Define genome, proteome, and transcriptome

Overview

The genome, proteome, and transcriptome represent three fundamental molecular inventories of a cell, each capturing genetic information at a different stage of gene expression. Understanding these concepts is crucial for genomics, systems biology, and personalized medicine.

Core Concepts

The Mathematical Relationship (Information Flow)

Formula From Genome to Proteome

Let's derive the quantitative relationships:

Step 1: Define genome size $G$ (total base pairs) $G = G_{\text{genic}} + G_{\text{intergenic}}$ where $G_{\text{genic}}$ is the DNA occupied by genes (protein-coding AND non-coding RNA genes, including their introns), and $G_{\text{intergenic}}$ is the remaining regulatory/repetitive DNA between genes.

WHY separate these? Both protein-coding genes AND non-coding RNA genes (rRNA, tRNA, miRNA, lncRNA) produce functional transcripts. Only $G_{\text{genic}}$ is directly transcribed. Note that the protein-coding exon fraction (the sequence that actually specifies protein) is only ~1.5% of the human genome — much smaller than the full genic fraction, because most genic DNA is introns.

Step 2: Number of genes $N_g$ relates to the genic DNA: $N_g \approx \frac{G_{\text{genic}}}{L_{\text{avg}}}$ where $L_{\text{avg}}$ is the average full gene length (~27,000 bp for humans, including introns).

WHY use full gene length here (not the 1.5% exon figure)? To be internally consistent, the numerator and denominator must both refer to the same thing: whole genes (introns + exons). We do NOT divide the exonic 1.5% by the full gene length — that would mix two different measures. (Cross-check: $20{,}000 \times 27{,}000 \approx 5.4 \times 10^8$ bp ≈ 17% of the genome is genic, which matches known estimates.)

Step 3: Transcriptome size $T$ at time $t$ : $T(t) = \sum_{i=1}^{N_g} n_i(t) \cdot L_i$ where:

$n_i(t)$ = number of transcript copies of gene $i$ at time $t$
$L_i$ = length of mature transcript $i$ (after splicing)

WHY this form? Each gene contributes proportionally to its expression level AND its mature transcript length. A highly expressed gene with 1000 copies contributes more total RNA than a lowly expressed gene with 10 copies.

Step 4: Proteome size $P$ (number of distinct protein species): $P \geq N_g^{\text{coding}} \cdot S_{\text{avg}} \cdot M_{\text{avg}}$ where:

$N_g^{\text{coding}}$ = number of protein-coding genes (~20,000 in humans)
$S_{\text{avg}}$ = average number of splice isoforms per gene (~3-5 in humans)
$M_{\text{avg}}$ = average number of distinct co-existing modification states per protein. For most proteins this is small (roughly 1–3 dominant states); only a minority of heavily regulated proteins have many combinatorial states.

WHY the inequality? This is a lower bound. Protein complexes and rare highly-modified proteins add further diversity not captured by these average multipliers.

Worked example Worked Example 1: Prokaryote (

E. coli) Given: E. coli genome = 4.6 × 10⁶ bp, ~4,300 genes Find: Theoretical transcriptome size if all genes expressed equally at 10 copies each.

Solution:

Step 1: Average gene length
L_avg = 4.6 × 10⁶ bp / 4,300 genes 
     ≈ 1,070 bp per gene

WHY this step? We need transcript length. In prokaryotes, 
genes have no introns and the genome is gene-dense, so 
gene length ≈ mRNA length.

Step 2: Total transcriptome size
T = Σ(copies × length) for all genes
 = 4,300 genes × 10 copies × 1,070 bp
 = 4.6 × 10⁷ bp = 46 Mb of RNA

WHY? Each gene contributes (copies × length) to total RNA.
Step 3: Compare to genome
T/G = 46 Mb / 4.6 Mb = 10×

Interpretation: The transcriptome is 10× larger than the genome in total nucleotides because each gene is copied multiple times.

Worked example Worked Example 2: Human Liver Cell vs Neuron

Given: Both cells have identical genomes (3.2 Gb, ~20,000 genes) Scenario:

Liver cell: 5,000 genes active, expressing enzymes for detoxification
Neuron: 8,000 genes active, expressing synaptic proteins

Find: Why do proteomes differ despite identical genomes?

Solution:

Step 1: Transcriptome differences
Liver: Genes for P450 enzymes highly transcribed
      (albumin, cytochrome P450 family: 100-1000 copies each)
Neuron: Genes for neurotransmitter receptors highly transcribed
       (NMDA receptors, ion channels: 100-1000 copies each)

WHY different? Gene expression is REGULATED. Transcription 
factors activate different gene sets in each cell type.

