Function
Processes
Deforms a donor host membrane into the closed or necked geometry that defines the factory lumen.
A viral factory (VF) is a virus-induced intracellular microcompartment that concentrates the viral genome, the replication and transcription machinery, and the structural proteins of the progeny virion into one place, raising the local concentration of every reactant in the replication cycle while screening replication intermediates from host nucleic-acid sensors and nucleases. The module is deliberately mechanism-neutral at its core and taxon-spanning: what makes a VF a VF is the compartment boundary plus the co-localised replication, transcription, and assembly programme, not any one way of drawing the boundary. Three boundary mechanisms are modelled as variants along a single axis - a host-membrane-derived boundary (spherules, tubes, double-membrane vesicles, convoluted ER), a membraneless biomolecular condensate formed by liquid-liquid phase separation, and a self-assembled proteinaceous shell - because the same downstream programme runs inside all three. Around that core the module places the recruitment of host systems the factory depends on (cytoskeleton, mitochondria, donor membranes, host lipid kinases, in some lineages host transcription machinery), the shielding of replication intermediates from host defences, non-lytic egress, and the antagonists that act on the factory itself.
Scope decisions. (1) The module is ABSTRACT: the boundary it draws is the factory itself, which recurs across RNA and DNA viruses of eukaryotes and across jumbo phages of bacteria, so most leaf roles are selectors with orienting representative members rather than a species-specific member list. Representative members are examples only; where a strain-specific reference entry exists it is cited, and representatives within one family descriptor may come from different reference strains. (2) Phage nuclei are included as a boundary variant, with a caveat. The source review is explicit that the chimallin-shelled compartments of Chimalliviridae are "not considered true VFs", and GO has no term for them. They are modelled here anyway because they satisfy the module's functional definition exactly - a boundary that concentrates replication and transcription, excludes translation, and blocks host nucleases - and because treating the boundary mechanism as the variant axis is the whole point of the module. The caveat is recorded on the variant node itself, not hidden here. (3) Translation is deliberately outside the module. In every system where it has been localised - mimivirus, and the phage nucleus - translation happens outside the factory, and that exclusion is a property worth stating rather than a part to include. (4) Host restriction factors and virophages are grouped as one optional "antagonists" part rather than scattered as negative regulation on individual steps, because both act on the assembled factory as a unit. (5) Provenance caution on the GO groundings. The GO viral-factory branch used throughout this module is not independent of ViralZone. Only GO:0039713 carries the dbxref VZ:1951, and it is the only term in the branch that the SPKW ViralZone audit scores (0.82 best-sentence similarity, its CLOSE_PARAPHRASE band). The children - spherule, double membrane vesicle, tube, virogenic stroma, peristromal region viral factory - carry no VZ dbxref, so they are outside that generated table; that their definitions also track the sections of VZ:1951 is this curator's reading on comparing the texts, not a measured result. GO does drop ViralZone's claim that filoviruses produce viroplasms. Citing VZ:1951 alongside those GO terms would therefore count one source twice rather than corroborate, so it is recorded here and not as an evidence item. The primary literature cited on each boundary variant is the independent support.
knowledge_gaps[0].provenance[1] · supporting_text
(0/1)✗ none found
No MODULE:viral_factory deep-research report alongside the module YAML.
✓ representative grounding skipped for abstract module.
✓ every declared conforms_to bundle matches its template motif.
✓ every PRECEDES step chains, or its break is acknowledged via chaining_status.
2 gene-function check(s) reflected in core functions · 0 conflict(s) · 26 evidence or coverage gap(s) (33 comparison(s))
Retained noncore functions are reported separately; they do not require promotion to core functions. Support from representative genes does not establish conservation across the whole family. Missing reviews or identifiers leave a claim unverified.
