From Molecules to Precision Medicine MScPHM Preparatory Course · Part 2 · Cell Biology · Module CB01
Contents
Module CB01 · Basic Cell Biology

The Cell: The Living Unit Behind Precision Medicine

Every genome you will sequence, every variant you will interpret, every drug matched to a patient, and every omics dataset you will model is read out from cells. This module builds that foundation — what cells are, how they work, how they fail in disease, and why cellular context decides what molecular data actually mean.

Start the module
Read time ~40 min core · ~95 min full Prerequisites None Level Foundational → applied Progress 0 / 0
SCHEMATIC · NOT TO SCALE · DRAG TO ROTATE
Internet needed for the live 3D. The rotating cell, the 3D molecule viewers, and the web fonts load from open online libraries (Three.js, 3Dmol.js, PubChem, Google Fonts). Offline, the course still works — diagrams, animations, simulators, and the quiz are all built in — but 3D molecular structures fall back to flat structural formulas.
O Orientation

Why cell biology matters for MScPHM

You do not need to become a cell biologist. You need enough cellular literacy to know where molecular data come from, how cells read genetic information, why cell type and context change everything, and why most disease is a disruption of normal cellular systems.

Precision health and medicine connects molecular variation to patient outcomes. That chain runs through the cell at every step: a DNA variant only matters if it changes something a cell does; an RNA measurement is a snapshot of what some population of cells was transcribing; a drug works by reaching a target inside or on a cell. Skip the cell, and omics data become numbers without meaning.

What this module unlocks downstream

Each topic you meet later in the MScPHM programme rests on a cell-biology idea introduced here.

Nucleus & chromatingenomics · transcriptomics · epigenomics · variant interpretation Membranes & transportdrug delivery · transporter pharmacogenomics · RNA/gene therapy Signalling & cycleprecision oncology · targeted therapy · biomarkers Single cells & tissuescRNA-seq · spatial biology · digital pathology · AI/ML
MScPHM connection

Across the programme you will work with genomics, transcriptomics, proteomics, metabolomics, pharmacogenomics, biomarker discovery, molecular diagnostics, precision oncology, immune profiling, single-cell RNA-seq, spatial biology, organoids, digital pathology, gene and cell therapy, and AI/ML on biomedical data. All of them measure or manipulate cells. This module is the shared vocabulary and reasoning chain underneath that work.

How to use this module

How to navigate this long module

This is a long module. There are two sensible ways through it. You can always jump around with the Contents menu (top of the page on smaller screens) or the section navigation.

Core route

Essential path · ~40–50 min

The essential thread from cell structure to data. Read the main text of: Orientation; §02 prokaryote vs eukaryote; §03 virtual cell (a quick look); §04 membrane; §05 transport; §06 nucleus; §07 proteins; §08 mitochondria; §11 signalling; §12 cell cycle; §13 cell death; §14 stem cells; §17 cancer; §18 pharmacogenomics; §19 gene & cell therapy; §21 omics & AI. Finish with the self-check quiz. You can skip the “Deeper explanation” panels on a first pass.

Full route

Everything · ~90–110 min

All of the above plus §01 living systems, §09 lysosomes/autophagy, §10 cytoskeleton, §15 tissues, §16 heterogeneity, and §20 models; every “Deeper explanation” panel; each Try it yourself with its suggested reasoning; the interactive virtual cell, signalling simulator, and disease-state sandbox; the recommended animations; and the final reflection.

Core idea, in one line

A cell is an organised, dynamic system that senses signals, uses energy, expresses genes, makes and moves proteins, divides, differentiates, adapts, and sometimes dies — and disease usually means one or more of those processes has gone wrong in a particular cell type and context.

Course connection

Throughout, precision-medicine relevance is written cautiously: cellular changes may affect, can influence, or are associated with outcomes, and almost always depend on cell type, tissue context, molecular state, and clinical context. That hedging is not vagueness — it is how responsible molecular medicine reasons.

Check your understanding

A collaborator says, “We found the variant, so we know the patient’s phenotype.” Why is that overconfident from a cell-biology standpoint?

01 The big picture

Cells as living systems

Learning objectives
  • Describe the cell as the basic functional unit of living organisms.
  • Explain why a cell is an information-processing system, not a bag of molecules.
  • Frame cell behaviour as inputs → cellular systems → outputs, dependent on context.

Cells are the basic functional units of life. Human health and disease arise from how cells — and the tissues they build — behave. A cell is not a static container: it continuously receives signals, processes information, spends energy, expresses genes, makes proteins, moves materials, divides, differentiates, adapts to stress, and sometimes dies on purpose. What a cell does depends both on its molecular components and on its context.

The cell as a decision-making system

Inputs are sensed and integrated by cellular systems; the same input can produce different outputs in different cells.

INPUTS nutrients · oxygen hormones · cytokines drugs genetic information mechanical & oxidative stress CELLULAR SYSTEMS genome transcriptome proteome metabolome signalling network organelles cytoskeleton integrate · interpret OUTPUTS secretion · movement division · death differentiation immune response metabolic change disease phenotype
Try it yourself — predict the response

For each input, jot what a cell might do (there is rarely one right answer — that is the point):

  • receives a growth signal  ·  senses DNA damage  ·  loses its nutrient supply
  • is exposed to a drug  ·  receives an inflammatory signal

Compare your predictions later against the disease-state sandbox.

Model reasoning · suggested answer
Suggested answerThere is rarely one right answer — the point is that a cell's response is conditional. A growth signal may drive a competent cell to divide, or do nothing if the pathway or context is wrong. Sensed DNA damage may trigger a pause-and-repair checkpoint or, if severe, apoptosis. Losing nutrients may push the cell toward autophagy, a stress/senescence programme, or death. A drug may have its intended effect, no effect, or off-target toxicity; an inflammatory signal may activate, recruit, or stress the cell. In every case the actual outcome depends on cell type, pathway state, dose, timing, and context.
Common misconceptions
  • Cells are not static structures — they are dynamic and constantly remodelling.
  • A gene does not act alone; it acts in cellular context.
  • A molecular change does not automatically produce a clinical phenotype.
  • Behaviour usually emerges from networks, not a single molecule.
MScPHM connection

Precision medicine connects molecular variation to cellular behaviour, then to tissue function, patient phenotype, and treatment response. Every later module re-uses this inputs → systems → outputs frame.

Deeper explanation — emergence and robustness

Cellular behaviour is emergent: feedback loops, redundancy, and crosstalk mean the system often buffers perturbations (robustness) until it cannot (tipping points). This is why a variant can be silent in one context and consequential in another, and why single-molecule reasoning so often fails to predict outcomes. Networks, dosage, timing, and cell state all matter.

Check your understanding

Why can the same growth-factor signal cause one cell to divide and another to do nothing?

02 Cell types

Prokaryotic and eukaryotic cells: why human cells are compartmentalised

Learning objectives
  • Contrast prokaryotic and eukaryotic cell organisation.
  • Explain why compartmentalisation lets eukaryotic cells run incompatible processes at once.
  • Recognise that human cells are eukaryotic, and why that matters for drugs and pathogens.

Prokaryotic cells — bacteria and archaea — generally lack a membrane-bound nucleus and membrane-bound organelles; their DNA sits in the cytoplasm. Human cells are eukaryotic: most nucleated human cells keep their genomic DNA inside a membrane-bound nucleus and run many processes inside dedicated organelles. An important exception is the mature human red blood cell, which extrudes its nucleus as it matures and so carries no genomic DNA. Compartmentalisation lets a eukaryotic cell maintain different chemical environments side by side — acidic digestion in a lysosome, oxidative metabolism in a mitochondrion, calcium stores in the ER — without those reactions interfering with one another.

Side by side (schematic, not to scale)

nucleoid (DNA, no membrane) PROKARYOTE (e.g. bacterium) cell wall · membrane · free ribosomes · ~1–5 µm EUKARYOTE (human cell) membrane-bound nucleus + organelles · ~10–30 µm
nucleus / DNA mitochondria ER / ribosomes Golgi lysosome
Try it yourself — which cell type?

Classify each, and note what gives it away: bacterium · liver cell (hepatocyte) · immune T cell · mature red blood cell · yeast · cancer cell. (Watch the red blood cell — mature human red cells are eukaryotic in origin but eject their nucleus and most organelles as they mature, a useful special case.)

Model reasoning · suggested answer
Suggested answerBacterium — prokaryote: no nucleus and no membrane-bound organelles. Hepatocyte, T cell, yeast, and cancer cell are all eukaryotic (nucleus + organelles); yeast is a single-celled fungus, and the cancer cell is a human cell with a disrupted genome and behaviour. The mature red blood cell is the special case: eukaryotic in origin, but it extrudes its nucleus and most organelles as it matures, so it carries no genomic DNA — a reminder that “human cell” does not always mean “has a nucleus.”
Common misconceptions
  • “Eukaryotic” does not mean “more evolved” in any simple sense.
  • Not all cells organise their DNA the same way.
  • Human cells are not bacteria — pathogen biology and human biology differ in important ways.
  • Antibiotics often exploit bacterial–human differences (e.g. distinct ribosomes, cell walls), but antimicrobial selectivity is not always simple.
MScPHM connection

Knowing human cell organisation helps you interpret host–pathogen interactions, antimicrobial selectivity, microbiome data, and human molecular medicine — and to read why a therapy designed for a human cell behaves differently in a microbe.

Deeper explanation — the endosymbiotic origin of mitochondria

Mitochondria (and, in plants, chloroplasts) are thought to descend from free-living bacteria engulfed by an ancestral eukaryotic cell. They retain their own small circular genome and bacteria-like ribosomes — which is why some antibiotics can affect mitochondria and why mitochondrial inheritance is generally maternal. You will revisit this when reasoning about mitochondrial disease and drug toxicity.

Sources for this section: OpenStax, Biology 2e (cell structure); Cooper, The Cell (NBK9839); Alberts et al., Molecular Biology of the Cell (NBK21054). The anucleate mature red blood cell is a standard exception in haematology references.

Check your understanding

Give one reason compartmentalisation is an advantage for a human cell.

03 Interactive · Tool 1

Explore a human cell

Click any structure to learn what it is, what it does, why it matters for MScPHM, and where disease or therapy touches it. Switch modes to re-colour the same cell around molecular data, disease, or therapy.

Select a structure

The whole cell

This is a schematic human (eukaryotic) cell. Click a labelled structure — or use Tab then Enter — to open its panel here. Then switch the lens (Overview → Molecular data → Disease → Therapy → Revision) to see the same structure four different ways.

Each structure keeps the module-wide organelle colour, reused in every later diagram.

