Biotechnology
Reading the genome: genes, their switches and the non-coding 98 per cent
Builds on: DNA as the code of life (the central dogma) and Stem cells (iPS cells).
In a nutshell
Reading the whole human genome showed that genes are only about 1.5 to 2 per cent of it, and that about 20,000 genes produce many times that number of proteins. What matters is not only which genes you have but which are switched on, when, and how much: gene expression. Much of the "junk" turns out to be switches and regulators, and medicine is learning to act on them.
- Before transcription: switches on the DNA decide which recipes a cell reads.
- After transcription: the RNA copy is cut and joined (splicing), so one gene can give several proteins.
- Before translation: small RNAs can silence or dim the message.
Same genome in every cellDifferent genes expressedDifferent cells
A hair cell and a nail cell in your body carry exactly the same 3.2 billion letters of DNA. Both make a tough protein called keratin, yet hair and nail are different, and the hair on your head grows long while that on your arm stays short. If the instructions are identical, where does the difference come from?
This topic zooms in on the central dogma. It is the most demanding part of the chapter, and it repays the effort: RNA interference, mRNA therapies, gene silencing sprays and epigenetic editing all appear in the news, and all make sense only once you see where in the chain they act.
What reading the whole genome revealed
Before the Human Genome Project, the natural assumption was simple: a gene is a recipe, and one recipe gives one dish. One gene, one protein. Reading the genome letter by letter (draft 2001, finished 2003) overturned this in two ways.
- Genes are a small part of the book. Humans have only about 20,000 protein-coding genes, fewer than many expected and about as many as a tiny roundworm has. Together they make up about 1.5 to 2 per cent of the genome.
- One gene, many proteins. The human body makes far more kinds of protein than it has genes; estimates run to several times as many. So the same gene must be used to make different proteins in different cells.
The conclusion was that the expression of genes, which genes are read, in which cells, when and in what amount, matters as much as the genes themselves. Think of a chef given a recipe: before cooking he needs to know for how many people, and whether to make the dish plain or rich. The instructions that answer those questions are also written in the genome, mostly in the part that does not code for proteins.
Switching genes on and off
Every cell has the whole book, but a hair cell must read only the hair recipes. Picture the DNA as a long railway track along which the RNA polymerase (the photocopier) travels. At some stretches the signal is green and it starts copying; at others it is red and it stops.
- Promoters are stretches of DNA just before a gene that act as the starting signal: the polymerase attaches there and begins copying.
- Transcription factors are proteins that bind near genes and switch them on or off. Changing them can change a cell's identity. The four factors Shinya Yamanaka used to turn adult cells into iPS cells (see Stem cells) are transcription factors of exactly this kind.
- Packaging. DNA is wound around spool-like proteins called histones, like thread on beads. Where the winding is tight, the photocopier cannot get in and the genes stay silent; where it is loose, they can be read.
- Chemical tags. Small chemical groups attached to the DNA or to histones change how tightly it is packed. The best known is the methyl group: DNA methylation near a gene generally switches it off.
The study of these controls, which change what a gene does without changing its letters, is epigenetics ("above the genes"). Epigenetic marks differ from cell to cell, which is why the same genome makes a neuron in one place and a skin cell in another. They also change with age and can respond to diet, stress or toxins, and many are copied when a cell divides. Because these marks are reversible, researchers are now testing epigenetic editing: switching a faulty gene off or a silent one on without cutting the DNA.
Cutting and joining the copy: splicing
The first RNA copy of a gene is not yet the final message. Within most genes, the protein-coding pieces are interrupted by stretches that do not code. The coding pieces are called exons (they are expressed); the stretches in between are introns (they intervene). The first copy, the pre-mRNA, contains both. Before it leaves the nucleus, the introns are cut out and the exons joined together. This is RNA splicing, and the result is the mature mRNA.
Splicing is not always done the same way. A cell can keep some exons and skip others, joining them in different combinations. This alternative splicing is the main reason about 20,000 genes can produce many more proteins, and part of the answer to the hair-and-nail puzzle. Errors in splicing cause disease. In spinal muscular atrophy, an inherited disease that destroys the nerves controlling muscles, a backup gene is spliced wrongly; one of the approved treatments is a short synthetic strand that corrects the splicing.
Where the protein is made
The mature mRNA goes out to the ribosomes. Some ribosomes float free in the cytoplasm; others sit on a network of membranes called the endoplasmic reticulum (ER), giving it a rough appearance. The rough ER makes proteins and checks that they are folded correctly; the smooth ER has no ribosomes and makes fats and breaks down toxins. So in a plant cell, RNA is found in the nucleus, cytoplasm, ribosomes, mitochondria and chloroplasts, and on the rough ER, but not on the smooth ER.
