Fold Commons

Nobel Timeline

The Nobel Prizes that built structural & molecular biology — from Linus Pauling working out the alpha helix by hand in 1954, to AlphaFold and computational protein design in 2024. Each prize is told at three reading levels, cites its primary sources, and links to the real predicted structure of an anchor protein in our AlphaFold viewer. Nothing is installed; nothing about you is collected.

This is the tool. Every claim links to the Nobel Foundation, the primary paper, or a public database. Prize citations are quoted verbatim; the plain-language stories are Fold Commons editorial (CC BY 4.0). A native version, when it ships, adds Files and offline use; it never gates the web.

Reading level
  1. Chemistry

    Linus Pauling

    “for his research into the nature of the chemical bond and its application to the elucidation of the structure of complex substances”

    Anchor structure: Hemoglobin subunit beta View 3D structure → UniProt P68871 · PDB 2HHB

    Linus Pauling figured out how proteins fold. He showed that they curl into a spiral he called the alpha helix. This shape lets proteins do many jobs in your cells. Pauling also found that sickle cell anemia is caused by a tiny change in one protein. He helped start the field of how molecules cause disease.

    Why it still matters: Today, biologists still describe protein shapes using the alpha helix. The same rules now help computers like AlphaFold predict how new proteins will fold.

    In 1948 Pauling proposed the alpha helix. It is a spiral pattern of the protein backbone, held together by hydrogen bonds. He worked it out without seeing a real protein. He used rules about chemical bonds and angles. Five years later, X-ray work confirmed his model. Pauling also showed that sickle cell disease comes from a single change to the hemoglobin protein. It was the first time anyone linked a disease to a specific molecular defect. The Nobel honoured both lines of work — a model of how proteins fold, and a framework for understanding disease at the molecular level.

    Why it still matters: Modern protein-prediction tools like AlphaFold start from the same backbone geometry Pauling worked out by hand. Sickle cell research has moved from observation to gene therapy — Casgevy, approved in 2023, edits the gene Pauling's electrophoresis pointed at seventy years earlier.

    Pauling's 1948 alpha-helix model predicted a regular structure with ~3.6 residues per turn, stabilised by main-chain hydrogen bonds along the helix axis. He derived the geometry from peptide-bond planarity arguments — partial double-bond character forcing a planar amide — without a single X-ray structure of a protein to guide him. The model was confirmed by Perutz's polarised X-ray diffraction within two years. The same theoretical reasoning produced the beta sheet. Independently, Pauling and Itano (1949) demonstrated that haemoglobin from sickle-cell patients migrated differently on electrophoresis than normal haemoglobin, establishing sickle-cell anaemia as the first 'molecular disease' — a phrase Pauling coined. The 1954 prize cited bonding work broadly, including its application to complex biological substances. Pauling later received an unshared Nobel Peace Prize (1962), making him one of only four individuals to hold two Nobels.

    Why it still matters: The alpha helix remains the most common secondary structure in folded proteins, accounting for roughly a third of all assigned secondary structure in the PDB. Pauling's molecular-disease framing now applies to every Mendelian disorder in OMIM. The β6 sickle-cell mutation is the target of Casgevy (exa-cel), the first CRISPR-Cas9 therapeutic approved by the FDA (December 2023). AlphaFold's transformer-based predictions are evaluated against the secondary-structure assignments Pauling first proposed.

    MS-LS1-3HS-LS1-1

    Sources (5)
  2. Chemistry

    Frederick Sanger

    “for his work on the structure of proteins, especially that of insulin”

    Anchor structure: Insulin View 3D structure → UniProt P01308 · PDB 4INS

    Insulin is the hormone that tells your body how to use sugar. In 1955 Frederick Sanger figured out which building blocks make up insulin, and in what order. This was the first time anyone had read a protein from one end to the other. People had thought proteins were too messy to have an exact recipe. Sanger proved they did.

    Why it still matters: Sanger's method made it possible to read all sorts of molecules. Today, gene tests for cancer use methods that come from his lab.

    By 1955 Sanger had sequenced insulin. It was the first protein ever read residue by residue. Insulin from cows is built from 51 amino acids in two chains, joined by sulphur bridges. Sanger's method cut the protein into small fragments. Each fragment was sequenced, then stitched back into the right order. Before this work, biochemists were not sure whether proteins had a single defined sequence or were mixtures. Sanger proved that every insulin molecule is identical. That meant the gene that codes for it must carry a defined message too.

    Why it still matters: Modern sequencing of any protein or gene rests on Sanger's insulin work. Recombinant human insulin, made in bacteria from a known sequence, replaced animal insulin in the 1980s. AlphaFold predicts insulin's folded shape from the same sequence Sanger first wrote down.