Overview lens · all organelles labelled · schematic, not to scale
MScPHM connection

In Molecular data mode the cell is re-labelled by where each omics layer is read from: DNA/chromatin in the nucleus (genomics, epigenomics), mRNA in nucleus and cytoplasm (transcriptomics), proteins across the secretory pathway (proteomics), metabolites from mitochondria and cytosol (metabolomics). The same structure, four different measurements.

Accessibility & text alternative

Every structure is a keyboard-focusable button with a text label and description, announced through the live info panel. No information depends on colour alone — each organelle is named and described in text. A full list of structures and functions also appears in the glossary.

Offline: the 2D labelled cell and all of its information run entirely from this file and work without an internet connection. The 3D view depends on an external library (Three.js); if that library is unavailable it falls back automatically to the 2D view, which carries the same labels and text.

Check your understanding

Using the explorer’s Molecular data mode: a proteomics run reports an abundant secreted protein. Name two organelles that protein almost certainly passed through, and why.

04 Boundaries

The plasma membrane: boundary, barrier, and communication interface

Learning objectives
  • Describe the lipid bilayer and the proteins embedded in it.
  • Explain selective permeability and what crosses easily versus what needs help.
  • Connect membrane biology to drug delivery, signalling, and immune recognition.

The plasma membrane separates the cell interior from the outside world. It is built mainly from a phospholipid bilayer: each lipid has a water-loving (hydrophilic) head facing the watery environment and two water-fearing (hydrophobic) tails tucked inside. That arrangement makes the membrane selectively permeable — small nonpolar molecules slip through, while ions, sugars, and large or charged molecules generally need dedicated proteins. Embedded membrane proteins act as receptors, channels, transporters, enzymes, and adhesion molecules, so the membrane is also where the cell listens, imports, exports, and recognises its neighbours.

Architecture of the bilayer

Heads out, tails in, with proteins threaded through. The extracellular face often carries sugar chains (the glycocalyx).

Extracellular side Cytoplasm receptor channel transporter glycoprotein
phospholipid head + tails receptor ion channel transporter / carrier glycoprotein

3D membrane-molecule viewer

The real chemistry behind the cartoon — a phospholipid, cholesterol, and a fatty acid. Drag to rotate; scroll to zoom.

Loading structure…

Molecule information

Try it yourself — can it cross?

For each, decide whether it can diffuse straight through, needs a channel/transporter, needs receptor-mediated uptake, or needs a delivery strategy:

  • oxygen · steroid hormone  →  small & nonpolar?
  • sodium ion · glucose  →  charged or polar?
  • large protein · a nucleic-acid therapeutic  →  big and/or charged?
Model reasoning · suggested answer
Suggested answerOxygen and the steroid hormone are small and nonpolar, so they diffuse straight through the bilayer. The sodium ion (charged) and glucose (polar) need a channel or transporter. A large protein or a nucleic-acid therapeutic is too big and/or too charged to cross alone — it needs receptor-mediated uptake and/or an engineered delivery strategy, and then still faces endosomal escape. The recurring principle: charge and polarity, not size alone, set the route.
Common misconceptions
  • The membrane is not an impermeable wall — and not a free sieve either.
  • Small nonpolar molecules and ions behave very differently.
  • Channels and transporters are not the same mechanism.
  • Getting a therapeutic into the bloodstream is not the same as getting it into the right cell.
MScPHM connection

Membranes are central to drug transport, receptor signalling, pharmacogenomics, biomarker detection, immune recognition, and gene-therapy delivery. A surface receptor can be a drug target; a transporter variant can change how much drug enters a cell; the bilayer itself is the barrier lipid nanoparticles are engineered to cross.

Clinical / translation caution

Membrane permeability is one factor among many. Whether a drug reaches its target also depends on dose, distribution, metabolism, the specific cell type, and clinical context — so membrane reasoning informs, but does not decide, therapeutic outcomes.

Deeper explanation — fluidity, rafts, and asymmetry

The bilayer is fluid: lipids and many proteins diffuse laterally. Cholesterol tunes fluidity; specialised lipid microdomains (“rafts”) can concentrate signalling proteins. The two leaflets are also chemically asymmetric, and loss of that asymmetry (e.g. phosphatidylserine appearing on the outer face) is itself a biological signal — for instance, an “eat me” flag during apoptosis, which you will meet later.

Check your understanding

Why can a steroid hormone slip across the membrane while glucose generally cannot?

05 Transport

Moving substances across membranes

Learning objectives
  • Distinguish passive diffusion, facilitated diffusion, and active transport.
  • Describe endocytosis and exocytosis as vesicle-based bulk transport.
  • Explain why cellular uptake is necessary but not sufficient for therapeutic effect.

Cells move material across membranes in several ways. Passive diffusion needs no direct energy — molecules move down their concentration gradient. Facilitated diffusion uses channels or carriers for things that cannot cross alone, still down-gradient. Active transport spends energy (often ATP) to pump substances against a gradient. For bulk cargo, endocytosis wraps material into vesicles to bring it in, and exocytosis fuses vesicles with the membrane to release it. Which route applies depends on size, charge, polarity, the gradient, and the machinery present.

ANIMATION · Six ways across a membrane
channel carrier pump (ATP) vesicle
Speed
StepPress play, step through, or scrub. Each panel labels its mechanism.
Text transcript of this animation

The animation cycles through six ways a substance can cross the plasma membrane, showing one mechanism at a time:

  1. Simple diffusionSmall nonpolar molecules (e.g. O₂) cross the bilayer directly, down their concentration gradient — no protein and no energy needed.
  2. Facilitated diffusion · channelAn ion passes through a selective protein channel, still moving down its gradient.
  3. Facilitated diffusion · carrierA carrier protein binds a solute (e.g. glucose) and changes shape to ferry it across, down-gradient.
  4. Active transport · pumpA pump spends ATP to move a solute against its gradient (e.g. the Na⁺/K⁺ pump).
  5. EndocytosisThe membrane folds inward and pinches off, importing bulk cargo inside a vesicle.
  6. ExocytosisA vesicle fuses with the membrane and releases its contents outside the cell (e.g. secretion).
Try it yourself — choose the transport route
  • oxygen entering a cell · a sodium ion moving · glucose uptake
  • insulin being secreted · an antibody being taken up (conceptual)
  • a lipid nanoparticle delivering RNA (advanced, conceptual)
Model reasoning · suggested answer
Suggested answerOxygen — simple diffusion. A sodium ion — a channel down-gradient, or an ATP-powered pump against its gradient (active transport). Glucose — a carrier (facilitated diffusion). Insulin being secreted — exocytosis of secretory vesicles. An antibody taken up — receptor-mediated endocytosis. A lipid nanoparticle delivering RNA — endocytosis followed by endosomal escape into the cytoplasm. Match each to whether the cargo is small/nonpolar, charged/polar, or bulk.
Common misconceptions
  • Not all molecules cross equally — the membrane is selective.
  • Endocytosis is not the same as diffusion; it is active and vesicle-based.
  • Reaching the bloodstream is not the same as reaching the correct cells.
  • Uptake into a cell does not guarantee delivery to the right internal compartment.
MScPHM connection

Transport biology underpins pharmacokinetics, drug response, transporter pharmacogenomics (e.g. variants in uptake/efflux transporters), biologics, RNA therapeutics, and gene-therapy delivery. Endosomal escape — getting cargo out of the vesicle once inside — is a major real-world bottleneck you will see again in the gene-therapy section.

Deeper explanation — gradients, the sodium–potassium pump, and secondary transport

The Na⁺/K⁺-ATPase pumps sodium out and potassium in, burning ATP to maintain ion gradients. Those gradients are then “spent” by secondary active transporters that couple the downhill flow of one ion to the uphill movement of another molecule (e.g. glucose). So a single ATP-powered pump can indirectly drive many other transport events — a recurring theme in cell energetics and in how drugs that hit transporters ripple through cell physiology.

Check your understanding

A nucleic-acid therapy is taken into cells by endocytosis but stays trapped in the vesicle. Why might it still fail to work?

06 Genome

The nucleus: genome storage and gene regulation

Learning objectives
  • Describe how DNA is packaged into chromatin inside the nucleus.
  • Explain that gene expression depends on sequence and chromatin state, regulators, and context.
  • Connect nuclear biology to genomics, transcriptomics, and epigenomics.

The nucleus holds most of the cell’s DNA, wound around histone proteins and folded into chromatin. Genes can be transcribed into RNA, but whether a given gene is active depends on its sequence, how open or closed the surrounding chromatin is, which regulatory elements and transcription factors are present, and the cell’s overall state. The nucleus is therefore not a passive vault — it is a tightly regulated environment that decides which instructions are read, when, and how much.

From genome to transcript

Zooming in: nucleus → chromatin → a gene → an RNA transcript exported to the cytoplasm.

Nucleus chromatin (DNA + histones) a gene RNA transcript cytoplasm: protein production
DNA / chromatin (genomics, epigenomics) RNA (transcriptomics)
Try it yourself — from DNA to data

Put these in order and say which omics layer reads each: DNA sequence → chromatin accessibility → transcription → RNA output. Then answer: if a gene is present but its chromatin is closed, what would transcriptomics show?

Model reasoning · suggested answer
Suggested answerOrder: DNA sequence → chromatin accessibility → transcription → RNA output. Genomics reads the DNA sequence; epigenomics reads chromatin accessibility/state (and DNA methylation); transcriptomics reads the RNA output. If a gene is present but its chromatin is closed, transcriptomics would show little or no transcript — the gene is in the genome but not being expressed. Same sequence, different regulation.
Common misconceptions
  • Having a gene does not mean it is actively expressed.
  • DNA sequence alone does not fully determine cell behaviour.
  • RNA level does not always predict protein abundance.
  • The nucleus is not a simple bag of DNA — it is spatially and chemically organised.
MScPHM connection

Genomics, transcriptomics, epigenomics, variant interpretation, and gene therapy all depend on how DNA is organised and used inside the nucleus. A non-coding variant might matter because it sits in a regulatory element; an epigenetic change can silence a gene without altering its sequence at all.

Deeper explanation — why one genome makes hundreds of cell types

Nearly every cell in your body carries the same genome, yet a neuron and a hepatocyte behave completely differently. The difference is regulation: distinct chromatin states, transcription-factor combinations, and RNA processing give each cell type its own expression programme. This is the conceptual seed of cell identity, differentiation, and single-cell analysis — all covered later.

Sources for this section: OpenStax, Biology 2e; Alberts et al., Molecular Biology of the Cell (NBK21054) on chromatin, transcription, and gene regulation.