Step 2: Proteome consequences
Liver proteome: Rich in detoxification enzymes
               ~10,000 protein species (lower splicing complexity)
Neuron proteome: Rich in signaling molecules
               ~13,000 protein species (neurons use extensive 
               alternative splicing for receptor diversity)

WHY more protein species than transcripts? Alternative splicing 
and post-translational modifications multiply one transcript
into several protein variants.

Key insight: Same genome → different transcriptomes → different proteomes. Gene regulation creates cellular identity.

Worked example Worked Example 3: Temporal Dynamics - Heat Shock Response

Scenario: E. coli shifted from 37°C to 42°C

Time course (using E. coli's own chaperones DnaK (an Hsp70-family protein) and GroEL (Hsp60-family)):

t = 0 min (before heat):
 Heat shock genes (dnaK, groEL): 1-2 transcripts each
 Proteome: ~2,000-3,000 protein species; DnaK present 
           at a modest baseline level
t = 5 min (during heat):
 Heat shock genes: 50-100 transcripts each
 WHY? Heat activates the alternative sigma factor σ³² (RpoH), 
 which redirects RNA polymerase to heat-shock promoters
 Transcriptome has changed, but proteome lags 
 (translation takes time)
t = 15 min:
 Heat shock transcripts: still elevated
 Proteome: DnaK rises several-fold above baseline. 
 (Total protein content of one E. coli cell is only 
 ~2-4 × 10⁶ molecules, so DnaK stays well within that budget.)
 WHY the delay? Translation and accumulation are gradual
t = 60 min (return to 37°C):
 Heat shock transcripts: back to 1-2 (rapid mRNA decay)
 Proteins: still elevated (protein decays slowly)
 WHY proteins persist? Protein half-life = many minutes to hours, 
 mRNA half-life = a few minutes

Lesson: Transcriptome changes are fast (minutes), proteome changes are slow (minutes-to-hours). This creates a temporal lag in cellular responses.

Common Mistakes

Common mistake Mistake 1: "More DNA = More Genes = More Complexity"

The wrong intuition: "Humans have 3.2 Gb genome, E. coli has 4.6 Mb, so humans must have ~700× more genes." Why it feels right: Bigger genome → more information capacity → more genes (linear relationship assumed).

The reality:

Humans: 3.2 Gb → ~20,000 protein-coding genes (0.006 genes/kb)
E. coli: 4.6 Mb → ~4,300 genes (0.93 genes/kb)
Ratio: humans have only ~5× more genes, not 700×

WHY the paradox? The C-value paradox: genome size doesn't correlate with organism complexity. Humans have vast non-coding regions (introns, repetitive DNA). E. coli is gene-dense (~87% coding).

The fix: Complexity comes from:

Alternative splicing (one gene → many proteins)
Regulatory complexity (gene networks)
Post-translational modifications
Non-coding RNA genes (regulatory transcripts not counted as protein-coding genes)

Human proteome (very large) >> E. coli proteome (~4,000 species), even though gene count is similar.

Common mistake Mistake 2: "Transcriptome = Proteome"

The wrong intuition: "If I sequence all mRNA (transcriptome), I know all proteins (proteome)." Why it feels right: Central dogma says DNA → RNA → protein, so RNA directly predicts protein.

The reality: Correlation between transcript levels and protein levels is only ~0.4-0.6 (moderate).

WHY the disconnect?

Translational control: Some mRNAs are sequestered and not translated (e.g., maternal mRNAs in eggs)
Protein stability: A short-lived protein (half-life 10 min) won't accumulate even if its mRNA is abundant
Post-translational modifications: One mRNA → several protein forms (phosphorylated, glycosylated, etc.)

Example:

β-globin mRNA in reticulocytes: high copy number
β-globin protein: accumulates to very high levels
WHY? High translation rate + very stable protein

Conversely:

c-Myc mRNA in cancer cells: elevated
c-Myc protein: much less amplified relative to mRNA
WHY? Rapid protein degradation (half-life ~20 min)

The fix: Measure proteins directly (mass spectrometry) if you need accurate proteome data. Transcriptome is a proxy, not a mirror.

Active Recall Practice

Recall Feynman Technique: Explain to a 12-year-old

Imagine your body is made of trillions of tiny factories (cells). Each factory has three important things:

The genome is like a giant recipe book stored in the factory's vault. It has ALL the recipes (instructions) for making every tool and machine the factory could ever need. Every factory in your body has the same complete recipe book (that's why every cell has the same DNA).
The transcriptome is like photocopies of recipes that workers are actively using RIGHT NOW. The factory doesn't open the vault book directly—instead, workers photocopy the recipes they need today. A brain factory photocopies recipes for "thinking molecules." A stomach factory photocopies recipes for "digestion molecules." Same recipe book, different photocopies!
The proteome is the actual tools and machines built from those recipes. These are the physical things doing work: scissors cutting, hammers pounding, conveyor belts moving. One recipe (photocopy) can be used to build the same tool 100 times, or workers might modify the tool to make different versions (like adding a handle to a hammer).