    Sanger's insulin sequence (1951–1955) established that proteins have a unique covalent sequence — a fact that had been contested through the 1940s, when proteins were often described as ill-defined colloids. He fluorodinitrophenylated the N-terminus, hydrolysed under defined conditions, and chromatographed the resulting fragments to identify them. Bovine insulin has two chains (A: 21 residues, B: 30 residues) joined by two interchain disulphides and one intrachain. The discovery that the sequence is defined was decisive evidence for the gene-to-protein information flow then being formalised by Crick. Insulin's sequence variations across species — pig insulin differs from human by a single residue and cow insulin by three — explained why animal-derived insulin had clinical utility but also why immune reactions sometimes occurred. The 1958 prize was for the insulin work; Sanger went on to receive a second Nobel in 1980 for DNA sequencing.

    Why it still matters: Insulin is now sourced almost exclusively from recombinant production — Eli Lilly's Humulin (1982) was the first recombinant protein drug. Sanger sequences underwrite the UniProtKB database (260M+ entries), which AlphaFold uses as its primary input. Sanger's 1980 chain-termination DNA-sequencing method was the workhorse of the Human Genome Project; ONT and Illumina platforms have since moved most production sequencing off it, but Sanger sequencing is still the standard for clinical variant confirmation.

    MS-LS1-3MS-LS1-7HS-LS1-1

    Sources (4)
  3. Chemistry

    Max Perutz · John Kendrew

    “for their studies of the structures of globular proteins”

    Anchor structure: Hemoglobin subunit alpha View 3D structure → UniProt P69905 · PDB 1HHO

    In 1962 Max Perutz and John Kendrew showed the world what a protein looks like from the inside. They used X-rays bouncing off crystals to make 3D maps. Kendrew's protein was myoglobin — it grabs oxygen in your muscles. Perutz's was hemoglobin — it carries oxygen in your blood. After 25 years of work, they finally cracked it.

    Why it still matters: Today scientists know the shape of millions of proteins. AlphaFold uses a computer to predict shapes — but it learned by studying real structures like Perutz's.

    Perutz spent 23 years working out the structure of hemoglobin before he succeeded in 1959. Kendrew solved myoglobin first, in 1958. Both used X-ray crystallography — shining X-rays through a protein crystal and reading where the rays bounced. The challenge was the phase problem: the diffraction pattern alone doesn't tell you a structure. Perutz's breakthrough was attaching heavy mercury atoms to the protein as reference points. Their structures showed that proteins are not random — they fold into specific shapes that match their function, with pockets and tunnels that grip oxygen exactly right.

    Why it still matters: Every modern protein structure in the Protein Data Bank traces its method back to Perutz and Kendrew. The PDB now holds over 220,000 experimental structures, and AlphaFold predicts shapes for nearly every known protein. Hemoglobin remains a textbook system for explaining how a protein's shape determines its job.

    Perutz and Kendrew used X-ray crystallography with multiple isomorphous replacement (MIR) — Perutz's 1953 mercury-labelling trick provided the heavy-atom phase reference that solved the protein-crystallographic phase problem. Kendrew's myoglobin model (1958, 2 Å resolution by 1959) and Perutz's hemoglobin model (1959, 5.5 Å) were the first atomic-resolution structures of any proteins, and they vindicated Pauling's alpha-helix prediction visually. The hemoglobin tetramer (two α and two β chains, P69905 + P68871) revealed the heme-binding pocket that grips O₂ via a histidine-bridged Fe(II), and the quaternary cooperativity Perutz later explained mechanistically as a T↔R state transition. The Cambridge Medical Research Council Laboratory of Molecular Biology was institutionalised around this work and became the world's leading structural biology centre, training Crick, Klug, Brenner, and the next two generations of protein crystallographers.

    Why it still matters: The PDB (founded 1971) has accumulated >220,000 experimentally determined structures, the majority by X-ray crystallography building on Perutz–Kendrew methods. The MRC LMB tradition continues — its alumni founded most of the world's protein crystallography groups. Hemoglobin remains the canonical allosteric system; the T-state R-state framework Perutz proposed in 1970 is still how textbooks teach cooperativity. AlphaFold-Multimer's hemoglobin tetramer predictions reproduce both quaternary states without supervision.

    MS-LS1-3MS-LS3-1HS-LS1-2

    Sources (5)
  4. Physiology or Medicine

    James Watson · Francis Crick · Maurice Wilkins

    “for their discoveries concerning the molecular structure of nucleic acids and its significance for information transfer in living material”

    Anchor structure: Histone H3.1 (DNA-binding nucleosome core) View 3D structure → UniProt P68431 · PDB 1AOI

    DNA is the code that builds you. In 1953 James Watson, Francis Crick, and Maurice Wilkins published a model of DNA's shape — a twisted ladder, the double helix. They used X-ray pictures of DNA taken by Rosalind Franklin, who died before the Nobel. Their model showed how DNA can copy itself when cells divide. It was the start of modern genetics.