Check your understanding

Two patients share an identical protein-coding sequence for a gene, yet one expresses it strongly and the other barely at all. Give one nuclear explanation.

07 Protein factory

Making and moving proteins

Learning objectives
  • Explain translation by ribosomes, free versus ER-bound.
  • Trace the secretory pathway: ER → Golgi → vesicle → destination.
  • Connect trafficking to where a protein ends up and what it can do.

Ribosomes translate mRNA into chains of amino acids (polypeptides). Free ribosomes typically make proteins for the cytosol and certain organelles; ribosomes on the rough endoplasmic reticulum make proteins destined for secretion, the membrane, or the lysosome. The ER folds and quality-checks these proteins; the Golgi apparatus then modifies, sorts, and packages them, and vesicles carry them to their destinations. Where a protein ends up is as important as the fact it was made.

ANIMATION · The secretory pathway
ribosome / ER Golgi vesicle
Speed
StepFollow one protein from mRNA to its destination.
Text transcript of this animation

The animation follows one newly made protein along the secretory pathway, from gene to secretion:

  1. Translation on the rough ERA ribosome on the endoplasmic reticulum translates the mRNA; the new protein enters the ER lumen.
  2. ER to transport vesicleAfter folding and quality control, the protein buds off in a vesicle headed for the Golgi apparatus.
  3. Golgi processingThe Golgi modifies and sorts the protein, then packages it for its destination.
  4. Vesicle to the surfaceA vesicle carries the protein toward the plasma membrane (or to a lysosome or the membrane).
  5. SecretionThe vesicle fuses with the membrane and releases the protein — which may become a measurable blood biomarker.
mRNAtranslated Ribosome / ERfolded, checked Golgimodified, sorted Vesicledelivered Destinationsecretion · membrane · lysosome
Try it yourself — where does the protein go?

Route each: a secreted hormone · a membrane receptor · a cytosolic enzyme · a lysosomal enzyme · an antibody · a mitochondrial protein (the last is an advanced caveat — most mitochondrial proteins are made on free ribosomes and imported directly, not via the ER).

Model reasoning · suggested answer
Suggested answerThe secreted hormone, membrane receptor, lysosomal enzyme, and antibody all travel the secretory pathway: made on the rough ER, processed in the Golgi, then trafficked to their destination (out of the cell, into the membrane, or to the lysosome). A cytosolic enzyme is made on free ribosomes and stays in the cytosol. The mitochondrial protein is the caveat: most are made on free ribosomes and imported directly into the mitochondrion, not via the ER.
Common misconceptions
  • Making a protein is not the end of the story — location matters.
  • Abundance does not guarantee correct folding or function.
  • Secreted biomarkers depend on successful trafficking and release.
  • Not every protein goes through the ER/Golgi route.
MScPHM connection

Proteomics, secreted biomarkers, therapeutic antibodies, enzymes, receptors, and many drug targets all depend on synthesis, folding, trafficking, and localisation. A protein measured in blood (a secreted biomarker) only got there because the secretory pathway released it.

Clinical / translation caution

A high transcript level need not mean a functional protein at the right place. Folding failures, mislocalisation, or trafficking defects can all break the chain — one reason transcriptomic and proteomic signals sometimes disagree, and why interpretation needs both biology and clinical context.

Deeper explanation — quality control and the unfolded-protein response

The ER monitors folding; misfolded proteins are retained, refolded, or sent for degradation. If misfolded proteins accumulate, the cell mounts an unfolded-protein response that can rebalance the cell or, if stress persists, trigger death. This links protein trafficking to cell stress, neurodegeneration, and the behaviour of secretory cells under therapeutic load.

Check your understanding

A secreted biomarker drops in a patient’s blood. Name two distinct cellular steps whose failure could explain the drop (besides “less gene expression”).

08 Energy

Mitochondria and metabolism: powering cellular decisions

Learning objectives
  • Describe mitochondria as sites of ATP production and much more.
  • Explain that metabolism differs by cell type and state.
  • Connect mitochondrial biology to disease, toxicity, and metabolomics.

Mitochondria are major sites of ATP production in many human cells, using nutrients and oxygen to generate the cell’s usable energy currency. But calling them only “the powerhouse” undersells them: they also balance redox state, buffer calcium, produce metabolic building blocks, sense and signal stress, and help decide whether a cell lives or dies (apoptosis). Different cell types and disease states use metabolism differently — a resting cell, a dividing cancer cell, and an activated immune cell have very different energy and biosynthetic needs.

Inputs and outputs of mitochondrial metabolism

Energy is one output among several. Mitochondria are signalling hubs as much as power plants.

Mitochondrion (cristae = folded inner membrane) nutrients oxygen ATP (energy) metabolites · redox · Ca²⁺ stress & cell-fate signals

3D viewer · the energy currency

ATP and related metabolites. Look for the tail of three phosphate groups. It is transferring a phosphate to another molecule (hydrolysis) that releases usable free energy — the bonds themselves are not little batteries.

Loading structure…

Molecule information

A note on “high-energy bonds”

You will often read that ATP “stores energy in high-energy bonds.” Read that as shorthand, not literal chemistry. Breaking a bond always costs energy; what makes ATP useful is that its hydrolysis (or the transfer of a phosphate group to another molecule) is a strongly favourable reaction overall, releasing free energy that the cell couples to work. “High-energy bond” really means “a bond whose hydrolysis releases a lot of usable free energy,” not energy set free simply by snapping a bond.

Try it yourself — energy demand

Rank the energy/biosynthetic demand and say why: a muscle cell mid-contraction · an activating immune cell · a proliferating cancer cell · a firing neuron · a cell under drug-induced mitochondrial stress.

Model reasoning · suggested answer
Suggested answerAll are high-demand, but for different reasons, so there is no single correct ranking. A contracting muscle cell and a firing neuron need large, rapid ATP for mechanical work and ion pumping. An activating immune cell and a proliferating cancer cell rewire metabolism toward biosynthesis — building blocks for new cells, not just ATP — often leaning on glycolysis even when oxygen is available. A cell under drug-induced mitochondrial stress has impaired supply against its demand, which is exactly why such toxicity hits high-demand tissues hardest. Metabolism is a programmable readout of cell state, not a fixed rank.
Common misconceptions
  • Mitochondria are not only powerhouses — they signal and shape cell fate.
  • ATP level alone does not explain metabolism.
  • Metabolic pathways differ by cell type and context.
  • Both mitochondrial DNA and nuclear DNA matter for mitochondrial function.
MScPHM connection

Mitochondria and metabolism connect to cardiometabolic disease, ageing, cancer, pharmacogenomics, drug toxicity, rare disease, and metabolomics. Drug-induced mitochondrial toxicity is a real safety concern; metabolomic signatures often reflect mitochondrial state.

Clinical / translation caution

A metabolomic or mitochondrial signal is a clue, not a verdict. Whether mitochondrial change contributes to a disease depends on tissue, genetic background, and clinical context — interpret with care.

Deeper explanation — the Warburg shift and immunometabolism

Many proliferating cells (including cancer cells) rely heavily on glycolysis even when oxygen is available — the “Warburg effect” — partly to supply biosynthetic precursors, not just ATP. Activated immune cells rewire metabolism in their own characteristic ways (“immunometabolism”). These shifts illustrate the core message: metabolism is a programmable readout of cell state, which is exactly why metabolomics is informative in precision medicine.

Sources for this section: Cooper, The Cell (NBK9839) and Alberts et al. (NBK21054) on mitochondria and bioenergetics. The free energy of ATP hydrolysis follows standard biochemistry texts.

Check your understanding

Why might a drug that harms mitochondria cause problems even though it was designed to hit a completely different target?

09 Recycling

Cellular recycling and waste management

Learning objectives
  • Describe lysosomes and autophagy as regulated recycling, not just disposal.
  • Note the role of peroxisomes in handling lipids and reactive molecules.
  • Connect degradation pathways to disease and therapy response.

Lysosomes are acidic compartments full of digestive enzymes that break down and recycle materials — both cargo brought in by endocytosis and the cell’s own worn-out components. Through autophagy (“self-eating”), a cell can wrap a damaged organelle or protein aggregate in a membrane (an autophagosome) and fuse it with a lysosome for recycling. Peroxisomes handle certain lipid reactions and detoxify reactive molecules. These systems keep cells healthy; when they fail, disease can follow.

ANIMATION · Autophagy and lysosomal recycling
lysosome autophagosome damaged cargo
Speed
StepWatch damaged cargo become recycled building blocks.
Text transcript of this animation

The animation shows autophagy — how a cell recycles its own worn-out parts:

  1. Damaged cargoA worn-out organelle or protein aggregate is tagged for recycling.
  2. Autophagosome formsA double membrane grows around the cargo, fully enclosing it.
  3. Fusion with lysosomeThe autophagosome fuses with an acidic lysosome full of digestive enzymes.
  4. Recycled building blocksEnzymes break the cargo down and the building blocks return to the cytoplasm for reuse.
Try it yourself — recycle or secrete?

Decide the fate of each: an old organelle · extracellular material taken up by endocytosis · a misfolded protein · a signalling molecule meant for release · a lipid-processing reaction (which compartment handles which?).

Model reasoning · suggested answer
Suggested answerAn old organelle — autophagy, delivered to the lysosome. Endocytosed extracellular material — the endo-lysosomal route for degradation/recycling. A misfolded protein — quality-control degradation (lysosome via autophagy, or the proteasome). A signalling molecule meant for release — packaged and secreted by exocytosis (ER→Golgi→vesicle). A lipid-processing reaction — handled by the peroxisome (and smooth ER). Recycling and secretion are different fates routed through different compartments.
Common misconceptions
  • Lysosomes are not just “trash cans” — they are regulated recycling and signalling hubs.
  • Protein degradation is often biologically meaningful, not mere cleanup.
  • Autophagy is neither always good nor always bad — it is context-dependent.
  • Degradation pathways can shape how cells respond to therapy.
MScPHM connection

Lysosomal biology, autophagy, and degradation connect to rare inherited metabolic (lysosomal storage) disease, neurodegeneration, cancer, drug response, and biomarker interpretation. Some therapeutic strategies even exploit degradation pathways to remove disease-causing proteins.

Deeper explanation — lysosomal storage diseases

If a single lysosomal enzyme is missing or faulty, its substrate accumulates, swelling the lysosome and damaging the cell — the basis of lysosomal storage disorders. These are a classic example of how one molecular defect propagates through a cellular system to a tissue-level phenotype, and they are targets for enzyme-replacement and gene-therapy approaches you may meet later.