RNA that never becomes protein
Much of the genome is copied into RNA that is never translated. These non-coding RNAs do jobs of their own. Two kinds we have met already: transfer RNA, which brings amino acids, and ribosomal RNA, part of the ribosome. Others regulate the mRNA.
| Non-coding RNA | Size | What it does |
|---|---|---|
| Long non-coding RNA (lncRNA) | More than 200 letters | Helps regulate gene activity and splicing; still poorly understood, so no therapies yet |
| Small interfering RNA (siRNA) | About 20–25 letters | Matches an mRNA exactly, sticks to it and gets it destroyed, so that protein is not made: gene silencing or RNA interference |
| MicroRNA (miRNA) | About 22 letters | Matches an mRNA partly and dims it, reducing how much protein is made; each miRNA fine-tunes many genes |
Both small RNAs work on the message, after transcription and before translation. RNA interference was discovered in worms by Andrew Fire and Craig Mello (Nobel Prize, 2006). MicroRNA was discovered by Victor Ambros and Gary Ruvkun, who received the Nobel Prize in Physiology or Medicine in 2024.
A third tool copies the same idea: an antisense oligonucleotide is a short, man-made single strand designed to pair with a chosen RNA. Depending on its design it can block the RNA, mark it for destruction or, as in spinal muscular atrophy, change how it is spliced.
Using RNA interference
Cells already use small RNAs to silence their own messages. Biotechnology copies the trick by supplying a laboratory-made siRNA that matches a chosen target.
- Crop protection. RNA interference sprays deliver double-stranded RNA that matches an essential gene of a pest or virus. When the insect eats the treated leaf, the RNA silences that gene and the insect is harmed, while the crop's own DNA is unchanged. Plants can also be made resistant to viruses this way. Because it targets one gene sequence, the approach can be very specific to one pest.
- Medicine. The first siRNA drug, patisiran, was approved in the United States in 2018 for a rare inherited nerve disease caused by a faulty protein; it silences the gene so less of the protein is made. In cancer, researchers aim to silence oncogenes, genes whose overactivity drives cancer.
- Diagnosis. Because some microRNAs act as brakes on cell division, abnormal microRNA levels in blood can signal cancer early, and microRNAs are being studied as drug targets.
Where medicine can intervene
Each step of the chain is a possible point of treatment, and the right point depends on the problem. The key question is whether you need to add a protein or remove one.
| Act on | How | Use it when | Example |
|---|---|---|---|
| DNA | Add, remove or edit a gene | A permanent fix is needed; but a change in sex cells passes to children, and one gene may affect several traits | Gene therapy for sickle cell disease (see Gene therapy) |
| mRNA, by supplying it | Inject an mRNA so the body's own ribosomes make a protein for a while | A protein is missing; or the immune system must be shown a pathogen's protein | mRNA vaccines for COVID-19, which carry the recipe for only the virus's spike protein (see Therapeutics and vaccines) |
| mRNA, by silencing it | siRNA, miRNA or antisense strands | An unwanted or faulty protein must be stopped or reduced | RNA interference sprays; patisiran |
| Protein | Conventional drugs that bind to a protein; or supplying the protein itself | Most of medicine today | Most tablets; insulin injections |
This logic answers a common question. RNA interference can help fight cancer (silencing a gene that drives it) and protect crops (silencing a pest's or virus's gene). It cannot replace a missing hormone, because silencing takes a protein away; replacing one requires adding the protein or its mRNA.
Two kinds of medicine
Most drugs are small molecules, chemicals made in a factory that bind to a target protein. Once the patent expires, others can make exact copies, called generics. A growing class are biologics: large molecules such as insulin, antibodies and vaccines, which are made by living cells. Before the 1980s, people with diabetes injected insulin extracted from cows or pigs; today human insulin is made by genetically engineered microbes (see From rDNA to gene editing). Because a biologic is made by living cells, a copy can never be exactly identical; a near-identical version approved after the original's patent expires is called a biosimilar.
The dark genome
A large part of the genome's function is still unknown, so it is sometimes called the dark genome. How much of it is functional is genuinely disputed: some large studies have found chemical activity across most of it, while others argue that a much smaller share does useful work. Two components are worth knowing.
- Transposons, or jumping genes. These are stretches of DNA that can cut or copy themselves and insert at a new place. They were discovered in maize by Barbara McClintock (Nobel Prize, 1983) and make up nearly half of our genome, mostly as inactive remnants. When one jumps into a gene it can disrupt it; over evolution, they have also created new genes and switches. Their cut-and-insert machinery is now being developed as a tool for gene editing (see From rDNA to gene editing).