| Annoton | Participant | Function | Assessment |
|---|---|---|---|
| vf_membrane_remodeling | ANY_WITH_FUNCTION Gene |
membrane bending activity GO:0180020 |
Participant Unchecked
No direct UniProt participant to compare. Abstract selectors and complex-wide activities cannot be inferred from donor or subunit gene reviews. |
| vf_alphavirus_spherule | Chikungunya virus non-structural polyprotein P1234 UniProtKB:Q8JUX6 Gene |
RNA-directed RNA polymerase activity GO:0003968 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_alphavirus_nsp1_membrane_shaping | Chikungunya virus mRNA-capping enzyme nsP1 UniProtKB:Q8JUX6 Gene |
membrane bending activity GO:0180020 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_tombusvirus_spherule | Tomato bushy stunt virus p92 replicase (and its p33 cleavage product) UniProtKB:P15962 Gene |
RNA-directed RNA polymerase activity GO:0003968 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_nucleoprotein_condensate_scaffold | N (rabies virus nucleoprotein) UniProtKB:P16285 Family Representative |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_nucleoprotein_condensate_scaffold | N (measles virus nucleoprotein) UniProtKB:P04851 Family Representative |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_nucleoprotein_condensate_scaffold | NP (Zaire ebolavirus nucleoprotein) UniProtKB:P18272 Family Representative |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_phosphoprotein_condensate_partner | P (rabies virus phosphoprotein) UniProtKB:P69479 Family Representative |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_phosphoprotein_condensate_partner | P (measles virus phosphoprotein) UniProtKB:P03422 Family Representative |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_coronavirus_n_condensate | SARS-CoV-2 nucleoprotein N UniProtKB:P0DTC9 Gene |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_poxvirus_h5_condensate | Vaccinia virus H5 (OPG110) UniProtKB:P07242 Gene |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_nuclear_condensate_ebna2 | Epstein-Barr virus nuclear antigen 2 (EBNA2) UniProtKB:P12978 Gene |
molecular condensate scaffold activity GO:0140693 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_chimallin_shell | ChmA (Pseudomonas phage 201phi2-1 chimallin gp105) UniProtKB:B3FIW8 Family Representative |
structural molecule activity GO:0005198 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_chimallin_shell | ChmA (Pseudomonas phage phiKZ chimallin gp54) UniProtKB:Q8SDA8 Family Representative |
structural molecule activity GO:0005198 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_vimentin_cage | human vimentin UniProtKB:P08670 Gene |
structural constituent of cytoskeleton GO:0005200 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_pi4k_recruitment | human phosphatidylinositol 4-kinase beta (PI4KB) UniProtKB:Q9UBF8 Gene |
1-phosphatidylinositol 4-kinase activity GO:0004430 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_phuz_spindle | PhuZ (Pseudomonas phage 201phi2-1 phage tubulin) UniProtKB:B3FK34 Family Representative |
GTPase activity GO:0003924 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_phuz_spindle | PhuZ (Pseudomonas phage phiKZ phage tubulin) UniProtKB:Q8SDC3 Family Representative |
GTPase activity GO:0003924 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_viral_dna_dependent_rna_polymerase | ANY_WITH_FUNCTION Gene |
DNA-directed RNA polymerase activity GO:0003899 |
Participant Unchecked
No direct UniProt participant to compare. Abstract selectors and complex-wide activities cannot be inferred from donor or subunit gene reviews. |
| vf_mononegavirales_l_polymerase | L (rabies virus large structural protein) UniProtKB:A3F5L9 Family Representative |
RNA-directed RNA polymerase activity GO:0003968 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_mononegavirales_l_polymerase | L (measles virus RNA-directed RNA polymerase L) UniProtKB:P12576 Family Representative |
RNA-directed RNA polymerase activity GO:0003968 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_mononegavirales_l_polymerase | L (Zaire ebolavirus RNA-directed RNA polymerase L) UniProtKB:Q05318 Family Representative |
RNA-directed RNA polymerase activity GO:0003968 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_viral_dna_polymerase | ANY_WITH_FUNCTION Gene |
DNA-directed DNA polymerase activity GO:0003887 |
Participant Unchecked
No direct UniProt participant to compare. Abstract selectors and complex-wide activities cannot be inferred from donor or subunit gene reviews. |
| vf_viral_rna_polymerase | ANY_WITH_FUNCTION Gene |
RNA-directed RNA polymerase activity GO:0003968 |
Participant Unchecked
No direct UniProt participant to compare. Abstract selectors and complex-wide activities cannot be inferred from donor or subunit gene reviews. |
| vf_capsid_assembly | ANY_WITH_FUNCTION Gene |
structural constituent of virion GO:0039660 |
Participant Unchecked
No direct UniProt participant to compare. Abstract selectors and complex-wide activities cannot be inferred from donor or subunit gene reviews. |
| vf_asfv_capsid_example | African swine fever virus major capsid protein p72 UniProtKB:P22776 Gene |
structural constituent of virion GO:0039660 |
Review Missing
No ai-review exists for this protein; its function and core-function coverage cannot yet be checked. |
| vf_fam111a_restriction | human serine protease FAM111A UniProtKB:Q96PZ2 Gene |
serine-type peptidase activity GO:0008236 |
Core Supported
The GO molecular function is reflected in core_functions (exact term or a reviewed subclass). |
| vf_fam111b_restriction | human serine protease FAM111B UniProtKB:Q6SJ93 Gene |
serine-type peptidase activity GO:0008236 |
Core Supported
The GO molecular function is reflected in core_functions (exact term or a reviewed subclass). |
| vf_nucleoprotein_condensate_scaffold | Mononegavirales nucleoprotein (N/NP) Family |
molecular condensate scaffold activity GO:0140693 |
Family Id Missing
The family has no identifier linking it to a curated family review. |
| vf_phosphoprotein_condensate_partner | Mononegavirales phosphoprotein (P) Family |
molecular condensate scaffold activity GO:0140693 |
Family Id Missing
The family has no identifier linking it to a curated family review. |
| vf_chimallin_shell | Chimallin (ChmA) phage nuclear shell protein Family |
structural molecule activity GO:0005198 |
Family Id Missing
The family has no identifier linking it to a curated family review. |
| vf_phuz_spindle | PhuZ phage tubulin Family |
GTPase activity GO:0003924 |
Family Id Missing
The family has no identifier linking it to a curated family review. |
| vf_mononegavirales_l_polymerase | Mononegavirales large structural protein (L) Family |
RNA-directed RNA polymerase activity GO:0003968 |
Family Id Missing
The family has no identifier linking it to a curated family review. |
✓ every negative regulator described as feedback has an upstream edge closing the loop.