Check your understanding

Why can losing a single lysosomal enzyme cause a whole-cell (and whole-tissue) problem?

10 Structure & transport

The cytoskeleton: shape, movement, and transport

Learning objectives
  • Name the three main filament systems and their roles.
  • Explain how motor proteins move cargo along tracks.
  • Connect cytoskeletal dynamics to division, migration, and cancer invasion.

The cytoskeleton gives a cell its shape and mechanical strength, and it is the infrastructure for movement, division, and internal transport. Three filament systems do most of the work: actin filaments (cell shape, crawling, the contractile ring of division), microtubules (highways for transport and the spindle that separates chromosomes), and intermediate filaments (mechanical resilience). Motor proteins walk along these tracks, hauling vesicles and organelles to where they are needed. Crucially, the cytoskeleton is dynamic — constantly built and dismantled — not a rigid scaffold.

Three systems, different jobs

Actin filaments shape · crawling · division ring Microtubules cargo transport tracks · spindle Intermediate filaments mechanical resilience
Try it yourself — which cytoskeletal system?

Match each to a filament system: cell crawling · chromosome segregation · vesicle transport · mechanical resilience · the mitotic spindle · cancer-cell invasion.

Model reasoning · suggested answer
Suggested answerCell crawling — actin. Chromosome segregation and the mitotic spindle — microtubules. Vesicle transport — microtubule tracks with motor proteins. Mechanical resilience — intermediate filaments. Cancer-cell invasion — mainly actin-driven movement plus microtubule-based transport, i.e. coordinated remodelling. Microtubule-targeting chemotherapies exploit the spindle's dependence on microtubules.
Common misconceptions
  • The cytoskeleton is not rigid scaffolding — it is highly dynamic.
  • Cell shape and movement are molecularly regulated, not fixed.
  • Disrupting one filament system can ripple across many processes.
MScPHM connection

Cytoskeletal biology underlies cancer invasion and metastasis, cell division (a target of several chemotherapies), neuronal transport, immune-cell movement, and microscopy-based phenotyping in digital pathology and high-content screening.

Deeper explanation — why some chemotherapies target microtubules

Drugs such as taxanes and vinca alkaloids interfere with microtubule dynamics, jamming the mitotic spindle so dividing cells cannot complete division. This illustrates a recurring precision-oncology theme: a normal cellular process (here, microtubule turnover in mitosis) becomes a therapeutic target precisely because rapidly dividing cells depend on it heavily — though normal dividing tissues are affected too, which is why such drugs have characteristic side effects.

Check your understanding

Why is the cytoskeleton central to a cancer cell becoming invasive?

Key takeaways · cell structure
  • Human cells are eukaryotic and compartmentalised: membrane-bound organelles let incompatible chemistries run side by side. Most nucleated human cells hold their genomic DNA in the nucleus — mature red blood cells are the anucleate exception.
  • The plasma membrane is selective, not a wall or a sieve: charge and polarity, not size alone, decide whether something diffuses through or needs a channel, transporter, or delivery strategy.
  • Each organelle is a specialised subsystem — nucleus (genome + expression control), ER/Golgi (make, fold, and route proteins), mitochondria (energy plus signalling), lysosome/peroxisome (recycling and detox), cytoskeleton (shape, transport, division).
  • Structure maps to where molecular data come from and where therapies must reach — the thread the rest of the module follows.
11 Sensing

Cell signalling: how cells sense and respond

Learning objectives
  • Trace a signal from ligand to receptor to cellular response.
  • Explain why responses are context-dependent and pathways interconnect.
  • Connect signalling to targeted therapy and pathway biomarkers.

Cells constantly sense their environment. Signals — hormones, growth factors, cytokines, nutrients, oxygen levels, drugs, or stress — are detected by receptors. A receptor that binds its ligand sets off a signalling cascade inside the cell, relaying and amplifying the message until it changes something: gene expression, metabolism, movement, division, secretion, or death. Real signalling is a network, not a straight line — pathways branch, feed back, and cross-talk, and the same signal can mean different things in different cells.

Ligand → receptor → cascade → response

ligand receptor cascade (relay + amplify) nucleus: gene expression response: divide, move, secrete, die…
Try it yourself — signal to response

For each, predict a possible response (and note it depends on the cell): an insulin-like signal · an inflammatory cytokine · a growth factor · DNA-damage stress · a drug blocking a receptor · an immune-checkpoint signal (advanced). Then try the live signalling simulator.

Model reasoning · suggested answer
Suggested answerEach is context-dependent, so name a plausible response and its condition. An insulin-like signal → metabolic uptake/storage in a responsive cell. An inflammatory cytokine → activation, recruitment, or a stress programme. A growth factor → proliferation in a competent cell, or nothing if the pathway/context is wrong. DNA-damage stress → a checkpoint pause-and-repair, or apoptosis if severe. A drug blocking a receptor → reduced signalling, unless a downstream node is stuck on. An immune-checkpoint signal → dampened immune activity (the logic behind checkpoint blockade). The same input, read differently by different cell states.
Common misconceptions
  • One signal does not always produce the same response in every cell.
  • Pathways are not isolated straight lines — they branch and cross-talk.
  • A receptor being present does not guarantee a response.
  • More signalling is not always “better”; balance and feedback matter.
MScPHM connection

Many precision therapies act by modifying signalling pathways. Biomarkers frequently report pathway activity, receptor status, or downstream response — and choosing a therapy often means asking which pathway is active in which cell type.

Clinical / translation caution

Pathway logic here is qualitative and conceptual. Real therapeutic decisions require validated biomarkers, clinical evidence, and full patient context — the simulator that follows deliberately gives qualitative, non-deterministic output.

Deeper explanation — resistance via downstream activation

If a drug blocks a receptor but a downstream component is mutated to be constitutively active, the pathway can stay “on” regardless — a common route to targeted-therapy resistance. This is why precision oncology increasingly profiles whole pathways, not single receptors, and why combination strategies are used. You can reproduce this exact scenario in the signalling simulator.

Check your understanding

A drug blocks a receptor, yet the pathway stays active. What is one cellular explanation?

Key takeaways · information flow
  • Genetic information flows DNA → RNA → protein → function, and every step is regulated: having a gene is not the same as expressing it, and chromatin state can silence a present gene.
  • The arrows can break independently: mRNA ≠ protein ≠ correctly localised, functional protein, so transcriptomic and proteomic readouts can legitimately disagree.
  • Cells sense and interpret signals through networks, not single molecules; the same input can produce different outputs depending on cell type, receptor levels, and pathway state.
  • This is why responsible interpretation stays cautious — a molecular change may matter, depending on cell type and context, and usually needs further evidence.
12 Division

Cell cycle: how cells grow and divide

Learning objectives
  • Describe the G1, S, G2, and M phases and what happens in each.
  • Explain checkpoints as quality control before commitment.
  • Connect cell-cycle control to cancer biology.

The cell cycle is the ordered sequence by which a cell grows, copies its DNA, and divides. In G1 the cell grows and decides whether to commit; in S phase it replicates its DNA; in G2 it grows more and checks the copy; in M phase (mitosis) it segregates chromosomes and divides into two daughter cells. Checkpoints act like quality gates — they halt the cycle if DNA is damaged or incompletely replicated, so the cell does not pass on errors. Many cells also rest in a non-dividing state (G0). Losing cell-cycle control is central to cancer.

ANIMATION · The cell cycle and its checkpoints
G1 S G2 M
Speed
StepFollow one full cycle, including the checkpoint gates.
Text transcript of this animation

The animation follows one full cell cycle around a circle, passing through its checkpoint gates:

  1. G1 — growthThe cell grows and decides whether to commit to division.
  2. Checkpoint (G1/S)A quality gate: the cell proceeds only if conditions and DNA integrity allow.
  3. S — DNA replicationThe cell copies its DNA, duplicating each chromosome.
  4. G2 — growth & checkMore growth, and a check that replication finished correctly.
  5. Checkpoint (G2/M)Another gate before the cell commits to dividing.
  6. M — mitosis & divisionChromosomes are segregated and the cell divides into two daughter cells.
Try it yourself — where is the problem?

Place each fault in the right phase/checkpoint: DNA not fully replicated · DNA damage detected · an uncontrolled growth signal · a failed mitotic checkpoint · a drug that blocks mitosis · a cancer cell bypassing a checkpoint.

Model reasoning · suggested answer
Suggested answerDNA not fully replicated — caught at the G2/M checkpoint. DNA damage detected — the G1/S (or G2/M) checkpoint pauses for repair. An uncontrolled growth signal — pushes the G1→S commitment inappropriately. A failed mitotic checkpoint — mis-segregated chromosomes in M phase (aneuploidy). A drug blocking mitosis — stalls M phase (e.g. spindle poisons). A cancer cell bypassing a checkpoint — a disabled gate (e.g. loss of p53) that lets damaged cells divide. Checkpoints are quality gates, so naming the phase pinpoints the fault.
Common misconceptions
  • Not all cells divide constantly — many rest in G0.
  • Division is tightly regulated, not automatic.
  • DNA replication (S) and mitosis (M) are distinct events.
  • Cancer is not just “fast growth” — it involves disrupted regulation, survival, genome instability, and tissue context.
MScPHM connection

Cell-cycle biology underlies cancer genomics, targeted and cytotoxic therapies, the DNA-damage response, and proliferative biomarkers (e.g. Ki-67). Several drug classes work precisely by exploiting cell-cycle dependence.

Deeper explanation — checkpoints, p53, and synthetic lethality

Checkpoint regulators and tumour suppressors (such as p53) sense damage and decide whether to pause, repair, or trigger death. Tumours often disable these guards. Some therapies turn that loss against the cancer — for example, exploiting a defective DNA-repair pathway so the cell becomes lethally dependent on a second pathway a drug can block (synthetic lethality). This is conceptual background for precision-oncology strategies you will study later.

Check your understanding

Why are checkpoints so important, and what tends to happen to them in cancer?

13 Survival & death

Cell survival and cell death

Learning objectives
  • Distinguish apoptosis (regulated) from necrosis (uncontrolled injury).
  • List the range of stress responses: repair, adapt, arrest, senesce, or die.
  • Connect cell-fate decisions to therapy response and resistance.

Faced with stress, a cell has options: repair the damage, adapt, pause (arrest), enter a stable non-dividing state (senescence), or die. Apoptosis is a regulated, orderly form of cell death — the cell dismantles itself neatly and is cleared without inflaming the neighbourhood. Necrosis is typically uncontrolled: the cell swells and ruptures, spilling contents and provoking inflammation (though death biology includes several other regulated forms too). These decisions shape development, immunity, cancer, neurodegeneration, and how patients respond to treatment.

