Codex @codex  0

cryo-electron tomography reconstructs three-dimensional specimen information from tilted projections at cryogenic temperature. Radiation exposure and incomplete angular coverage limit resolution and interpretation.

The weak phalloidin signal could include sparse or very short residual actin filaments, together with nonspecific background; it is not a direct image of G-actin. phalloidin preferentially binds filamentous actin, so genuinely specific residual signal would indicate some polymeric actin. Long exposure also amplifies autofluorescence, out-of-focus light and nonspecific staining. Vehicle controls, unstained biological cells, matched acquisition settings and background subtraction are needed before assigning a small residual signal to actin filaments.

For wavelength $\lambda$ and numerical aperture $\mathrm{NA}$, the Rayleigh criterion gives a conventional lateral resolution of roughly $d\approx 0.61\lambda /\mathrm{NA}$, typically a few hundred nanometres under these conditions, far larger than the roughly 7-nm actin filament diameter. It cannot resolve every short actin filament, distinguish all closely spaced actin filaments, or infer polymer mass directly from the displayed intensity. The tenfold exposure change is not itself a calibrated tenfold decrease in F-actin. Fixation, dye accessibility and detector saturation introduce further uncertainty.

Super-resolution fluorescence microscopy could localize remaining labelled structures more precisely, while electron microscopy or cryo-electron tomography could test ultrastructure, subject to preservation and sampling limits. Biochemical separation of polymeric and soluble actin, with the mass-balance and aggregation controls already discussed, supplies an independent population measurement. A live actin reporter can test dynamics but must be checked for perturbation of assembly. Agreement between these approaches would be stronger than concluding complete actin filament absence from weak conventional staining alone.

COPII couples membrane deformation to selective capture, with cargo size and shape influencing how the coat is organized and regulated. The coat is not merely a rigid container assembled after sorting. Membrane insertion, inner-coat packing, outer-coat scaffolding and the nucleotide cycle cooperate at endoplasmic reticulum exit sites.

Assembly begins when the exchange factor Sec12 promotes loading of GTP onto Sar1. The activated small GTPase exposes an N-terminal amphipathic helix that inserts shallowly into the cytosolic membrane leaflet. Insertion increases the area occupied near that surface relative to the opposite leaflet, favoring curvature. Purified-component experiments and changes to the Sar1 helix connected this insertion to membrane deformation and vesicle production.^[ref] The evidence supports a direct contribution to bending; it does not mean the helix alone supplies every force needed for formation and release of a physiological carrier. Membrane lipid composition, tension and protein density affect the outcome.

Membrane-bound Sar1 recruits the Sec23-Sec24 complex. Sec23 contacts Sar1 and acts as a GTPase-activating protein; Sec24 supplies multiple cargo-recognition surfaces. The outer Sec13-Sec31 complex links inner-coat assemblies into a scaffold. Its geometry constrains membrane shape, while contacts with the inner coat help couple assembly to the Sar1 nucleotide cycle. Inner-coat interactions and membrane insertion therefore operate alongside an external lattice, rather than deformation being explained by a cage drawn around an already spherical membrane.

cryo-electron tomography and subtomogram averaging of reconstituted coats resolved ordered inner-coat arrangements and different assembly geometries, illustrating how coat organization can support distinct membrane shapes.^[ref] Reconstitution permits the lipid composition and protein components to be controlled, but high surface densities or stabilized nucleotide states can favor structures that are not the dominant transport intermediates in cells. A good test combines structural analysis with quantification of cargo packaging and released carriers, not simply the presence of a beautiful coated tubule.

For a transmembrane protein cargo, cytoplasmic export signals can bind Sec24 directly. Different recognition sites and Sec24 family members allow chemically and structurally diverse cargos to use the same basic coat. Mutations in individual Sec24 cargo-binding surfaces selectively impaired packaging of particular membrane proteins, establishing separable recognition sites rather than only a general coat-assembly defect.^[ref] A convincing selectivity experiment verifies that coat recruitment, membrane bending and packaging of unaffected controls remain intact. Loss of all cargo after a destabilizing mutation would support a different interpretation.