- Viral fossils. About 8 per cent of our genome comes from retroviruses that infected our ancestors' sex cells long ago and inserted their DNA, which has been inherited ever since. Most are inactive, but some have been put to use: a protein essential for forming the human placenta comes from one of them. A virus such as HIV infects body cells, not eggs or sperm, so an infected person does not pass HIV DNA into the next generation's genome.
How genomes are read today
Genome sequencing means reading the letters in order. Methods differ mainly in how many letters they read at a time and how many pieces they read together.
- Sanger sequencing, used for the Human Genome Project, reads one piece of DNA at a time, a few hundred to about a thousand letters long. Reading 3.2 billion letters this way took over a decade.
- Next-generation sequencing breaks the genome into millions of fragments and reads them all at once, in parallel, like reading many pages of a book together. The pieces are short (short reads), and computers reassemble them.
- Long-read sequencing, such as nanopore sequencing, reads tens of thousands of letters, sometimes far more, in one go. Long reads made the complete telomere-to-telomere genome possible, because they span the repetitive stretches that short reads cannot place.
Whole genome sequencing simply means reading the entire genome of an organism, by whichever method. A human genome can now be read in hours for a few hundred dollars.
Scientists have also begun to write genetic code. In 2019 a US team created Hachimoji DNA (Japanese for "eight letters"), which adds four synthetic letters to A, T, G and C; the work, funded partly by NASA, explores what other chemistries life could use. Others are adding man-made amino acids to the 20 that cells use, to build proteins with new properties.
Where gene expression is controlled
| Stage | Control | What it decides |
|---|---|---|
| DNA, before transcription | Promoters, transcription factors, histone packing, methylation (epigenetics) | Which genes a cell reads at all |
| Pre-mRNA, after transcription | Splicing and alternative splicing (with help from long non-coding RNA) | Which version of the protein is made |
| mRNA, before translation | siRNA (silence), miRNA (dim); antisense in medicine | Whether, and how much, protein is made |
| Protein | Folding, modification, breakdown | Whether the protein works, and for how long |
Recent developments
Genome India: first findings from nearly 10,000 Indian genomes
The Genome India project, launched in 2020 by the Department of Biotechnology and run by more than twenty institutions, collected samples from about 20,000 people and sequenced whole genomes from 10,074 of them, drawn from 85 population groups, 32 tribal and 53 non-tribal. The dataset was released for research in January 2025, and the first findings were published as a Comment in Nature Genetics on 8 April 2025, based on 9,772 individuals after two populations were excluded. About 180 million genetic variants were catalogued, 130 million on the autosomes and 50 million on the X and Y chromosomes. The sequence data sit in the Indian Biological Data Centre at the Regional Centre for Biotechnology, Faridabad.
Why it matters: Most of what medicine knows about which DNA letters matter was learnt from European populations, so a variant that is common in India may be missing from the reference databases a clinical laboratory consults. Closing that gap is the point of the project, and India's endogamous population structure means some variants are confined to single communities, which is exactly what a global database cannot supply. What the catalogue does not yet do is treat anyone: variants become useful only once particular ones are tied to particular conditions and drug responses, and the April publication was a Comment reporting progress rather than a full analysis.
Remember: Genome India - Department of Biotechnology, launched 2020, more than 20 institutions; 10,074 genomes sequenced from 85 population groups (32 tribal, 53 non-tribal); first findings on 9,772 individuals, about 180 million variants (130 million autosomal, 50 million on sex chromosomes); data at the Indian Biological Data Centre, Faridabad. Distinguish it from IndiGen, the CSIR project that sequenced about 1,000 Indian genomes.
Source: The Hindu, April 2025
Last-month notes
- Human genome: about 20,000 protein-coding genes, about 1.5–2% of the DNA; far more proteins than genes.
- Gene expression controls: promoters, transcription factors, histones (tight = silent), DNA methylation (generally silences). Epigenetics: changes in gene activity without changing the DNA letters.
- Exons code and are kept; introns are cut out. Splicing makes mature mRNA; alternative splicing gives several proteins from one gene.
- Rough ER has ribosomes and makes proteins; smooth ER has none.
- Non-coding RNA: tRNA, rRNA, lncRNA (>200 letters), siRNA (silences mRNA: RNA interference), miRNA (dims mRNA). Nobel: RNAi 2006 (Fire, Mello); miRNA 2024 (Ambros, Ruvkun).
- RNAi uses: gene-silencing therapies, cancer research, virus- and pest-resistant crops, RNAi sprays. Not for replacing a missing protein. First siRNA drug: patisiran (US, 2018).