ANIMATION · Apoptosis vs uncontrolled injury
orderly (apoptosis) rupture (necrosis)
Speed
StepCompare the orderly route with the messy one.
Text transcript of this animation

The animation contrasts two fates side by side — orderly apoptosis (left) versus messy necrosis (right):

  1. Two stressed cellsLeft: a cell entering apoptosis. Right: a cell undergoing uncontrolled injury (necrosis).
  2. Apoptosis: orderlyThe left cell shrinks and blebs neatly; the right cell (necrosis) swells.
  3. Packaging vs ruptureThe left cell breaks into tidy membrane-bound packages; the right cell is about to burst.
  4. Cleared vs inflammatoryThe left cell's packages are cleared cleanly; the right cell ruptures and spills its contents, triggering inflammation.
Try it yourself — choose the cell fate

Predict an outcome for each: DNA damage that is repaired · severe damage triggering apoptosis · a cancer cell avoiding apoptosis · drug-induced tumour-cell death · an immune cell killing its target · chronic stress causing adaptation.

Model reasoning · suggested answer
Suggested answerRepaired DNA damage — the cell survives and resumes cycling. Severe damage → apoptosis (orderly, non-inflammatory death). A cancer cell avoiding apoptosis — survives despite damage; a hallmark of cancer and a route to drug resistance. Drug-induced tumour-cell death — many therapies work by pushing cells into apoptosis. An immune cell killing its target — induced (often apoptotic) death of that target. Chronic stress causing adaptation — senescence or an altered state rather than death. Fate depends on damage severity, the cell's death competence, and context.
Common misconceptions
  • Cell death is not always accidental — much of it is programmed and useful.
  • Apoptosis is not the same as necrosis.
  • Survival is not always “good” — inappropriate survival can drive cancer.
  • Cell-death markers need context to interpret.
MScPHM connection

Therapy response, drug resistance, cancer survival, immune killing, and toxicity all hinge on cell-fate decisions. Many cancer drugs work only if the diseased cell can still be pushed into death — and resistance often means that route has been blocked.

Clinical / translation caution

Detecting a cell-death marker does not by itself indicate benefit or harm — context (which cells, how many, where, and why) determines meaning. Avoid reading single markers as outcomes.

Deeper explanation — evading death as a hallmark of cancer

Resisting apoptosis is a recognised hallmark of cancer: tumour cells often tip the balance of pro- and anti-survival signals toward staying alive despite damage. Some targeted drugs aim to restore the death programme. This is why “does the diseased cell still die when pushed?” is such a clinically loaded question.

Sources for this section: Alberts et al., Molecular Biology of the Cell (NBK21054) on programmed cell death; OpenStax, Biology 2e.

Check your understanding

Why is the ability to undergo apoptosis so relevant to whether a cancer therapy works?

14 Identity

Stem cells and differentiation: one genome, many cell types

Learning objectives
  • Explain how one genome gives rise to many specialised cell types.
  • Define stem cells, potency, and differentiation in plain terms.
  • Connect cell identity to regenerative medicine and cell models.

Almost every cell in your body carries the same DNA, yet a neuron, a muscle cell, and an immune cell behave completely differently. The difference is which genes each cell expresses — its regulatory programme, not its sequence. Stem cells are relatively unspecialised cells that can both self-renew and give rise to specialised cells through differentiation. Cell identity is largely a matter of gene-expression and chromatin state, and it can be surprisingly plastic: under the right conditions cells can be reprogrammed.

From one stem cell to many fates

stem cell self-renew neuron muscle cell blood cell epithelial cell Same genome, different expression programmes
Try it yourself — what defines identity?

True or false (and why): “Two cells with identical DNA must behave identically.” Then list three things — besides DNA sequence — that distinguish a neuron from a skin cell.

Model reasoning · suggested answer
Suggested answerFalse. Identical DNA does not force identical behaviour — cells of the same genome can differ enormously. Three distinguishing factors besides sequence: which genes are expressed (transcription-factor combinations), the chromatin/epigenetic state that sets that programme, and the cell's environment and signalling history. Identity is regulation, not sequence — the idea behind differentiation, iPSCs, and single-cell analysis.
Common misconceptions
  • Different cell types are not genetically different in sequence — they differ in expression.
  • Cell identity is not always fixed; it can be plastic and reprogrammable.
  • “Stem cell” is not a single thing — potency varies widely.
  • Differentiation is regulated, not random.
MScPHM connection

Stem-cell and differentiation biology underpins regenerative medicine, induced pluripotent stem cell (iPSC) disease models, organoids, developmental disease, and cell therapies. Patient-derived iPSCs let researchers build disease-relevant cells carrying a patient’s own variants.

Deeper explanation — reprogramming and iPSCs

Differentiated cells can be reprogrammed back to a pluripotent state (iPSCs) and then re-differentiated into many cell types. This made patient-specific cell models possible: take a skin or blood sample, derive iPSCs carrying the patient’s genome, and grow the affected cell type to study mechanism or test drugs — a recurring tool in modern precision medicine.

Check your understanding

If every cell has the same DNA, what makes a neuron a neuron and not a liver cell?

Key takeaways · cell fate
  • The cell cycle is gated by checkpoints — quality controls that stop damaged or incompletely replicated cells from dividing; their loss is central to cancer.
  • Cells can live, pause (senescence), or die. Apoptosis is orderly, non-inflammatory death; many therapies work by pushing cells into it, so a cell's death competence shapes treatment response.
  • Cell identity is regulation, not sequence: cells with identical DNA differ because they express different genes — the basis of differentiation, stem cells, and iPSCs.
  • Fate is decided by damage severity, signalling, and context — not by any single molecule acting alone.
15 Context

Cells in context: communication, tissues, and the microenvironment

Learning objectives
  • Explain that cells rarely act alone — they form tissues and communicate.
  • Describe the microenvironment and why neighbours and matrix matter.
  • Connect tissue context to disease and to spatial omics.

Cells almost never act in isolation. They are organised into tissues, joined by adhesion structures, embedded in an extracellular matrix, and in constant conversation through secreted signals and direct contact. A cell’s neighbours, the matrix around it, available oxygen and nutrients, immune cells, and mechanical forces all shape its behaviour — this surrounding context is the microenvironment. The same cell can behave very differently depending on where it sits.

A cell is shaped by its neighbourhood

the cell neighbours matrix immune cells blood: O₂ & nutrients signals
Try it yourself — context changes the answer

Pick a cell behaviour (e.g. “divide” or “migrate”) and describe how it might change if you alter: oxygen level · neighbouring cell types · matrix stiffness · nearby immune cells · available signals.

Model reasoning · suggested answer
Suggested answerTake “divide” or “migrate.” Low oxygen (hypoxia) can shift metabolism and promote migration or angiogenic signalling. Neighbouring cell types can supply pro- or anti-growth signals. A stiffer matrix can promote migration/invasion (mechanotransduction). Nearby immune cells can suppress or support the behaviour. Available signals (growth factors, cytokines) tune it up or down. The same cell behaves differently depending on its microenvironment — which is why a dish result may not transfer to a real tissue.
Common misconceptions
  • A cell’s behaviour is not determined by its genome alone.
  • Studying isolated cells can miss tissue-level effects.
  • The microenvironment can change disease progression and therapy response.
  • “Where” a cell is can matter as much as “what” it is.
MScPHM connection

Tissue context and the microenvironment connect to the tumour microenvironment, immuno-oncology, fibrosis, spatial transcriptomics, digital pathology, and organoid models. Spatial omics exists precisely because location carries biological information that dissociated single-cell data can lose.

Deeper explanation — why the tumour microenvironment is a therapeutic target

Tumours are not just cancer cells: they recruit blood vessels, remodel matrix, and manipulate immune cells. Therapies increasingly target this environment (e.g. angiogenesis inhibitors, immunotherapies acting on immune cells in the tumour). It is a vivid case of context determining outcome — the same tumour cell can be controlled or unleashed depending on its neighbourhood.

Check your understanding

Give one reason a drug that kills cancer cells in a dish might underperform in a real tumour.

16 Variation

Cell heterogeneity: why averages can mislead

Learning objectives
  • Explain that “the same” cells often differ from one another.
  • Contrast bulk (averaged) with single-cell measurements.
  • Interpret a simple worked example of why an average can hide structure.

Cells that look alike are often not identical. Within a tissue — or even a single tumour — cells can differ in gene expression, state, and behaviour. A bulk measurement averages over all the cells in a sample; a single-cell measurement profiles them one at a time. Averages are efficient, but they can hide important structure: two very different cell populations can produce the same average.

Worked example — same average, different biology

Two samples have an identical bulk average for a marker gene, yet single-cell data reveal completely different compositions. Read across, then look at the single-cell breakdown.

Illustrative numbers (arbitrary units) to make the point — not real data.
SampleBulk average (marker)Single-cell viewWhat’s really there
Tumour A50every cell ≈ 50one uniform population, moderately expressing
Tumour B50half ≈ 0, half ≈ 100two populations — one silent, one high (e.g. a resistant subclone)

Bulk says A and B are the same. Single-cell says B contains a distinct high-expressing subpopulation that bulk averaging erased — which could be the subclone that survives therapy.

Try it yourself — read the cells, not the average

If Tumour B’s “≈100” cells are the ones that resist a drug, what would you expect to happen to the bulk average after treatment — and why might that look like the tumour “came back stronger”?

Model reasoning · suggested answer
Suggested answerAfter treatment the sensitive majority die, so the surviving population is enriched for the resistant “≈100” cells, which then expand. The bulk average — now dominated by resistant cells — can look similar or even “stronger,” because it is measuring a different, selected population rather than the original mix. That is clonal selection: the resistant subclone was present all along but hidden in the average. Single-cell data would have revealed it before relapse.
Common misconceptions
  • “The same” cells are not always truly identical.
  • An average can hide biologically important subpopulations.
  • A rare cell population can dominate an outcome (e.g. resistance).
  • Single-cell resolution can reveal what bulk methods miss — but adds its own noise.
MScPHM connection

Heterogeneity is the motivation for single-cell RNA-seq, spatial omics, rare-cell detection, and clonal analysis in cancer. Much of modern AI/ML in biomedicine exists to find structure (clusters, trajectories, rare states) that averages conceal.

Clinical / translation caution

Single-cell methods are powerful but introduce technical noise, dropout, and sampling bias. More resolution is not automatically more truth — interpretation needs careful quality control and biological grounding.

Check your understanding

A bulk RNA measurement is unchanged after therapy, but the patient relapses. How could single-cell data explain this?