1 complete review(s) · 0 with deep research · 24 missing review · 2 reviewed but lacking deep research
| Gene | Review | Complete | Deep research |
|---|---|---|---|
| Rotavirus A SA11 non-structural protein 2 A2T3P0 | ✗ | — | — |
| Rotavirus A SA11 non-structural protein 5 A2T3Q9 | ✗ | — | — |
| L (rabies virus large structural protein) A3F5L9 | ✗ | — | — |
| ChmA (Pseudomonas phage 201phi2-1 chimallin gp105) B3FIW8 | ✗ | — | — |
| PhuZ (Pseudomonas phage 201phi2-1 phage tubulin) B3FK34 | ✗ | — | — |
| FAM111A Q96PZ2 | ✓ | ✓ | ✗ |
| FAM111B Q6SJ93 | ✓ | 16/19 | ✗ |
| P (measles virus phosphoprotein) P03422 | ✗ | — | — |
| N (measles virus nucleoprotein) P04851 | ✗ | — | — |
| Vaccinia virus H5 (OPG110) P07242 | ✗ | — | — |
| human vimentin P08670 | ✗ | — | — |
| SARS-CoV-2 replicase polyprotein 1a (nsp3 pore component) P0DTC1 | ✗ | — | — |
| SARS-CoV-2 nucleoprotein N P0DTC9 | ✗ | — | — |
| L (measles virus RNA-directed RNA polymerase L) P12576 | ✗ | — | — |
| Epstein-Barr virus nuclear antigen 2 (EBNA2) P12978 | ✗ | — | — |
| Tomato bushy stunt virus p92 replicase (and its p33 cleavage product) P15962 | ✗ | — | — |
| N (rabies virus nucleoprotein) P16285 | ✗ | — | — |
| NP (Zaire ebolavirus nucleoprotein) P18272 | ✗ | — | — |
| Dengue virus 2 genome polyprotein (NS4A/NS4B membrane-remodelling region) P18356 | ✗ | — | — |
| African swine fever virus major capsid protein p72 P22776 | ✗ | — | — |
| P (rabies virus phosphoprotein) P69479 | ✗ | — | — |
| L (Zaire ebolavirus RNA-directed RNA polymerase L) Q05318 | ✗ | — | — |
| Chikungunya virus non-structural polyprotein P1234 Q8JUX6 | ✗ | — | — |
| ChmA (Pseudomonas phage phiKZ chimallin gp54) Q8SDA8 | ✗ | — | — |
| PhuZ (Pseudomonas phage phiKZ phage tubulin) Q8SDC3 | ✗ | — | — |
| human phosphatidylinositol 4-kinase beta (PI4KB) Q9UBF8 | ✗ | — | — |
Biogenesis and operation of the virus-induced compartment: a boundary is drawn in the host cell, host systems are recruited to and around it, viral transcription and genome replication run inside it, progeny virions are assembled and packaged within it, and replication intermediates are screened from host defences throughout.
Establishment of the physical boundary that separates factory contents from bulk cytoplasm or nucleoplasm. The boundary is what makes the factory a compartment; the variants below are the three physically distinct ways it is drawn.
ONE_OR_MORE rather than EXACTLY_ONE: coronaviruses run a membrane-bounded replication organelle and a separate nucleoprotein condensate in the same cell, and Chimalliviridae pass through a membrane-bound vesicle before building the protein shell.
The boundary is a host membrane deformed by viral proteins into an invagination (spherule), a tube, a double-membrane vesicle, or a convoluted network. The lumen is topologically separated from the cytosol, so a pore or neck is required to import nucleotides and export progeny RNA.
Deforms a donor host membrane into the closed or necked geometry that defines the factory lumen.
Solves the topological problem created by a sealed membrane boundary: exports newly synthesised genomes and mRNAs from the DMV lumen to the cytosol for translation and packaging. No molecular-function term is asserted - the pore's transport activity has not been reconstituted.