The key insight: The recipe book never changes, but which recipes you photocopy and which tools you build changes every second based on what the factory needs to do. That's why your muscle cells look totally different from your brain cells—same recipe book, totally different choices of what to build!

Connections

Central Dogma of Molecular Biology - the flow from genome → transcriptome → proteome
Gene Expression Regulation - how cells choose which genes to transcribe
Alternative Splicing - mechanism that expands proteome beyond gene count
Post-Translational Modifications - further diversifies the proteome
Non-coding RNA - functional transcripts from non-protein-coding genes
RNA-Seq Technology - method for measuring transcriptome
Mass Spectrometry Proteomics - method for measuring proteome
Systems Biology - integrates genome, transcriptome, proteome data
C-Value Paradox - genome size doesn't predict organism complexity
Single-Cell Genomics - reveals transcriptome/proteome heterogeneity within tissues

Diagram

#flashcards/biology

What is the genome? :: The complete set of genetic material (DNA) in an organism, including protein-coding genes, non-coding RNA genes, and all non-coding/regulatory sequences

What is the transcriptome? :: The complete set of RNA transcripts (mRNA, rRNA, tRNA, non-coding RNA) in a cell at a specific time

Do only protein-coding genes produce functional transcripts?

No — non-coding RNA genes (rRNA, tRNA, miRNA, lncRNA) also produce functional transcripts

What is the proteome?

The entire complement of proteins expressed in a cell, tissue, or organism at a given time, including all isoforms and modifications

Why is the proteome larger than the number of genes?

Alternative splicing creates multiple mRNA isoforms from one gene, and post-translational modifications create additional protein variants

Is the genome dynamic or static?

Static - it remains mostly the same in all cells of an organism (except mutations)

Are the transcriptome and proteome dynamic or static?

Dynamic - they change based on cell type, developmental stage, environmental conditions, and time

What is the C-value paradox?

The observation that genome size does not correlate with organism complexity; humans have only ~5× more genes than E. coli despite having ~700× more DNA

Why doesn't the transcriptome perfectly predict the proteome?

Because of translational control, differences in protein stability, post-translational modifications, and variable translation rates

Which changes faster after an environmental stimulus: transcriptome or proteome?

Transcriptome (minutes) because mRNA is rapidly made and degraded; proteome changes more slowly because proteins are generally more stable

What heat-shock chaperones does E. coli use, and what activates them?

DnaK (Hsp70-family) and GroEL (Hsp60-family), activated by the alternative sigma factor σ³² (RpoH) upon heat stress

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, is concept ko samajhna hai toh ek factory ki tarah socho. Genome matlab poori blueprint library — saara DNA jo cell ke andar hai, complete instruction set. Yeh cheez static hoti hai, matlab tumhare body ke har cell mein — chahe neuron ho ya liver cell — genome bilkul same rehta hai. Transcriptome matlab wo active work orders jo abhi copy ho rahe hain, yaani sab RNA transcripts jo cell abhi bana rahi hai. Aur proteome matlab wo actual workforce aur machinery — saare proteins jo un orders se banate hain. Yeh Central Dogma ke teen snapshots hain, ek hi information flow ke teen alag stages par.

Ab core intuition yeh hai ki genome fixed hai par transcriptome aur proteome dynamic hain — yeh badalte rehte hain cell type, developmental stage aur environment ke hisaab se. Isliye ek neuron aur liver cell ka genome same hone ke baad bhi, unke proteomes ekdum alag hote hain, kyunki har cell apne kaam ke hisaab se alag genes ko "on" karta hai. Aur ek interesting baat — ek gene se ek se zyada proteins ban sakte hain alternative splicing aur post-translational modifications ki wajah se. Isiliye humare paas sirf ~20,000 protein-coding genes hain, par proteome mein lakhon distinct protein forms nikal aate hain.

Yeh why-it-matters isliye important hai kyunki genomics, systems biology aur personalized medicine ka poora foundation inhi teen inventories par khada hai. Jab tum samajh jaate ho ki information kaise DNA se RNA se protein tak flow karti hai aur kahan pe variation aati hai, tab tumhe pata chalta hai ki bimariyan kaise hoti hain aur treatment kaise design karein. Mathematical relationship bhi bas yahi batata hai ki genome ka kitna hissa actually genes hai (~17% genic, par sirf ~1.5% exon jo protein banata hai) — baaki regulatory aur repetitive DNA hai. Toh yeh teen terms yaad rakhna future ke saare molecular biology topics ki neenv hai.

Test yourself — Genomics

Connections