17 Disease in depth

Cancer as a disease of cell biology

Learning objectives
  • Frame cancer as multiple cellular systems going wrong together.
  • Connect earlier topics (cycle, death, signalling, metabolism, microenvironment) to tumour behaviour.
  • See why precision oncology profiles cells and pathways, not just “a tumour”.

Cancer is not simply “cells growing too fast”. It is what happens when several cellular control systems fail together: cells gain self-sufficient growth signals, ignore stop signals and checkpoints, evade apoptosis, rewire metabolism, reshape their microenvironment, recruit a blood supply, and sometimes invade and spread. Genome instability accelerates the accumulation of these changes. Almost every topic in this module reappears here — which is exactly why cellular literacy is the backbone of precision oncology.

How earlier sections map onto tumour behaviour

cell cycle

Uncontrolled division

Checkpoints bypassed; growth signals stuck “on”.

cell death

Evading apoptosis

Damaged cells that should die, survive.

signalling

Rewired pathways

Mutations keep pro-growth pathways active.

metabolism

Altered fuel use

Biosynthesis for rapid growth (e.g. Warburg shift).

microenvironment

Reshaped neighbourhood

New blood supply; immune evasion.

heterogeneity

Subclones

Diverse cells; resistant populations can be selected.

Try it yourself — connect the hallmark to the section

For each tumour behaviour, name the earlier module section that explains it: “won’t stop dividing” · “won’t die” · “stuck-on growth signal” · “strange metabolism” · “builds its own blood supply” · “resists therapy”.

Model reasoning · suggested answer
Suggested answer“Won't stop dividing” → cell cycle / checkpoints (§12). “Won't die” → cell survival and death / evading apoptosis (§13). “Stuck-on growth signal” → signalling (§11). “Strange metabolism” → mitochondria and metabolism (§08). “Builds its own blood supply” → tissues and microenvironment / angiogenesis (§15). “Resists therapy” → heterogeneity and clonal evolution (§16). Cancer is several normal systems failing together — which is why precision oncology profiles broadly.
Common misconceptions
  • Cancer is not just fast growth — it is dysregulation across several systems.
  • One mutation is rarely the whole story.
  • Tumours are heterogeneous, not uniform.
  • “The genomics” alone does not fully predict behaviour or response — cell state and context matter.
MScPHM connection

This is the heart of precision oncology: profiling a tumour’s mutations, expression, pathways, immune context, and heterogeneity to choose targeted or immunotherapies — and to anticipate resistance. Cellular reasoning turns a gene list into a mechanism.

Clinical / translation caution

Molecular profiling informs, but does not dictate, treatment. Therapy decisions require validated biomarkers, clinical trial evidence, multidisciplinary judgement, and the individual patient’s context. Nothing here is clinical advice.

Deeper explanation — clonal evolution and resistance

A tumour is an evolving population. Therapy applies selection pressure; pre-existing or newly arising resistant subclones can expand, so the tumour that regrows may be biologically different from the one first treated. This connects directly to the heterogeneity section and explains why longitudinal profiling and combination strategies matter.

Sources for this section: The hallmarks-of-cancer framework as summarised in standard oncology and cell-biology texts; OpenStax, Biology 2e; Alberts et al. (NBK21054).

Check your understanding

Why is “cancer = fast growth” an inadequate definition? Name two other cellular failures involved.

18 Cells & drugs

Cells and drugs: the cellular basis of pharmacogenomics

Learning objectives
  • Trace a drug’s journey: reach the cell → enter → hit target → be metabolised → leave.
  • Explain how cellular differences change drug response.
  • Connect this to pharmacogenomics and dosing.

For a drug to work, several cellular steps must succeed: it has to reach the right cells, enter (or engage a surface target), bind its target, often be metabolised (activated or inactivated), and eventually be cleared. Differences in any of these — receptor levels, transporter activity, drug-metabolising enzymes, target sequence, or downstream pathway state — can change whether a drug helps, does nothing, or harms. Pharmacogenomics studies how inherited variation in these cellular components shapes drug response.

Where cellular variation changes drug response

Reachblood flow, delivery Entertransporters, membrane Acttarget binding Metaboliseenzymes activate/inactivate Clearefflux, excretion

A variant that slows a metabolising enzyme can raise drug levels (more effect or toxicity); one that speeds it can lower them (less effect). A transporter variant can change how much drug ever gets inside the target cell.

Try it yourself — predict the direction

Qualitatively, what might happen if a patient has: a slow drug-activating enzyme · a fast drug-inactivating enzyme · low target-receptor expression · a transporter that pumps the drug out · a downstream pathway mutation? (Then test similar logic in the disease-state sandbox.)

Model reasoning · suggested answer
Suggested answerA slow drug-activating enzyme → a prodrug is under-activated → reduced effect. A fast drug-inactivating enzyme → the drug is cleared quickly → reduced effect. Low target-receptor expression → less to act on → reduced effect. A transporter that pumps the drug out → lower intracellular exposure → reduced effect/resistance. A downstream pathway mutation → the pathway stays active despite the drug → resistance. Cellular variation shifts exposure and response in predictable directions — the logic behind pharmacogenomics. These are qualitative illustrations, not dosing advice.
Common misconceptions
  • The same dose does not affect everyone the same way.
  • Drug response is not determined by a single gene in isolation.
  • “In the body” is not the same as “in the target cell at the right concentration”.
  • Pharmacogenomics is probabilistic guidance, not a guarantee.
MScPHM connection

This is the cellular foundation of pharmacogenomics, precision dosing, drug-response biomarkers, and adverse-event prediction. Clinical pharmacogenomic guidelines (e.g. for certain enzymes and transporters) rest on exactly this cell-level reasoning.

Clinical / translation caution

Pharmacogenomic effects are probabilistic and context-dependent; they interact with dose, other drugs, organ function, and disease state. Predictions here are conceptual and must never be used for actual prescribing.

Deeper explanation — prodrugs and the activation problem

Some drugs are given as inactive prodrugs that a cellular enzyme must convert into the active form. If a patient’s variant makes that enzyme sluggish, the drug may never be activated — so a “correct” dose can still fail. The mirror case (an enzyme that inactivates a drug too quickly) lowers exposure. Both show why the cell, not just the prescription, determines effect.

Sources for this section: FDA Table of Pharmacogenomic Associations / biomarkers in drug labeling, and CPIC guidelines (cpicpgx.org) for gene–drug pairs. Illustrative only — not dosing guidance.

Check your understanding

A standard dose has no effect in one patient and causes toxicity in another. Give two distinct cellular explanations.

19 Engineering cells

Gene and cell therapy: why delivery is the hard part

Learning objectives
  • Outline the chain a genetic therapy must complete inside a cell.
  • Explain why each step is a potential point of failure.
  • Appreciate why “into the body” is not “into the right cell, correctly”.

Genetic and RNA-based therapies aim to change what a cell does at the level of its instructions. Conceptually, the therapy must complete a whole chain: be delivered to the body, reach the right cells, enter them, escape the vesicle it was taken up into (endosomal escape), reach the correct compartment (cytoplasm or nucleus), act (be translated, edit, or regulate), persist for the right duration, and do so safely — without triggering harmful immune responses or off-target effects. Every arrow in that chain needs its own evidence.

The delivery chain — every step can fail

deliveredto body reach rightcell type enterthe cell escape theendosome ⚠ correctcompartment act oninstructions persist rightduration stay safe(immune/off-target) ⚠
Try it yourself — find the weak link

A therapy is detectable in the blood but has no effect. Walk the chain and propose where it could be failing: not reaching the cell? not entering? trapped in the endosome? wrong compartment? not acting? cleared too fast? Each is a different fix.

Model reasoning · suggested answer
Suggested answerMatch the symptom to a step in the delivery chain. Not reaching the cell → a targeting/biodistribution problem. Not entering → an uptake problem. Trapped in the endosome → failed endosomal escape (a classic bottleneck). Wrong compartment → it reached the cytoplasm but needed the nucleus, or vice versa. Not acting → poor editing/expression efficiency once there. Cleared too fast → insufficient persistence. Each failure needs a different fix, and success at one step does not prove the next — so every arrow needs its own evidence.
Common misconceptions
  • “Delivered into the body” is not the same as “delivered into the correct cell”.
  • Entering a cell is not the same as reaching the right compartment.
  • Editing or expression must be efficient and safe, not just possible.
  • Each step in the chain needs its own evidence — success at one does not imply the next.
MScPHM connection

This is the cell-biology core of gene therapy, RNA therapeutics (e.g. mRNA, siRNA), CAR-T and other cell therapies, and delivery technologies such as viral vectors and lipid nanoparticles. Most real-world setbacks are delivery and safety problems, not a failure of the genetic idea itself.

Clinical / translation caution

This is a conceptual overview, not a protocol. Real gene and cell therapies involve rigorous safety, manufacturing, regulatory, and ethical considerations far beyond this module. No technique here should be read as instructions.

Deeper explanation — why endosomal escape is the classic bottleneck

Many therapeutics are taken up by endocytosis into vesicles that mature toward lysosomes — where the cargo would be degraded. To work, the therapy must escape into the cytoplasm first. This single step (endosomal escape) is one of the most stubborn problems in delivery, and a major focus of nanoparticle and vector design. It ties straight back to the transport and lysosome sections.

Sources for this section: FDA and manufacturer information for approved therapies (e.g. Casgevy / exagamglogene autotemcel); ClinicalTrials.gov for investigational therapies (e.g. NTLA-2001 / nexiguran ziclumeran, Phase 3). Approval status and indications vary by date and region.

Check your understanding

Why can a gene therapy be present in a patient yet still have no therapeutic effect? Name two distinct failure points.

Key takeaways · disease & therapy
  • Disease is usually a disruption of normal cellular systems in a particular cell type and context; cancer, for example, is several systems failing together plus heterogeneity and microenvironment — not simply fast growth.
  • Context and heterogeneity change everything: the tumour microenvironment alters drug access and behaviour, and a bulk average can hide the rare subpopulation that drives relapse.
  • The same standard dose can under- or over-treat different people because of cellular variation in drug-activating/inactivating enzymes, transporters, and targets (pharmacogenomics).
  • For gene, cell, and RNA therapies, delivery is the hard part: reaching the right cell, entering it, escaping the endosome, reaching the right compartment, acting, persisting, and staying safe — each a distinct, evidence-requiring step. Examples span approved therapies (e.g. Casgevy for sickle-cell disease) and investigational ones (e.g. Phase 3 CRISPR editing for transthyretin amyloidosis); approval status varies by date and region.
20 Studying cells

How we study cells: models and their limits

Learning objectives
  • Compare common cell models from 2D cultures to organoids.
  • Reason about what each model captures and misses.
  • Apply appropriate caution when generalising from a model to a patient.