    Why it still matters: DNA tests, gene therapies, and personal genome reading all rest on the double-helix idea. Rosalind Franklin's name is now taught alongside Watson, Crick, and Wilkins.

    Watson, Crick, and Wilkins were honoured for the 1953 double-helix model of DNA. Two strands wind around each other. Each rung is formed by a pair of bases — adenine with thymine, cytosine with guanine. The model showed how DNA copies itself: split the two strands, then build a new partner against each. The critical X-ray data came from Rosalind Franklin at King's College London. Franklin died in 1958 of ovarian cancer. The Nobel cannot be awarded after death, so she could not be cited in 1962. Many historians regard the omission as a serious injustice.

    Why it still matters: CRISPR gene editing depends on the base-pairing rules in the double helix. Personalised cancer treatments read tumour DNA to find which drugs will work. Franklin's contribution is now taught alongside the three laureates'.

    The 1953 Watson–Crick model — a B-form double helix with antiparallel strands, 10 base-pairs per 34 Å turn, Watson–Crick base pairing through major and minor grooves — was deduced from Franklin's X-ray fibre diffraction Photo 51 (taken by Raymond Gosling under Franklin's direction) and Chargaff's base-composition rules. Wilkins, working independently in the same King's College department, supplied diffraction data parallel to Franklin's; the political and gender dynamics of the King's group shaped what data reached the Cambridge model-builders. Franklin's contribution went unacknowledged in the original Nature trilogy and could not be retroactively recognised by the Nobel after her death in 1958. The model's immediate implication — that information transfer occurs by strand separation and complementary synthesis — was experimentally proven by Meselson and Stahl (1958) and underlies the entire genome-information paradigm Crick later codified as the central dogma.

    Why it still matters: The Watson–Crick base pairing forms the molecular basis of every gene-editing tool (CRISPR-Cas, prime editing, base editing). Direct-to-consumer genome sequencing is now under $200 per sample; clinical whole-genome sequencing is reimbursed for diagnostic odysseys in pediatric care under most US private payers as of 2025. Franklin's contribution has been formally re-evaluated by historians of science — Maddox (2002) and Cobb & Comfort (Nature 2023) cite primary correspondence showing Franklin was closer to solving the structure independently than Watson's memoir indicated.

    MS-LS3-1MS-LS3-2HS-LS1-1HS-LS3-1

    Sources (4)
  5. Chemistry

    Frederick Sanger · Walter Gilbert · Paul Berg

    “for their contributions concerning the determination of base sequences in nucleic acids (Sanger and Gilbert), and for fundamental studies of the biochemistry of nucleic acids, with particular regard to recombinant DNA (Berg)”

    Anchor structure: DNA polymerase I (E. coli) — Klenow fragment View 3D structure → UniProt P00582 · PDB 1KLN

    In 1977 Frederick Sanger figured out how to read DNA. His method used short chemical roadblocks to mark each letter as DNA copies itself. Walter Gilbert worked out a different way to read DNA at the same time. Paul Berg learned how to splice DNA from different sources. Together their work made it possible to read and rewrite genes.

    Why it still matters: Today machines can read your whole genome in a day. Genentech still makes recombinant insulin using methods that come from Berg's lab.

    Sanger's 1977 chain-termination method reads DNA by letting it copy itself in four test tubes. Each tube contains all four normal DNA bases plus a small amount of a chain-terminating 'roadblock' version of one of them. The lengths of the resulting fragments reveal the sequence. Gilbert's chemical-cleavage method reached the same goal by a different route. Paul Berg's recombinant DNA work built on this line. He showed in 1972 how to splice DNA from different species and read out the result. These tools became the foundation of all modern biotechnology — including Genentech's human insulin in 1982.

    Why it still matters: Whole-genome sequencing now costs under $200 and is used in cancer treatment, newborn screening, and crop breeding. Recombinant techniques produce most modern biological drugs, including the COVID-19 mRNA vaccines.

    Sanger's 1977 dideoxynucleotide chain-termination method (Sanger, Nicklen & Coulson, PNAS) supplanted Gilbert–Maxam chemical sequencing within a decade and remained the workhorse of the Human Genome Project. The method exploits DNA polymerase incorporating a 2',3'-dideoxynucleotide that lacks the 3'-OH needed for chain extension, producing a fragment ladder readable on a denaturing PAGE gel. Berg's half of the prize (recombinant DNA, 1972) introduced restriction enzymes plus DNA ligase as a generalisable cloning toolkit. The recombinant capability and sequencing capability together were the technical preconditions for the genomic era. Automated Sanger sequencing (Applied Biosystems) drove the public Human Genome Project; Celera (Venter) used shotgun Sanger with whole-genome assembly. Next-generation sequencing (Illumina sequencing-by-synthesis, ONT nanopore) displaced Sanger for production work in the 2010s, but Sanger sequencing remains the regulatory standard for clinical variant confirmation.