- Small molecules → generics; biologics (insulin, antibodies, vaccines) → biosimilars.
- Transposons (jumping genes; McClintock; nearly half the genome). Endogenous retroviruses: about 8%.
- Sequencing: Sanger (one piece at a time) → next-generation (short reads, massively parallel) → long-read (nanopore). Hachimoji DNA: eight letters (2019).
Statements to watch for
- One gene does not mean one protein.
- Exons are the coding parts; introns are removed.
- Epigenetic changes alter gene activity, not the DNA sequence.
- RNA interference acts on mRNA, not on DNA; it does not change the genome.
- A biosimilar is a near-copy of a biologic, not a chemical imitation of one.
- Methylation of DNA generally switches genes off.
Revise and apply
Activities to check your understanding, not new content. Answer first, then press Check or Show answers. Multiple-choice questions are original, in the UPSC style, not past UPSC questions. A printable workbook with all activities is on the theme page.
Type the missing words, then check.
(a) promoter (b) histones (c) methyl (d) exons (e) introns (f) alternative (g) sirna (h) mirna (i) transposons (j) retroviruses
- The study of changes in gene activity that do not alter the DNA letters is called .
- The first RNA copy of a gene, before splicing, is called .
- The 2024 Nobel Prize in Physiology or Medicine was awarded for the discovery of .
- A near-identical copy of a biologic drug, approved after its patent expires, is a .
- Proteins that bind near genes and switch them on or off are called factors.
1. epigenetics · 2. pre-mRNA · 3. microRNA · 4. biosimilar · 5. transcription
- A pest needs a particular protein to survive
- A patient's body makes too little of a protein
- The immune system must learn a virus's spike protein
- A faulty gene must be fixed permanently
A pest needs a particular protein to survive: Silence the pest's mRNA with RNA interference · A patient's body makes too little of a protein: Supply the protein, or an mRNA that makes it · The immune system must learn a virus's spike protein: An mRNA vaccine · A faulty gene must be fixed permanently: Gene therapy or gene editing
- RNA interference changes the DNA of the organism it is applied to. It destroys or blocks mRNA; the DNA is unchanged.
- Alternative splicing helps explain why humans make more proteins than they have genes.
- Introns are the protein-coding parts of a gene. Exons code for protein; introns are cut out.
- The smooth endoplasmic reticulum carries ribosomes. Ribosomes sit on the rough ER; the smooth ER has none.
- HIV DNA is passed on to the children of an infected person through their genome. HIV infects body cells, not eggs or sperm.
1. False: It destroys or blocks mRNA; the DNA is unchanged. 2. True. 3. False: Exons code for protein; introns are cut out. 4. False: Ribosomes sit on the rough ER; the smooth ER has none. 5. False: HIV infects body cells, not eggs or sperm.
"Farmers may soon spray RNA instead of chemical pesticides." (a) How would such a spray kill a pest? (b) Why might it harm fewer other insects than a chemical pesticide? (c) Does it make the crop genetically modified?
(a) It carries RNA matching an essential gene of the pest; once eaten, it triggers RNA interference that destroys the pest's mRNA, so the protein is not made. (b) It matches one gene sequence, so other species with different sequences are largely unaffected. (c) No; the crop's DNA is not changed.
Consider the following uses of RNA interference (RNAi) technology:
- Developing gene-silencing therapies
- Developing therapies for the treatment of cancer
- Developing hormone replacement therapies
- Producing crop plants resistant to viral pathogens
Which of the above are possible uses?
Answer: (a). RNAi silences a chosen mRNA. That can stop a cancer-driving gene or a virus's gene, but it cannot supply a hormone the body lacks; replacement needs the protein or its mRNA.
What it tests: the difference between removing a protein and adding one.
The production of several different proteins from a single gene is mainly the result of:
Answer: (b). Different cells join the exons of the same pre-mRNA in different combinations. Crossing over shuffles chromosomes in sex cells; it does not create several proteins from one gene within a cell.
What it tests: one gene, many proteins.
Explain why gene expression, and not only the genes a person carries, determines health and disease. Answer in about 50 words.
Points to include: Every cell has the same genome but reads different genes; promoters, transcription factors, methylation and histones decide which genes are read; splicing decides which protein; small RNAs decide how much; faults in any step cause disease (e.g. spinal muscular atrophy, cancer); hence therapies acting on RNA and epigenetic marks.
The next question If any two people share 99.9 per cent of their DNA, which parts of the remaining 0.1 per cent can tell individuals apart, identify a species from a drop of water, or link a population to its ancestors?