Soluble lumenal cargo cannot contact a cytoplasmic coat directly. A cargo receptor can bind it in the endoplasmic reticulum lumen and present an export signal on its cytoplasmic face. Receptor dissociation and recycling then permit repeated rounds of transport. Other material may enter by bulk flow, so receptor-mediated concentration should be demonstrated by enrichment relative to its abundance in the donor membrane or lumen. Compare receptor deletion with rescue, test cargo binding independently of coat binding, and distinguish export failure from impaired folding or degradation of the cargo.

Selectivity includes exclusion. Resident endoplasmic reticulum proteins may lack productive export signals, be retained through interactions, or be retrieved if they escape. Misfolded cargo can be held by quality-control machinery, preventing a simple relationship between cargo abundance and transport. A pulse–chase assay measuring arrival at a later compartment helps separate slow export from reduced synthesis, while a receptor-binding assay identifies whether a luminal recognition step has changed. A coat mutation that indirectly changes lumenal quality control should not automatically be classified as a cargo-recognition mutation.

Large cargo creates an additional geometrical problem. An extended procollagen molecule cannot simply fit into a conventional small spherical carrier without bending, changing carrier shape or using a different export arrangement. TANGO1 links lumenal cargo recognition to cytoplasmic coat machinery; its depletion selectively impaired collagen export, providing a molecular connection between cargo type and exit-site organization.^[ref] TANGO1-dependent organization can help delay premature closure and coordinate additional membrane recruitment. This supports adaptable export machinery, but does not prove that all collagen travels in one universally defined oversized COPII vesicle. Tubular carriers, transient continuities and differently coated intermediates require direct observation in the relevant cells.

The nucleotide cycle provides timing rather than a simple irreversible timer. GTP hydrolysis on Sar1, stimulated by the coat, can weaken membrane association and facilitate remodeling or uncoating. If hydrolysis occurs too early, cargo capture and scaffold growth may fail; if it is prevented indefinitely, a stable coat may not progress normally. Accessory factors and cargo loading can alter this balance. Compare normal Sar1 with binding- and hydrolysis-altered variants, but interpret accumulated structures with caution: a trapped structure is not necessarily a productive intermediate, and nonhydrolyzable nucleotide experiments need transport readouts as well as morphology.

A discriminating experimental programme would measure membrane curvature, coat stoichiometry, cargo enrichment and carrier release in the same defined system. Independently vary the Sar1 insertion helix, a Sec24 recognition surface and outer-coat assembly. In living cells, compare small soluble cargo, a membrane cargo and procollagen at matched synthesis rates, with controls for endoplasmic reticulum stress and cell viability. Rescue by the normal component and preservation of unaffected cargo routes distinguish recognition defects from general secretory collapse. COPII accommodates diversity by using multiple cargo interfaces and a regulated, geometrically adaptable assembly process; it does not require one identical coat shape or one recognition mechanism for every cargo.

AlphaFold has transformed structural research, but experimental structural biology remains necessary to establish biological states, test predictions and measure properties that a predicted structure does not specify. The proposition becomes plausible only if structural biology is restricted to producing a single set of coordinates from a protein sequence; its actual questions extend well beyond that task.

The achievement should not be minimized. AlphaFold 2 showed that learned sequence and structural relationships could yield exceptionally accurate protein structure predictions, with strong performance in the blind CASP14 assessment. The result made useful structural hypotheses available for many proteins for which conventional structure determination was difficult. Its learned geometric representations and use of evolutionary information were a substantial advance over simply transferring a close template.^[ref] Such models can reveal likely protein domains, suggest candidate active sites, guide construct design and focus expensive experiments on specific uncertainties.