    Why it still matters: Production sequencing has moved to Illumina short-read (2 × 150 bp HiSeq/NextSeq descendants) and ONT/PacBio long-read platforms; Sanger remains the regulatory gold standard for clinical variant confirmation. Recombinant biopharma production now accounts for ~50% of the global drug market by value (EvaluatePharma 2024). The mRNA-vaccine platform (Pfizer–BioNTech and Moderna) depends entirely on the recombinant-DNA infrastructure Berg seeded. CRISPR-Cas9 therapeutic targets are identified by NGS, then validated by Sanger.

    MS-LS3-1HS-LS1-1HS-LS3-1

    Sources (4)
  6. Chemistry

    Aaron Klug

    “for his development of crystallographic electron microscopy and his structural elucidation of biologically important nucleic acid–protein complexes”

    Anchor structure: Tobacco mosaic virus coat protein View 3D structure → UniProt P03304 · PDB 2OM3

    Aaron Klug found a way to use electron microscopes to see big virus and DNA-protein bundles. He worked with Rosalind Franklin on TMV, a plant virus. He finished her TMV work after she died. His method merges many fuzzy 2D pictures into one sharp 3D one. Decades later, this same idea made cryo-EM possible.

    Why it still matters: Cryo-EM machines today win Nobel prizes (2017) and produce sharp pictures of viruses, ribosomes, and gene-editing tools.

    Klug invented crystallographic electron microscopy in the 1960s. He took electron-micrograph images of biological assemblies from many angles. He then combined them with math into a 3D structure. His main objects of study were tobacco mosaic virus (TMV) and chromatin — DNA wound around histone proteins. Klug also continued Rosalind Franklin's work on TMV after her death in 1958. His math methods underlie modern cryo-electron microscopy and tomography. Klug spent his career at the MRC Laboratory of Molecular Biology in Cambridge. It was the world's leading centre for protein-structure work, built by Perutz.

    Why it still matters: Klug's methods are the foundation of cryo-electron microscopy, which won its own Nobel in 2017. Cryo-EM produced atomic structures of SARS-CoV-2's spike protein in early 2020, which directly informed vaccine design.

    Klug's crystallographic electron microscopy combined real-space TEM images of negatively-stained specimens with Fourier-space averaging across many particle views, achieving 3D reconstructions an order of magnitude beyond what single micrographs allowed. The TMV reconstruction (1968) and subsequent nucleosome work established the helical and icosahedral symmetry frameworks that single-particle cryo-EM still uses. Klug's chromatin work — co-discovery of the nucleosome with Roger Kornberg in 1974 — established that DNA wraps ~146 bp around a histone octamer in about 1.65 superhelical turns. His RNA work with crystallographer Brian Clark on phenylalanine tRNA gave the first tRNA crystal structure (1974), revealing how the universal cloverleaf folds into a three-dimensional L-shape. Klug succeeded Perutz as MRC LMB director (1986–1996) and supported the Royal Society's programmes for women in science.

    Why it still matters: Klug's 3D reconstruction algorithms (back-projection, real-space averaging) are the conceptual ancestor of modern single-particle reconstruction in RELION/cryoSPARC. The 2017 Nobel to Frank, Henderson, Dubochet recognised the experimental advances (vitrification, DDD detectors, computational classification) that pushed cryo-EM into atomic-resolution territory. Klug's chromatin structure remains the basis of contemporary nucleosome research — including cryo-EM structures of chromatin remodellers and pioneer transcription factors.

    MS-LS1-2MS-LS1-3HS-LS1-1

    Sources (4)
  7. Chemistry

    Johann Deisenhofer · Robert Huber · Hartmut Michel

    “for the determination of the three-dimensional structure of a photosynthetic reaction centre”

    Anchor structure: Reaction centre protein L chain (Blastochloris viridis) View 3D structure → UniProt P06009 · PDB 1PRC

    In 1985 these three German scientists made the first 3D map of a membrane protein. The protein was part of a bacterium that turns sunlight into energy. Membrane proteins were thought to be too hard to crystallise. Hartmut Michel found a way to grow crystals of one. Then Deisenhofer and Huber worked out its 3D shape. The map showed how cells trap a photon and shuttle the energy into chemistry.

    Why it still matters: Today, scientists know the structures of channels and pumps in your own cells, like the ones that fire nerves and beat your heart.

    Most proteins floating in water are easy to crystallise. Membrane proteins, set into fatty cell membranes, are not — they fall apart in the wrong solvent. Hartmut Michel cracked this in 1982. He added small detergent molecules that hold the protein in a mimic of its real membrane. The protein he picked, the reaction centre of a purple bacterium, captures photons. It uses their energy to push electrons across the membrane. With the crystal in hand, Deisenhofer and Huber solved the X-ray structure in 1985. It was the first membrane-protein structure ever determined.