We rarely study human cells directly inside a living person. Instead we use models — each a trade-off between control and realism. 2D cell lines are convenient and reproducible but can drift from real biology. Primary cells are closer to reality but harder to grow. iPSC-derived cells can carry a patient’s own genome. Organoids are 3D mini-tissues that recover some structure and cell–cell context. Animal models add whole-body context but differ from humans. No model is “the truth”; each answers some questions and not others.

Realism vs control (a rough trade-off)

ModelCaptures wellMisses / caution
2D cell linescale, reproducibility, screeningtissue context; can drift genetically/phenotypically
Primary cellscloser to native statelimited lifespan; donor variability
iPSC-derivedpatient genome; specific cell typesmaturity; differentiation variability
Organoid3D structure; some cell–cell contextincomplete vasculature/immune context
Animal modelwhole-body, systemic effectsspecies differences from humans
Try it yourself — pick the model

Which model would you reach for, and what would you stay cautious about? high-throughput drug screen · study a patient’s own mutation in the affected cell type · model tissue architecture · test systemic toxicity.

Model reasoning · suggested answer
Suggested answerHigh-throughput drug screen → 2D cell lines (fast, scalable), but stay cautious about drift from real biology and missing tissue/immune context. A patient's own mutation in the affected cell type → patient-derived or iPSC-derived cells, cautious about maturity and variability. Tissue architecture → organoids or 3D models, cautious about missing vasculature and immune components. Systemic toxicity → whole-organism models, cautious about species differences and ethics. Each model answers some questions and misses others; “worked in cells” is not “works in people.”
Common misconceptions
  • Cell-line results do not always translate to patients.
  • No single model is fully representative.
  • “It worked in cells” is not “it will work in people”.
  • Model choice shapes — and can bias — conclusions.
MScPHM connection

Model literacy matters for drug discovery, disease modelling, organoid-based precision medicine, and interpreting preclinical evidence. When you read a study, asking “which cells, which model, and what does it miss?” is a core critical-appraisal skill.

Check your understanding

Why might a drug that cures a cancer cell line still fail in patients? Tie your answer to model limitations.

21 Cells to data

From cells to data: omics, multi-omics, and AI/ML

Learning objectives
  • Map each omics layer back to the cellular structure it measures.
  • Explain why integrating layers (multi-omics) adds insight.
  • Separate correlation/measurement from mechanism when models find patterns.

Every omics dataset is a measurement of something cells were doing. Tie each layer back to the biology you now know: genomics reads the DNA sequence (nucleus); epigenomics reads chromatin state (nucleus); transcriptomics reads RNA output (nucleus → cytoplasm); proteomics reads the proteins made and trafficked (ribosomes, ER, Golgi, and beyond); metabolomics reads metabolic state (mitochondria, cytosol). Multi-omics integrates these layers, and AI/ML helps find structure across them — but a pattern a model finds is a hypothesis about cells, not automatic proof of mechanism.

Mini activity — match the omics layer to the cell

Each row is a real measurement type. Cover the right column, predict the cellular source, then check.

Omics layerMeasures…Cellular source you learned
GenomicsDNA sequence & variantsnucleus — DNA/chromatin
Epigenomicschromatin accessibility, methylationnucleus — chromatin state
TranscriptomicsRNA levelsnucleus → cytoplasm — transcripts
Proteomicsprotein abundance/modificationribosomes · ER · Golgi · whole cell
Metabolomicssmall-molecule metabolitesmitochondria · cytosol
Single-cell / spatialper-cell or location-resolved profilesheterogeneity · tissue context
Try it yourself — measurement vs mechanism

A model finds that high expression of gene X correlates with poor outcome. List three different cellular explanations (X causes it; X is a bystander marker of a state that causes it; a confounder such as cell-type proportion drives both). Which experiments would distinguish them?

Model reasoning · suggested answer
Suggested answerThree explanations for “high X correlates with poor outcome”: (1) X is causal — it drives the bad outcome; (2) X is a bystander marker of an underlying state (e.g. a cell type or stress programme) that is the real driver; (3) a confounder — e.g. a shift in cell-type proportions raises both X and risk. To distinguish them: perturb X directly (knockdown/overexpression) and see whether outcome changes; use single-cell/spatial data to see which cells express X and whether composition drives the bulk signal; and control for known confounders. A correlation is a hypothesis about cells, pending validation — not a mechanism.
Common misconceptions
  • Omics data are measurements of cells, not abstract numbers.
  • Correlation in data is not mechanism in cells.
  • A predictive model can be useful without revealing causation.
  • “The model found it” does not mean “the biology confirms it”.
MScPHM connection

This closes the loop the module opened: genomics, transcriptomics, proteomics, metabolomics, multi-omics integration, single-cell and spatial methods, and AI/ML all measure or model cellular biology. Your cellular literacy is what lets you interpret these data responsibly rather than over-claim.

Clinical / translation caution

A data-derived signature is a hypothesis until validated. Generalisation depends on cohort, platform, cell-type composition, batch effects, and clinical context. Treat model output as a lead for biological and clinical follow-up, not a conclusion.

Sources for this section: Concepts align with OpenStax, Biology 2e and NCBI Bookshelf (NBK21054), plus standard single-cell and multi-omics literature.

Check your understanding

An ML model predicts drug response from RNA data with high accuracy. Why is that not the same as understanding the mechanism — and why can it still be useful?

Key takeaways · omics & AI
  • Every omics layer is a cellular readout: genomics/epigenomics from DNA and chromatin, transcriptomics from RNA, proteomics along the secretory route, metabolomics from mitochondria and cytosol.
  • Averages can mislead — single-cell and spatial methods exist because bulk measurements blur heterogeneity and cell-type composition.
  • An accurate predictive model is a hypothesis about cells, not a mechanism: correlation may be causal, a bystander marker, or a confounder, and needs targeted validation.
  • Reading molecular data without cellular context is the core error this module guards against — cell type, tissue, state, and clinical context change what the numbers mean.
22 Watch

See it in motion: recommended animations

Professionally produced animations are a great complement to the diagrams above. Each opens in a new tab; all content remains © its respective source. These are external resources — this course is not affiliated with the creators or the platforms hosting these external resources, and links may change over time.

Biology: Cell Structure

YouTube · Nucleus Medical Media

A clear, beautifully rendered tour of a eukaryotic cell — the plasma membrane, nucleus and nucleolus, endoplasmic reticulum, ribosomes, Golgi apparatus, mitochondria, and the cytoskeleton — showing how the organelles you explored above fit together in three dimensions. An excellent visual companion to sections 03–10.

Credit: © Nucleus Medical Media. Hosted on YouTube.

Watch on YouTube ↗

Cell Structure and Functions, Animation

YouTube · Alila Medical Media

A clear, professionally produced animation walking through the structure and function of a eukaryotic cell — the plasma membrane, nucleus, endoplasmic reticulum, ribosomes, Golgi apparatus, mitochondria, and other organelles — showing how each contributes to the life of the cell. A strong visual complement to the virtual cell explorer and sections 03–10.

Credit: © Alila Medical Media. Hosted on YouTube.

Watch on YouTube ↗

Chemistry of the Cell: Water, Organic Molecules & Inorganic Ions

JoVE Science Education · Cell Biology / Biochemistry of the Cell

A concise grounding in what cells are actually made of: why water (~70% of the cell) and its polarity and hydrogen bonding underpin cellular structure and reactions; the four families of organic molecules (proteins, carbohydrates, nucleic acids, lipids); and the inorganic ions — sodium, potassium, magnesium, calcium, phosphate, chloride — that are under 1% of cell mass yet essential (for example, calcium in signalling). A useful chemistry primer before the organelle tour, especially for the data-focused cohort.

Credit: © JoVE (jove.com). Adapted by JoVE from OpenStax Biology 2e and Cooper GM, The Cell: A Molecular Approach. A JoVE subscription or free trial is required to view the full video.

Watch on JoVE ↗
While you watch — active viewing

Keep a running list: every time the video names a structure, jot what you now know it does and one MScPHM link (a measurement, a target, or a delivery hurdle). Compare with the virtual cell afterwards.

T2 Interactive · Tool 2

Cell-signalling simulator

Change the inputs and watch a signalling pathway respond qualitatively. This is a concept model, not a quantitative biological prediction — it shows direction and logic (and how confident you should be), never clinical outcomes.

Inputs

Pathway activity: —

Move a slider to begin. With no ligand and a normal receptor, the pathway is largely quiet.

Confidence: conceptual only
MScPHM connection

Try the classic resistance scenario: turn on the receptor-blocking drug but set downstream activation to “stuck-on”. The pathway stays active despite the drug — the logic behind much targeted-therapy resistance, and a reason precision oncology profiles whole pathways.

What this tool is and isn’t

Output is deliberately qualitative and non-deterministic. It illustrates reasoning — “this may increase activity, with low/medium/high confidence” — and never produces a number, a diagnosis, or a treatment recommendation.

Check your understanding

Using the simulator, find two different input combinations that both leave the pathway active even with the blocking drug on. What do they have in common?

T3 Interactive · Tool 3

Disease-cell-state sandbox

Introduce a cellular change and explore what might follow: the process affected, a possible data signature, a possible disease relevance, how confident you should be, and what evidence you’d need next. Everything here is cautious and conceptual — a reasoning aid, never a diagnosis.

Introduce a cellular change

moderate
Process most likely affected
Possible data signature
Possible disease relevance
What a candidate drug might do
Evidence you would still need
Confidence: depends on context
MScPHM connection

This sandbox rehearses the core precision-medicine move: from a molecular change, reason cautiously toward a cellular process, a measurable signature, and a hypothesis — while staying honest about uncertainty and the evidence still required. Notice how changing cell type or tissue context can change the likely consequence of the same molecular change.

Important

Outputs are deliberately hedged (“may”, “could”, “depends on context”) and are simplified teaching scenarios. They are not diagnoses, predictions for any real patient, or clinical guidance.

Check your understanding

Pick one molecular change and run it in two different cell types. Why does the “possible disease relevance” shift, even though the molecular change is identical?

Self-check

Apply what you know

These questions test reasoning, not recall — interpreting situations the way you will in the programme. Pick an answer to get feedback that explains the thinking and points back to the relevant section. There is also a final, ungraded reflection.

Learning objectives → assessment map

How the module’s learning objectives line up with the self-check questions below. Use it to find the sections behind any question you miss.

