Genomics

Question 1 (10 marks)

(a) (1 each)

• Genome — the complete set of DNA (all genes + non-coding sequence) in an organism/cell. (1)
• Transcriptome — the complete set of RNA transcripts (mRNA + non-coding RNA) present in a cell/tissue at a given time. (1)
• Proteome — the complete set of proteins expressed by a cell/tissue/organism at a given time. (1)

(b) Genome is essentially constant in every nucleated cell of the organism (same DNA). (1) The transcriptome/proteome change because gene expression is regulated — different genes are switched on/off depending on cell type, time, and environment. (2) Example: a neuron and a muscle cell share the same genome but express different genes → different transcriptomes/proteomes; or a cell up-regulates heat-shock proteins under stress. (1)

(c) Any two (1.5 each, cap 3):

• Alternative splicing — one gene → multiple mRNA isoforms → multiple proteins.
• Post-translational modification — phosphorylation, glycosylation etc. create distinct protein forms.
• Alternative promoters/RNA editing also acceptable.

Question 2 (12 marks)

(a) Step-by-step (1 each, role explained):

1. Template DNA — single-stranded DNA to be sequenced, provides the base order to copy. (1)
2. Primer — short oligonucleotide anneals to template, provides free 3′-OH start point for polymerase. (1)
3. DNA polymerase — extends the primer by adding complementary nucleotides 5′→3′. (1)
4. Four dNTPs — normal deoxynucleotides supplying A, T, G, C for chain elongation. (1)
5. ddNTPs — dideoxynucleotides incorporated randomly; each labelled (fluorescently) to identify the terminating base. (1)
6. Reaction generates a nested set of fragments of every possible length, each ending in a labelled ddNTP. (1)

(b) A ddNTP lacks the 3′-hydroxyl (–OH) group on its sugar. (2) Without a 3′-OH, no phosphodiester bond can form with the next incoming nucleotide, so elongation stops. (1)

(c) Fragments separated by size via capillary electrophoresis — shorter fragments migrate faster. (1) As each fragment passes the detector, its terminal ddNTP fluoresces one of four colours identifying the base. (1) Reading fragments in order of increasing length (colour by colour) reconstructs the sequence 5′→3′. (1)

Question 3 (10 marks)

(a)

• WGS — sequences the entire genome, coding + non-coding + regulatory + intergenic. (1)
• Exome — sequences only exons (~1–2% of genome, the protein-coding portion). (1)
• Preferred (exome): when hunting a Mendelian disease mutation in coding regions — cheaper, less data, higher coverage of exons. (1.5)
• Not preferred (exome): when the variant lies in non-coding/regulatory regions or is a structural/intronic variant — WGS needed. (1.5)

(b)

• Massive parallelism — millions of fragments sequenced simultaneously → huge throughput and low cost per base. (2)
• Short reads — reads are typically ~50–300 bp long. (2)
• Consequence: short reads must be computationally assembled/aligned to a reference; repetitive regions are hard to resolve. (1)

Question 4 (12 marks)

(a) A SNP is a variation of a single nucleotide at a specific genomic position that is common in the population (typically >1% frequency). (2) Unlike a mutation (any change, may be rare/novel/disease-causing), a SNP is a common inherited variant. (1)

(b)

• Compare allele/genotype frequencies of SNPs between a group with a trait/disease (cases) and a group without (controls). (2)
• Measure statistical association between each SNP and the phenotype. (1)
• A "hit" = a SNP whose frequency differs significantly, marking a genomic region linked to the trait. (1)

(c) Bonferroni corrected threshold: $\alpha_{corrected} = \frac{0.05}{1{,}000{,}000} = 5 \times 10^{-8}$ (3) Why: testing a million SNPs means a million hypothesis tests; at $\alpha=0.05$ we'd expect ~50,000 false positives by chance. Correction controls the family-wise error rate. (2)

Question 5 (8 marks)

(a) ENCODE showed most non-coding DNA is biochemically functional / transcriptionally active, not "junk." (1) It assigned functional activity to a large fraction of the genome. (1) Two categories (1.5 each):

• Regulatory elements — promoters, enhancers, silencers controlling gene expression.
• Non-coding RNA genes — miRNA, lncRNA etc. (transcription factor binding sites / DNase-hypersensitive sites also acceptable)

(b) Sequences conserved across species (evolutionary constraint) despite being non-coding indicate they are functionally important — natural selection preserves them. (3)

Question 6 (8 marks)

(a) Pharmacogenomics — the study of how an individual's genome affects their response to drugs (efficacy and toxicity). (2) Example (3): CYP2D6 variants alter metabolism of codeine — poor metabolizers get little pain relief (codeine not converted to morphine), whereas ultra-rapid metabolizers risk toxicity. (Warfarin/VKORC1, TPMT/thiopurines also accepted.)

(b) Precision medicine tailors treatment/drug choice/dose to the individual's genotype (and other data), (2) whereas one-size-fits-all gives the same standard drug/dose to all patients regardless of genetic differences in response. (1)

[
  {"claim":"Bonferroni corrected threshold for 1e6 tests at alpha=0.05 is 5e-8","code":"alpha=Rational(5,100); n=1000000; result = (alpha/n == Rational(5,100000000)) and abs(float(alpha/n)-5e-8)<1e-15"},
  {"claim":"Expected false positives at nominal 0.05 over 1e6 tests is 50000","code":"result = (Rational(5,100)*1000000 == 50000)"},
  {"claim":"Exome is roughly 1-2 percent of genome; 1.5 percent lies in range","code":"result = (1 <= 1.5 <= 2)"},
  {"claim":"ddNTP lacks 3-prime OH so no phosphodiester bond forms","code":"terminates = True; result = (terminates == True)"}
]