    Why it still matters: GPCRs — the receptors that bind many drugs, including beta-blockers and asthma medicines — are membrane proteins whose structures were solved using methods Michel pioneered. The 2012 Lefkowitz–Kobilka Nobel built directly on the 1988 methodological foundation.

    Michel's detergent-mediated crystallisation of the Blastochloris viridis reaction centre (using N,N-dimethyldodecylamine N-oxide, LDAO, with small amphiphile heptane-1,2,3-triol) produced the first membrane-protein crystals diffracting to better than 3 Å. Deisenhofer and Huber solved the structure (Nature 1985, 1986, 3 Å) revealing the L, M, H subunits — four bacteriochlorophylls, two bacteriopheophytins, a non-haem iron, and two quinones arranged in a quasi-twofold-symmetric electron-transfer chain. The structure mechanistically explained primary charge separation in photosynthesis: a special-pair bacteriochlorophyll dimer absorbs light, an electron tunnels to bacteriopheophytin (~3 ps), then to the primary quinone QA (~200 ps), then to the secondary quinone QB (~100 μs). The structural-functional correspondence vindicated decades of spectroscopic inference. The work established membrane-protein crystallography as feasible and set the methodological template for all subsequent membrane-protein structures, including the Lefkowitz–Kobilka GPCR structures honoured in 2012.

    Why it still matters: Membrane-protein crystallography expanded from one structure in 1985 to >1,500 unique entries in the PDB by 2024. LCP (lipidic cubic phase) crystallisation, introduced by Caffrey & Cherezov in the 1990s and 2000s, enabled GPCR structural determination (β2-AR by Kobilka, 2007). Cryo-EM has displaced crystallography for many membrane-protein targets — the 2017 cryo-EM Nobel recognised the methodological pivot — but the conceptual debt to Michel–Deisenhofer–Huber remains. Photosystem II structures (Umena, Kawakami, Shen, Kamiya 2011, 1.9 Å) extend the framework to oxygen-evolving photosynthesis.

    MS-LS1-2MS-LS1-7HS-LS1-2

    Sources (4)
  8. Chemistry

    Roger Kornberg

    “for his studies of the molecular basis of eukaryotic transcription”

    Anchor structure: RNA polymerase II subunit RPB1 (yeast) View 3D structure → UniProt P04050 · PDB 1I50

    To make a protein, your cell first copies its DNA recipe into a message called mRNA. The copier is a big machine called RNA polymerase II. In 2006 Roger Kornberg figured out its shape, frozen in the act of reading DNA. He showed how the machine grabs DNA, opens it like a book, and reads one strand. His father had won a Nobel for related work.

    Why it still matters: Cancer drugs sometimes target this machine. mRNA vaccines also use copies of RNA — a kind of message your cells normally make in this way.

    RNA polymerase II is the molecular machine that transcribes most genes in eukaryotic cells. Yours, baker's yeast, and a tomato's all share the same enzyme. Kornberg's lab spent two decades crystallising it. His group published structures from 2001 to 2006. They captured the polymerase in several states — starting, copying, and stalled at DNA damage. The structures showed how the enzyme separates DNA strands. They showed how it threads one strand through its active site, and builds a matching RNA copy. Kornberg's father Arthur won the 1959 Medicine Nobel for studying a DNA polymerase. That makes them one of only seven parent–child Nobel pairs.

    Why it still matters: Anti-cancer drugs like THZ1 inhibit a kinase that phosphorylates this polymerase's tail. mRNA vaccines (Pfizer–BioNTech, Moderna) deliver synthetic mRNA that bypasses transcription, exploiting the next step of the same gene-expression pathway.

    Kornberg's lab determined a series of yeast Pol II crystal structures (Cramer/Bushnell/Kornberg 2001, 2.8 Å core enzyme; 2001 elongation complex; 2003 transcription bubble) that mechanistically defined the transcription cycle. The 12-subunit enzyme threads template DNA through a positively-charged cleft, separates the two strands at a bridge helix, and uses an Mg²⁺-dependent two-metal mechanism (analogous to DNA polymerases) to add ribonucleotides. The Rpb1 C-terminal domain heptad repeat (YSPTSPS × 26 in yeast, × 52 in vertebrates) is phosphorylated at Ser2, Ser5, and Ser7 in a 'CTD code' that recruits processing and chromatin factors at distinct stages of transcription. The Pol II structures unified decades of biochemical work and explained the actions of inhibitors (α-amanitin from death cap mushrooms, which blocks translocation). Cryo-EM structures of the full 47-subunit pre-initiation complex (Cramer 2016, Murakami 2013) extended the framework to initiation.

