygote is intermediate between the two homozygotes → you can read genotype from phenotype

Incomplete dominance

• RR red · Rr pink · rr white

• Most common fatal AR disease in Caucasians; CFTR at 7q31; 27 exons; >2,000 pathogenic variants
• ΔF508 most common; p.W1282* on 60% of Ashkenazi CF chromosomes; c.3120+1G>A common in African Americans (12.3%)

1. PCR-amplify the CFTR region; denature to single strands
2. Add a wild-type probe and a mutant probe (different labels) plus a common probe that sits right next to them
3. DNA ligase joins probes only when the base at the junction is a perfect match
4. Separate the ligated products by electrophoresis

OLA principle

• Ligase seals only a perfect match

Homozygous ΔF508

• Mutant peak only at F508

• Genotype–phenotype; therapy depends on the variant
• Diagnosis when sweat chloride is equivocal or the baby is too young
• Prenatal diagnosis; routine prenatal/Ashkenazi carrier screening; male infertility work-up (CBAVD)
• A negative carrier screen ≠ not a carrier: panels miss rare variants → calculate residual risk with Bayes

• X-linked; dystrophin is huge (2.4 Mb, 79 exons)
• Large deletions ≈60% (98% detectable by multiplex PCR) · premature stops ≈15% · duplications ≈5%
• Deletion size doesn't predict phenotype; the reading frame does: in-frame → truncated but partly functional protein (amino + carboxy ends kept) → Becker (milder, allelic)

Dystrophin deletions

• A 46% in-frame deletion → mild Becker

• Like OLA: paired probes hybridize side-by-side and are ligated, then all are amplified with one primer pair
• Each probe pair has a unique length → separated by electrophoresis
• Peak height = relative copy number vs a reference: shorter peak = deletion; taller = duplication

MLPA

• Quantitative: dosage of each exon

• Guides therapy: exon skipping (eteplirsen) restores the reading frame; ribosomal read-through (ataluren) for premature stops
• Females may have a classic dystrophinopathy or be asymptomatic carriers
• Males with DMD don't reproduce; males with BMD can

• Most common inherited cause of intellectual disability in males
• CGG repeat in the 5′ UTR of FMR1 expands to >200 → promoter methylation → gene silenced → loss of function
• Repeats are dynamic: they expand as they pass through generations (tri-, tetra-, pentanucleotide repeats)

One gene, three disorders

• Premutation → FXTAS, POI; full → fragile X

• Three primers: a labeled forward primer 5′ of the repeat, a reverse primer 3′ of it, and a second reverse primer that binds within the CGG repeat at many sites
• Products form a "ladder" of peaks; run on a capillary sequencer → count repeats; a ladder that runs off the scale = full mutation

TP-PCR primers

• One primer lands inside the repeat

Results

• (a) male 32 · (b) female 20/64 premutation · (c) female 29/>200 full

• Different modification and expression of alleles depending on the parent of origin
• The imprinted allele = the silent, inactivated one
• Paternally imprinted gene → expressed from the maternal allele: UBE3A (Angelman)
• Maternally imprinted genes → expressed from the paternal allele: SNRPN, ZNF127 (PWS)
• During gametogenesis imprints are erased and reset by the individual's own sex → you re-imprint the chromosome from your opposite-sex parent

Imprint reset

• Each generation reprograms by sex

• >96% of Cs in the PWS/AS critical region are methylated on the maternal allele; none on the paternal
• Bisulfite converts unmethylated C → U; methylated C is protected
• Primers specific for C (maternal, 174 bp) or U (paternal, 100 bp) + a common primer
• Only maternal band → PWS; only paternal band → AS
• Detects deletions, UPD and imprinting defects (≈99% of PWS, ≈78% of AS); misses UBE3A mutations

Methylation PCR gel

• NL: both · AS: paternal only · PW: maternal only

• Dideoxynucleotides (no 3′-OH) stop chain extension; color-labeled ends are read in order (T red, A green, C blue, G black)
• Detects point mutations (missense, nonsense, synonymous, splice), small indels (frameshifts), mosaicism

Heterozygous variant

• Double peak at the variant base

Mosaicism

• Small second peak (arrow)

ACMG/AMP five tiers: Pathogenic · Likely pathogenic · Variant of uncertain significance (VUS) · Likely benign · Benign.

• VUS: find others with the same phenotype (e.g., GeneMatcher) or do functional lab work

• Phenotypes caused by many genes → multigene panels (can cost ≈$6,000) or ES/GS
• No answer? Heterogeneity, mutation in an un-analyzed region, or too many candidate variants (homozygosity)
• CF by exome still must test the 23-variant ACMG panel: 2 deep intronic variants are invisible to exomes
• ≈4–5% of patients with a clinical exome have two single-gene disorders (more with consanguinity or UPD)

• Incidental findings: results unrelated to the reason for testing. ACMG keeps a list of highly penetrant, medically actionable genes to return regardless of indication (ACMG-59 in 2016 → SF v3.0 with 73 genes, updated yearly)
• Pre-test counseling: VUS, misattributed paternity, false positives/negatives, carrier status, late-onset disease, susceptibility alleles
• Minors: normally not tested for adult-onset disorders (choose at 18), but ACMG recommends returning some such results found in a minor because of benefit to the family

• Mostly SNP-based; companies interpret risk differently; patients bring reports to their physician
• "Talent" tests make unsupported claims
• Ancestry ("recreational genetics"): can strip ethnic identity, reveal misattributed paternity and family secrets, and feed forensic databases
• Health tests: e.g., BRCA1/2 for only 3 Ashkenazi founder variants (not the most common in the general population); ≈40% of variants in DTC raw data are false positives