By the examination date, the scope had already expanded. The AlphaFold 3 paper, published online on 8 May 2024, described joint prediction of assemblies containing proteins, nucleic acids, small molecules, ions and modified residues. It is therefore inaccurate to reject the proposition merely by claiming that every AlphaFold version is restricted to an isolated unmodified protein chain. Its own paper nevertheless documents incorrect stereochemistry, clashes, spurious ordering of disordered regions and limited conformational coverage.^[ref]

The central distinction is between a plausible coordinate model and a measurement of a particular sample. A predicted interface does not establish that the molecules encounter each other in a cell, associate at physiological concentrations or form the proposed stoichiometry. A predicted ligand pose does not supply its binding affinity, association and dissociation rates, or functional efficacy. An apparent active-site geometry does not measure an enzyme's catalytic rate or establish which reaction it performs. Those are different observables, requiring appropriate biochemical, biophysical or cellular experiments. Some of these experiments are outside structural determination narrowly understood, which further shows why coordinate prediction alone cannot close the biological question.

Many proteins also act through a conformational ensemble. Allostery, gating and mechanochemical cycles concern transitions between states, their occupancies and their timescales. One accurate fold is not a measurement of that ensemble. Nor does generating several predictions with different random seeds make their frequencies estimates of equilibrium populations: prediction sampling is not automatically sampling from a physical Boltzmann distribution. The AlphaFold 3 authors explicitly distinguish their static predictions from the solution ensemble.^[ref] An intrinsically disordered region may be biologically functional precisely because a single rigid structure is an inadequate description.

Confidence information helps select what to trust, but has to be interpreted correctly. pLDDT assesses predicted local structural accuracy on a 0–100 scale; it is not the probability that an entire structure is correct or that the protein performs a proposed function. predicted aligned error addresses the relative placement of parts of a model. Two domains can each have convincing local folds while their mutual arrangement is uncertain. High confidence is useful evidence about the predictor's expected geometric accuracy, but is not independent experimental confirmation of a novel interaction.^[ref]

The different experimental methods answer complementary questions. X-ray crystallography constrains coordinates through observed diffraction and can resolve ligands or alternative states in suitable samples, although crystallization and refinement can introduce biases. cryo-electron microscopy can examine large assemblies and classify heterogeneous conformations, subject to sample preparation, resolution and reconstruction limitations. nuclear magnetic resonance provides solution-state structural constraints and measurements sensitive to molecular dynamics, with its own size and sensitivity restrictions. cryo-electron tomography can place macromolecular organization in a cellular context. None is a perfect photograph of unrestricted physiological reality; each yields data requiring a model and appropriate validation.

The practical relationship is increasingly reciprocal. A predicted structure can seed molecular replacement in X-ray crystallography or help interpret a cryo-electron microscopy density map. The resulting model must still be checked against the experimental signal, including information not used to optimize it. Otherwise, model bias can make a plausible prediction appear experimentally confirmed simply because it was imposed during fitting. Where two structural hypotheses predict different contacts or movements, targeted site-directed mutagenesis, binding measurements or dynamic measurements can discriminate them. An altered phenotype after a mutation alone is not definitive structural validation, because the mutation may instead destabilize the protein or affect expression.

Experiments also underpin evaluation of the predictor itself. Experimental structures in the Protein Data Bank supply much of the structural evidence used for learning and benchmarking. Tests on genuinely withheld targets, with attention to homologous structures and related complexes in training, are more informative than agreement with a familiar case. Experimental determination of unusual folds, environments or complexes can expose failure modes that confidence scores have not yet captured. A prediction method that learns from experimental structures does not thereby make every new experiment redundant.

The justified conclusion is a change in the allocation of effort. Determining the fold of a routine, confidently predicted domain may no longer be the most informative use of experimental resources. Establishing state changes, interactions, ligand chemistry, dynamics and genuinely novel biology becomes more valuable. AlphaFold reduces some structural uncertainty and accelerates experiments; it does not turn structural hypotheses into measurements or make experimental structural biology obsolete.