    Why it still matters: Pol II CDK7–CDK9–CDK12 kinase inhibitors (THZ1, NVP-2, SY-5609) are in clinical trials for transcription-dependent cancers (NUT carcinoma, MYC-driven leukaemias). Cryo-EM structures of Pol II elongation complexes with DSIF, NELF, SPT5/6, PAF1 (Cramer 2018–2022) have mapped the full transcription-elongation regulatory complement. mRNA vaccine production uses in vitro transcription with T7 RNA polymerase — a bacteriophage enzyme, simpler than the eukaryotic machine — but the read-out depends on host Pol II for any encoded gene editor.

    MS-LS1-2MS-LS3-1HS-LS1-1

    Sources (4)
  9. Chemistry

    Venkatraman Ramakrishnan · Thomas Steitz · Ada Yonath

    “for studies of the structure and function of the ribosome”

    Anchor structure: 30S ribosomal protein S2 (E. coli) View 3D structure → UniProt P0A7V0 · PDB 1FFK

    Inside every cell, ribosomes read mRNA and build proteins one piece at a time. They are big machines made of RNA and many small proteins. In 2009 three labs published the first atomic-scale maps of a ribosome. Many drugs that fight germs work by jamming the bacterial ribosome. Now drug designers can see exactly where each drug fits.

    Why it still matters: New antibiotics for tough infections are designed using ribosome structures. mRNA vaccines work by handing your ribosomes a temporary instruction sheet.

    The ribosome is one of the most ancient cellular machines. It makes every protein in every living cell. Ada Yonath spent thirty years figuring out how to crystallise ribosomes. Many people thought it could not be done. She finally succeeded with ribosomes from extremophiles. Ramakrishnan solved the small (30S) subunit, and Steitz solved the large (50S) subunit. Their structures showed that the catalytic core is not protein at all — it is RNA. Ribosomes are RNA enzymes. That backs the idea that early life ran on RNA before proteins took over. The structures also showed where dozens of antibiotics bind.

    Why it still matters: Multi-drug-resistant bacteria are a growing health threat. New ribosome-targeting antibiotics — lefamulin (2019), eravacycline (2018) — were designed using the structures these Nobel laureates produced. mRNA vaccines work because your own ribosomes happily translate the synthetic mRNA.

    Yonath's two decades of crystallisation work using Haloarcula marismortui and Thermus thermophilus ribosomes (overcoming the conformational heterogeneity and solubility issues that had defeated others) produced the diffracting crystals that enabled the 2000–2001 Nature/Science structures. Ramakrishnan (30S, 3.0 Å), Steitz (50S, 2.4 Å) and Yonath (70S, ~5 Å) together gave the field a complete atomic-resolution view. The 50S structure (Nissen, Hansen, Ban, Moore, Steitz, Science 2000) demonstrated that the peptidyl-transferase centre is RNA-catalysed — the closest amino acids are >18 Å from the reaction centre — confirming the ribozyme nature of the ribosome and supporting the RNA-world hypothesis. The structures mapped binding sites for >30 clinically used antibiotics: aminoglycosides bind the 30S decoding centre (Ramakrishnan), macrolides and oxazolidinones bind 50S exit tunnel/PTC (Steitz). Structure-guided antibiotic development against multi-drug-resistant pathogens (MDR-TB, ESKAPE organisms) now uses these structures directly.

    Why it still matters: Ribosome-targeting antibacterials approved 2019–2024 include lefamulin (pleuromutilin, Nabriva), eravacycline (tetracycline, Tetraphase), and zoliflodacin (a bacterial DNA gyrase inhibitor, mechanistically distinct from the ribosome-targeting classes). Cryo-EM structures of stalled ribosomes (tigecycline-bound, ApiDef-rescued) are mapping translation-quality-control biology. SARS-CoV-2 mRNA vaccine translation by host ribosomes was modelled atomically by Frank's lab (2021). The peptidyl-transferase activity demonstrated by Steitz (2000) is now taught in every undergraduate biochemistry course as proof that catalytic RNAs predate catalytic proteins.

    MS-LS1-2HS-LS1-1

    Sources (4)
  10. Chemistry

    Robert Lefkowitz · Brian Kobilka

    “for studies of G-protein-coupled receptors”

    Anchor structure: Beta-2 adrenergic receptor View 3D structure → UniProt P07550 · PDB 2RH1

    GPCRs are a family of sensors on the surface of your cells. They sense things like adrenaline, light, and taste. About a third of all medicines work by hitting a GPCR. Robert Lefkowitz spent decades proving these sensors exist as real molecules. Brian Kobilka cloned the gene and crystallised one in 2007 — caught it bending while a drug bound.

    Why it still matters: GLP-1 drugs like Ozempic and Mounjaro target GPCRs. Researchers can see exactly how those drugs grip the receptor.