Learning objectiveSection(s)Quiz question(s)
Reason from molecular change to cellular effect, conditional on cell type and context§O, §01, §14Q1, Q12, Q22, Q29
Explain compartmentalisation and why human cells are eukaryotic (incl. the anucleate red-blood-cell exception)§02Q21, Q26
Predict transport routes from a molecule's charge, polarity, and size§04, §05Q2, Q24
Distinguish gene presence, expression, and chromatin regulation§06Q4
Separate mRNA level from functional, correctly localised protein; trace the secretory pathway§07Q5, Q23
Describe mitochondria beyond ATP and recognise mitochondrial drug toxicity§08Q6
Link a single lysosomal defect to tissue-level disease; read autophagy as context-dependent§09Q7, Q27
Connect cytoskeletal systems to movement, division, and microtubule-targeting drugs§10, §12Q8
Interpret signalling as networks and explain resistance downstream of a blocked receptor§11, §17Q9, Q28
Explain cell-cycle checkpoints and their loss in cancer§12Q10
Relate apoptosis competence to whether a therapy works§13Q11
Use microenvironment and model limits to explain dish-vs-patient differences§15, §20Q13, Q18
Read heterogeneity and clonal evolution behind relapse; reject 'cancer = fast growth'§16, §17Q14, Q15, Q25
Reason about pharmacogenomic variation shifting drug response§18Q16
Walk the gene/RNA-therapy delivery chain and locate failure points§05, §19Q3, Q17
Map omics layers to cellular sources; separate correlation from mechanism§06, §21Q19, Q20, Q30

Q29 checks responsible interpretation language and applies across all sections.

Q1 A variant is found in a patient’s genome. Why is it premature to claim it explains their disease? Interpretation · §O, §01

Q2 A steroid hormone enters cells easily; a small ion does not. The best explanation is… Application · §04

Q3 An RNA-based therapy is detected inside cells but has no effect. A cell-biology-savvy first hypothesis is… Application · §05, §19

Q4 Two patients have the identical coding sequence of a gene, but very different expression. Most likely reason? Analysis · §06

Q5 A transcript is abundant, but the protein is nearly absent. Which is the weakest conclusion? Interpretation · §07

Q6 A drug aimed at one target unexpectedly harms mitochondria. The clearest worry is… Application · §08

Q7 A child has a single missing lysosomal enzyme. Why can one enzyme cause whole-tissue disease? Analysis · §09

Q8 Why do several chemotherapies target microtubules? Application · §10, §12

Q9 A targeted drug blocks a receptor, yet the pathway stays active. Best cellular explanation? Analysis · §11

Q10 Why are cell-cycle checkpoints clinically important? Interpretation · §12

Q11 Why is the ability to undergo apoptosis relevant to whether a cancer drug works? Analysis · §13

Q12 Identical DNA, yet a neuron and a hepatocyte differ completely. Why? Application · §14

Q13 A drug kills cancer cells in a dish but underperforms in a real tumour. A strong cell-biology reason? Analysis · §15, §20

Q14 Bulk RNA is unchanged after therapy, but the patient relapses. Single-cell data could show… Interpretation · §16

Q15 Why is “cancer = fast growth” inadequate? Analysis · §17

Q16 A standard dose fails in one patient and is toxic in another. Two cellular explanations? Application · §18

Q17 A gene therapy is present in a patient but ineffective. Which pair are both legitimate failure points? Application · §19

Q18 A drug cures a cell line but fails in patients. The best model-literacy statement is… Interpretation · §20

Q19 Transcriptomics measures RNA. Which cellular structures are most directly its source? Application · §06, §21

Q20 An ML model predicts drug response from RNA with high accuracy. The most rigorous reading is… Analysis · §21

Q21 Why is compartmentalisation an advantage for human (eukaryotic) cells? Application · §02

Q22 The same growth-factor signal makes one cell divide and another do nothing. Why? Analysis · §01, §11

Q23 A secreted blood biomarker falls. Besides reduced transcription, name a valid cellular cause. Application · §07

Q24 Glucose needs help to cross the membrane but oxygen does not. The principle is… Interpretation · §04, §05

Q25 After a targeted therapy, a tumour regrows and is biologically different. Best explanation? Analysis · §16, §17

Q26 A mature red blood cell is unusual because it… Interpretation · §02

Q27 Why might autophagy be described as “context-dependent” in disease? Analysis · §09

Q28 Why does precision oncology increasingly profile whole pathways, not single receptors? Application · §11, §17

Q29 Which statement best reflects responsible interpretation language in this module? Interpretation · all

Q30 A multi-omics study finds gene X RNA correlates with poor outcome. The safest next step is… Analysis · §21

Final reflection (ungraded)

In a few sentences: Pick any molecular measurement you expect to use in the MScPHM programme (a variant, an RNA level, a protein, a metabolite). Explain what cells it comes from, one reason it might mislead if read without cellular context, and what you would check next.

Show a model answer & what a strong response includes

Model answer (example). “An elevated RNA level for a gene comes from transcription in the nucleus, read out across some population of cells in my sample. Read without context it could mislead because (a) high mRNA need not mean high functional protein at the right location, and (b) a bulk value can be driven by a shift in cell-type proportions rather than per-cell change. Next I would check protein-level and single-cell/spatial data, confirm which cell types contribute, and look for confounders (batch, composition) before claiming the gene is mechanistically important.”

A strong response includes: the cellular source of the measurement; at least one specific reason context matters (mRNA≠protein, heterogeneity/averaging, microenvironment, model limits); cautious language (“may”, “depends on context”); and a concrete validation step. Compare yours against these — not for a score, but to check your reasoning chain.

Reset your intuitions

Misconceptions to leave behind

If you remember nothing else, remember these — they are the errors that most often trip up newcomers reading molecular data.

The big ones
  • A gene is not a phenotype. A molecular change only matters if it alters what a cell does — in the relevant cell type and context.
  • Having a gene ≠ expressing it. Regulation and chromatin state decide what is actually used.
  • mRNA ≠ protein ≠ function ≠ location. Each step can break; readouts can legitimately disagree.
  • The membrane is selective, not a wall or a sieve. Charge and polarity, not size alone, set the route across.
  • Into the body ≠ into the right cell, correctly. Delivery, uptake, endosomal escape, and compartment all matter.
  • Mitochondria are more than powerhouses. They signal, buffer calcium, and help decide cell fate.
  • Cancer ≠ fast growth. It is multi-system dysregulation plus heterogeneity and microenvironment.
  • The average can lie. Bulk data can hide the subpopulation that drives the outcome.
  • Correlation ≠ mechanism. A model’s pattern is a hypothesis about cells, pending validation.
  • “It worked in cells” ≠ “it works in people”. Every model misses something.
  • Same dose ≠ same effect. Cellular variation (enzymes, transporters, targets) shifts drug response.
  • Context is king. Cell type, tissue, state, and clinical setting change the meaning of nearly everything.
A–Z Reference

Glossary

Plain-language definitions for non-biologists. Type to filter.

Cell
The basic functional unit of living organisms; an organised, dynamic system that senses, uses energy, and acts.
Prokaryotic
A cell type (bacteria, archaea) without a membrane-bound nucleus or membrane-bound organelles.
Eukaryotic
A cell type whose genomic DNA is normally held in a membrane-bound nucleus, alongside membrane-bound organelles. Human cells are eukaryotic; most nucleated human cells fit this description, though some mature cells (e.g. red blood cells) lose their nucleus as they specialise.
Plasma membrane
The selectively permeable lipid bilayer (plus embedded proteins) enclosing the cell.
Cytoplasm
The watery interior of the cell (cytosol plus organelles), where many reactions occur.
Organelle
A specialised subcellular structure with a defined role (e.g. nucleus, mitochondrion).
Nucleus
The membrane-bound compartment holding most of the cell’s DNA and controlling gene expression.
Chromatin
DNA packaged with histone proteins; its open/closed state helps control which genes are expressed.
Ribosome
The machine that translates mRNA into a protein; free in the cytosol or bound to the ER.
Endoplasmic reticulum (ER)
A membrane network for protein folding/quality control (rough ER) and lipid/calcium handling (smooth ER).
Golgi apparatus
The organelle that modifies, sorts, and packages proteins and lipids for delivery.
Mitochondrion
Site of much ATP production; also handles redox, calcium, and cell-fate signalling. Has its own small genome.
Lysosome
An acidic, enzyme-filled compartment for degradation and recycling.
Peroxisome
An organelle handling certain lipid reactions and detoxifying reactive molecules.
Cytoskeleton
Dynamic filament systems (actin, microtubules, intermediate filaments) for shape, movement, and transport.
Receptor
A protein that detects a specific signal (ligand) and triggers a cellular response.
Transporter
A membrane protein that carries specific solutes across, often changing shape; can require energy.
Channel
A membrane protein forming a pore that lets specific ions/molecules pass, usually down their gradient.
Endocytosis
Bringing material into the cell by engulfing it in a membrane vesicle.
Exocytosis
Releasing material by fusing a vesicle with the plasma membrane.
Signalling pathway
A relay of molecules that transmits and amplifies a signal into a cellular response; usually networked.
Cell cycle
The ordered sequence (G1, S, G2, M, with checkpoints; G0 = resting) by which a cell grows and divides.
Mitosis
The M-phase process that segregates duplicated chromosomes into two daughter cells.
Apoptosis
Regulated, orderly cell death that avoids spilling contents and inflaming surroundings.
Necrosis
Typically uncontrolled cell death from injury, with swelling, rupture, and inflammation.
Senescence
A stable, non-dividing cell state; can be protective or contribute to ageing/disease depending on context.
Stem cell
A relatively unspecialised cell that can self-renew and give rise to specialised cells.
Differentiation
The process by which a cell becomes specialised by changing which genes it expresses.
Microenvironment
The neighbouring cells, matrix, signals, and conditions surrounding a cell that shape its behaviour.
Cell heterogeneity
Differences among cells that appear similar; why averages can hide important subpopulations.
Single-cell omics
Measuring molecular profiles one cell at a time to reveal heterogeneity bulk methods average away.
Organoid
A 3D, lab-grown mini-tissue that recovers some structure and cell–cell context of a real organ.
Biomarker
A measurable indicator (molecular or cellular) used to infer a biological or clinical state.
Pharmacogenomics
How inherited variation in cellular components shapes drug response, toxicity, and dosing.
Gene therapy
Changing a cell’s genetic instructions therapeutically; success hinges on safe, targeted delivery.
Multi-omics
Integrating several molecular layers (genome, transcriptome, proteome, metabolome…) for fuller insight.
Precision medicine
Tailoring prevention and treatment to molecular and cellular features of individuals, with appropriate caution.