Fundamental Biology

DNA (deoxyribonucleic acid) is a double-stranded helix made of nucleotide monomers. Each nucleotide has three parts: a deoxyribose sugar (the backbone), a phosphate group (linking sugars), and one of four nitrogenous bases (A, T, C, G). The two strands run antiparallel (one 5'→3', the other 3'→5') and pair via specific hydrogen bonds - this base complementarity is the foundation of all molecular biology.

• Adenine (A) ↔ Thymine (T) - 2 hydrogen bonds (weaker pairing; AT-rich regions melt first)
• Cytosine (C) ↔ Guanine (G) - 3 hydrogen bonds (stronger; GC-rich genomes are more thermostable)
• Chargaff's rule: in any dsDNA, %A = %T and %C = %G - key quality check for DNA purity
• The sugar-phosphate backbone runs 5'→3'; all polymerases synthesise new DNA in this direction
• DNA coils around octamers of histone proteins (H2A, H2B, H3, H4) to form nucleosomes (~147 bp per nucleosome)
• Nucleosomes compact further into 30 nm fibres → loops → chromosome domains → chromosomes
• Histone modifications (H3K27ac = active enhancer; H3K4me3 = active promoter) regulate gene expression

The central dogma describes the directional flow of genetic information in all living cells. Understanding each step is essential for interpreting variant consequences - a frameshift in an exon disrupts the coding sequence; a variant at a splice site disrupts pre-mRNA processing; a variant in the 3' UTR may destroy a miRNA binding site.

• Transcription: RNA Pol II reads the template strand (3'→5') and synthesises pre-mRNA (5'→3')
• 5' capping: 7-methylguanosine cap added co-transcriptionally - protects mRNA from exonucleases, required for ribosome binding
• Splicing: spliceosome removes introns at GT...AG splice sites (GU-AG rule in RNA); ~95% of human genes are multi-exonic
• Alternative splicing: one gene can produce multiple protein isoforms - BRCA1 has 23 splice variants; dystrophin has 78 exons
• Poly-A tail: ~200 adenines added after poly-A signal (AATAAA); essential for nuclear export and mRNA stability
• Translation: ribosome reads mRNA codons (3-base triplets) 5'→3'; tRNAs deliver amino acids matching each codon
• Start codon: AUG (methionine); stop codons: UAA, UAG, UGA - nonsense variants create premature stop codons
• Post-translational modification (PTM): phosphorylation, ubiquitination, glycosylation - variants can disrupt PTM sites

A typical human gene spans 10–100 kb of genomic DNA but produces only 1–3 kb of protein-coding mRNA. Understanding every element of gene structure is critical for interpreting variant consequences: a variant 2 bp from an exon-intron boundary is a potential splice-site variant even if it's technically intronic.

• Promoter (−2000 to TSS): TATA box at −30 bp binds TFIID; CpG islands at TSS in ~70% of human genes
• Transcription start site (TSS): where RNA Pol II begins - variants here can reduce transcription initiation
• 5' UTR: from TSS to start codon; upstream ORFs (uORFs) in 5' UTR can dramatically reduce translation efficiency
• Exon: protein-coding sequence; on average 300 bp; human BRCA1 exon 11 is unusually large at 3.4 kb
• Intron: spliced out; average 3,400 bp; intronic variants at ±1/±2 positions from exon are classified as splice-site (PVS1)
• 3' UTR: from stop codon to poly-A tail; miRNA binding sites here regulate mRNA degradation (e.g., VEGFA 3' UTR)
• Enhancers: can be up to 1 Mb away but loop to contact promoters via CTCF-cohesin; ENCODE has mapped >1 million
• Pseudogenes: non-functional copies of genes; PMS2 has multiple pseudogenes that confound variant calling

Variants differ in their size, mechanism of pathogenicity, and detection method. Choosing the right sequencing approach depends on which variant type you're looking for.

• SNP (Single Nucleotide Polymorphism): >1% frequency; mostly benign; e.g., rs334 (HBB Val6Glu) causes sickle cell - rare exception where a common SNP is pathogenic
• SNV (Single Nucleotide Variant): any frequency; clinically relevant rare variants; e.g., BRCA1 c.5266dupC is actually a single-base insertion but historically called "185delAG" in old nomenclature
• Missense: amino acid change; e.g., TP53 p.Arg248Trp is a gain-of-function hotspot in many cancers; classified by SIFT, PolyPhen, CADD
• Nonsense: creates premature stop codon; e.g., CFTR c.1624G>T p.Glu542* causes cystic fibrosis; typically Pathogenic via PVS1
• Frameshift indel: shifts reading frame; e.g., BRCA1 c.5266dupC p.Gln1756Profs*74 - founder mutation in Ashkenazi Jewish population; truncates protein
• In-frame indel: maintains reading frame; may preserve partial function; e.g., DMD in-frame deletions cause Becker MD (milder than frameshift = Duchenne)
• CNV deletion/duplication: e.g., SMN1 deletion (both copies) causes spinal muscular atrophy; PMP22 duplication causes Charcot-Marie-Tooth 1A
• Trinucleotide repeat: CAG expansion >36 in HTT causes Huntington disease; FMR1 CGG expansion >200 causes Fragile X (anticipation)
• Chromosomal translocation: BCR::ABL1 fusion (Philadelphia chromosome) in CML; creates constitutively active tyrosine kinase

A genetic counsellor refers a family with Duchenne muscular dystrophy (DMD). Cascade testing is requested for at-risk male relatives. The affected proband's DMD variant is a large exon 45-52 deletion - a CNV spanning ~170 kb.

• WES misses large CNVs: this deletion spans multiple exons but WES shows normal coverage in some exons due to read depth averaging
• Correct test: multiplex ligation-dependent probe amplification (MLPA) or chromosomal microarray to detect the deletion
• WES result: appears normal - a false negative for the family's mutation
• Lesson: WES detects SNVs and small indels well, but large deletions/duplications require CNV-specific tools (CNVkit, GATK gCNV) or MLPA
• ATGC Flow includes CNV detection; always check coverage plots for DMD/PARK2/PMP22/SMN1 in WES reports