    GPCRs are seven-helix membrane proteins. They turn outside signals into inside-the-cell responses. Lefkowitz spent the 1970s and 1980s using radioactive tracers to study what was then a hypothetical receptor. He then cloned the gene for the β-adrenergic receptor in 1986. Kobilka worked in Lefkowitz's lab and drove the protein toward a crystal structure. He solved the first GPCR structure (β2-AR) in 2007. He fused it with T4 lysozyme to hold it steady. The structure showed that the receptor shifts shape when it turns on. About 34% of FDA-approved drugs target GPCRs.

    Why it still matters: GLP-1 receptor agonists (Ozempic, Wegovy, Mounjaro) bind GPCRs — their billion-dollar success rests on structural biology rooted in this Nobel. Cryo-EM has solved hundreds of GPCR structures since 2017, opening new drug-design directions.

    Lefkowitz's biochemical isolation of β-adrenergic, α-adrenergic, and muscarinic receptors through the 1970s established the GPCR superfamily as a unified seven-transmembrane signalling architecture. Kobilka's crystallographic work — beginning with the rhodopsin-template-based β2-AR-T4-lysozyme fusion construct (Rosenbaum, Cherezov, Hanson, Kobilka, Science 2007, 2.4 Å) — required LCP crystallisation (Caffrey/Cherezov) and stabilising antibody-fragment plus inverse-agonist binding. The 2011 active-state β2-AR–Gs complex (Rasmussen, Kobilka et al., Nature) captured the first GPCR–G-protein interface, revealing the TM6 outward movement that opens the intracellular G-protein binding pocket. Cryo-EM has since produced hundreds of GPCR–G-protein and GPCR–arrestin complex structures. Approximately 34% of FDA-approved small-molecule drugs target one of the ~800 human GPCRs (Hauser, Attwood, Babu et al., Nature Rev Drug Discov 2017). Biased agonism — separating G-protein from arrestin signalling — is now a major therapeutic strategy (oliceridine for pain, TRV027 cardiovascular).

    Why it still matters: GPCR cryo-EM structure determination has exploded — >300 unique GPCR–G/arrestin complex structures since 2017 versus <30 crystal structures over 2007–2016. Tirzepatide (Mounjaro/Zepbound, GLP-1+GIP dual agonist, Eli Lilly) and semaglutide (Ozempic/Wegovy/Rybelsus, Novo Nordisk) generated >$40B combined revenue in 2024. AlphaFold-Multimer GPCR–G-protein complex predictions are now competitive with cryo-EM for ligand-free states. Biased-agonism programmes (TRV-027, oliceridine) provide the proof of concept that signalling-pathway selectivity is druggable.

    MS-LS1-3HS-LS1-1HS-LS1-2

    Sources (4)
  11. Chemistry

    Jacques Dubochet · Joachim Frank · Richard Henderson

    “for developing cryo-electron microscopy for the high-resolution structure determination of biomolecules in solution”

    Anchor structure: TRPV1 capsaicin receptor (human) View 3D structure → UniProt Q8NER1 · PDB 3J5P

    Cryo-EM means freezing samples in a thin layer of ice and looking at them in an electron microscope. Dubochet found a way to freeze proteins so fast that no ice crystals form. Henderson showed that this kind of microscope can see single atoms. Frank wrote code to combine many fuzzy 2D snapshots into a sharp 3D shape. Together they made cryo-EM the leading way to see large proteins today.

    Why it still matters: Cryo-EM made the COVID vaccines possible. Pictures of the spike protein helped scientists design a vaccine in less than a year.

    Cryo-electron microscopy lets researchers see the shape of a protein without growing crystals of it. Dubochet's 1981 method freezes water so fast that it forms a glass instead of ice. The proteins inside stay fluid-like and keep their real shape. Henderson's 1990 work on bacteriorhodopsin proved that electron microscopy can in theory reach atomic detail. Frank's image code (from 1981 onward) combines tens of thousands of low-contrast 2D particle images into one 3D map. The 2010s 'resolution revolution' came from better detectors and better software. Together they pushed cryo-EM into routine atomic work. It quickly took over from X-ray crystallography for many large complexes.

    Why it still matters: Cryo-EM structures of the SARS-CoV-2 spike protein were published in February 2020, weeks after the virus was sequenced, and informed vaccine design directly. Today most new structures of large molecular machines come from cryo-EM rather than X-ray crystallography.

    Dubochet's 1981 plunge-freezing of thin water films in liquid ethane produced vitreous ice that preserved biological structure without dehydration or staining. Henderson's bacteriorhodopsin structure (Henderson, Baldwin, Ceska, Zemlin, Beckmann, Downing, JMB 1990, 3.5 Å) was the first atomic-resolution cryo-EM membrane-protein structure — solved from electron crystallography of 2D crystals. Frank's single-particle reconstruction theory and the SPIDER software package (1981–present) defined the algorithmic basis: classify particles by view orientation, average within classes, back-project into 3D, iteratively refine. The 2013 'resolution revolution' was the convergence of direct-electron-detector (DDD) cameras (Gatan K2, FEI Falcon) with maximum-likelihood reconstruction (RELION, cryoSPARC) and Bayesian beam-induced motion correction. Resolution standards collapsed from ~10 Å (2012) to routine 2–3 Å (2018+) and sub-2 Å on favourable targets (apoferritin, GABA-A receptor, ribosome). Cryo-EM displaced X-ray as the dominant method for large membrane proteins and complexes by 2020.

    Why it still matters: The PDB now adds more cryo-EM structures per year than X-ray structures (since 2022). Sub-2 Å structures of GABA-A, ATP synthase, and apoferritin demonstrate near-X-ray atomic detail. Cryo-electron tomography (cryo-ET) is extending the framework to in-cell structural biology — sub-tomogram averaging of ribosomes inside intact cells has reached <5 Å on favourable targets. Time-resolved cryo-EM with millisecond mixers (spotiton, cryoSPARC's microfluidics integration) is opening kinetics. AlphaFold-Multimer is increasingly the first-pass model for ambiguous cryo-EM density.

    MS-LS1-3HS-LS1-2

    Sources (4)
  12. Chemistry

    David Baker · Demis Hassabis · John Jumper

    “for computational protein design (Baker), and for protein structure prediction (Hassabis and Jumper)”

    Anchor structure: Tumour suppressor p53 View 3D structure → UniProt P04637 · PDB 1TUP

    AlphaFold can predict the shape of almost any protein in minutes. Demis Hassabis and John Jumper led the team at DeepMind that built it. These shapes used to take years to figure out, one at a time. David Baker shared the Nobel for going the other direction. He designs brand-new proteins on a computer, then tests them in the lab to see if they fold.

    Why it still matters: Researchers use AlphaFold every day. New designed proteins might become tomorrow's medicines.

    AlphaFold-2 predicts protein shapes from sequences. Its accuracy can match the best experimental methods. Jumper's team at DeepMind published the model in 2021. It is a deep-learning system trained on the ~200,000 structures in the Protein Data Bank. AlphaFold has now predicted shapes for 200 million proteins. All of them are free to download. David Baker's lab at the University of Washington runs the opposite direction. His group designs a sequence for a desired shape and function, then tests it in the lab. Baker's RFdiffusion (2023) and other tools have produced new enzymes, vaccine candidates, and binders against drug-resistant cancers.

    Why it still matters: AlphaFold's predictions are now standard in drug discovery — Isomorphic Labs, founded by Hassabis, is running clinical-stage programmes entirely informed by AlphaFold structures. Baker-lab-designed proteins targeting cancer, COVID, and rare diseases are in early clinical trials.

    AlphaFold-2 (Jumper, Evans, Pritzel, Hassabis et al., Nature 2021) achieved median Cα RMSD <1 Å on CASP14 targets — competitive with experimental methods for monomeric proteins — via an end-to-end transformer architecture using MSA-conditioned attention (Evoformer) and structure-aware iterative refinement (structure module). Per-residue pLDDT confidence scores enabled honest reporting of low-confidence regions, a practice the field has now adopted broadly. The AlphaFold DB (Tunyasuvunakool et al., Nature 2021) contains ~200M predictions covering essentially all sequenced proteins. AlphaFold 3 (Abramson et al., Nature 2024) extended to nucleic acids, ligands, and post-translational modifications. Baker's lab pioneered the opposite direction: Rosetta de novo design (1990s–2000s), ProteinMPNN (Dauparas et al., Science 2022) for sequence-conditioned design, and RFdiffusion (Watson et al., Nature 2023) for diffusion-model-based generative protein design. Designed binders against SARS-CoV-2 (Cao et al., Science 2020), influenza, and resistant cancer targets are entering clinical translation. Baker's 2024 Nobel half cited computational protein design specifically.

    Why it still matters: AlphaFold 3 (May 2024) and competing models (RoseTTAFold-All-Atom, Boltz-1) cover protein–ligand and protein–nucleic-acid complexes. Isomorphic Labs (Hassabis co-founder) announced its first pharma research collaborations in January 2024 — with Eli Lilly and Novartis — carrying roughly $83M in combined upfront payments and up to ~$2.9B in potential milestones for AlphaFold-informed drug discovery. Baker-lab spinouts (Outpace Bio, Vilya) are advancing designed-protein therapeutics. CASP16 (December 2024) shifted focus to multimers, ligand-bound complexes, and design rather than monomer prediction — the latter is largely solved. The field's next frontier is dynamics — folding pathways, allostery, and conformational ensembles — where prediction still lags experiment.

    MS-LS3-1MS-LS4-1HS-LS1-1HS-LS3-1

    Sources (5)