The donor is a fixed, lineage-specific choice, which is why this axis is EXACTLY_ONE while the boundary-mechanism axis above is not.
Flaviviruses invaginate the rough ER into vesicle packets and convoluted membranes; coronaviruses derive double-membrane vesicles from the ER-Golgi intermediate compartment.
Representative of the flaviviral replication organelle: a single ER-derived network carrying both the replication vesicles and, apposed to their pores, the budding sites.
Alphaviruses generate spherules by deforming the host plasma membrane; in chikungunya virus these persist at the plasma membrane rather than being internalised.
Occupies the spherule and synthesises RNA inside it; the replication complex sits at the spherule neck, which stays open to the cytosol so nascent RNA is extruded rather than exported through a pore. The boundary-shaping half of this variant's claim is asserted separately on the nsP1 annoton below, which is the cleavage product that carries it.
Grounds the boundary-forming activity this part models in a concrete exemplar: a dodecameric ring that inserts amphipathic spikes into the host plasma membrane, anchors the replication complex to it, and forms the pore at the spherule neck.
Tombusviruses build spherules on the peroxisomal boundary membrane (and on the ER when peroxisomes are absent), the clearest case of a factory boundary drawn on an organelle unrelated to the secretory pathway.
Replicase that carries out RNA synthesis inside the spherule. The spherule-nucleating role is attributed to its p33 cleavage product rather than to the polymerase, and no membrane-shaping function is asserted here: unlike the alphavirus case above, no reference in this module establishes the mechanism, so this donor variant grounds the location and the polymerase only.
The boundary is a phase boundary rather than a membrane: viral scaffold proteins with intrinsically disordered regions, in concert with the viral genome, demix from the cytoplasm or nucleoplasm into a liquid condensate. Because the boundary is permeable to small molecules and to diffusing macromolecules, no pore is needed; concentration and selective partitioning do the work instead. The recurring architecture across Mononegavirales is a nucleoprotein-phosphoprotein pair, at least one of which contributes an essential disordered region.
Principal mass of the condensate and the RNA-binding partner; encapsidates nascent genomes into nucleocapsids within the droplet, where the encapsidation rate is higher than in the dilute phase.
Supplies the intrinsically disordered region and the dimerization domain required for demixing, and tethers the L polymerase to the N-RNA template inside the condensate. In rabies virus both elements of P are individually essential for Negri-body-like structures to form.
The second, condensate-based factory site of a virus whose RNA synthesis happens in a membrane-bounded organelle: packaging is spatially separated from replication, which is why the boundary-mechanism axis is ONE_OR_MORE.
Protein-RNA condensate in which the eleven genome segments of a segmented dsRNA virus are co-packaged; the one case where the factory's material properties are shown to change over the infection cycle.
Demonstrates that condensate boundaries are not restricted to RNA viruses: a DNA-binding poxviral protein phase separates through its N-terminal IDR, and DNA binding is required for both the phase separation and factory assembly. The same behaviour is seen for H5 orthologs of mpox virus and lumpy skin disease virus.
With its coactivator EBNALP, forms nuclear liquid-like puncta enriched at host superenhancers. Included as the nuclear counterpart of the cytoplasmic condensate variant; here the condensate organises viral control of host gene expression rather than genome replication, which is the main respect in which nuclear condensate factories differ.
The boundary is a single-layer protein lattice with ~2 nm pores, built de novo by the virus rather than borrowed from the host. It is selectively permeable: nucleotides and small proteins enter, but ribosomes and host nucleases are excluded, so transcription inside is physically separated from translation outside. In Chimalliviridae this shell is preceded by a transient membrane-bound compartment, making the two boundary mechanisms sequential within one infection.
Caveat carried from the source review: these phage compartments are conventionally described as analogous to, rather than instances of, viral factories, and GO has no term for them. They are retained here as a boundary variant because they meet this module's functional definition; see the module-level ontology knowledge gap.
Self-assembles as a flexible sheet into a closed micrometre-scale compartment around the replicating phage genome; the sole principal component of the shell.
A transcriptionally active, membrane-bound compartment that holds the injected genome before it is transferred into the chimallin shell; genome replication begins only after that transfer. Establishes that the two boundary mechanisms are used in sequence, not as alternatives, in this lineage.
The factory is not self-sufficient. It draws in host membranes and lipids for its boundary, host cytoskeleton for its position and structural integrity, host mitochondria for the ATP its morphogenesis consumes, and in some lineages host transcription factors it does not encode. These recruitments are modelled as a variant set because different viruses recruit different subsets, not as a fixed sequence.
Vimentin collapses from its dispersed cytoplasmic network into a cage surrounding the perinuclear factory, the same rearrangement the cell makes around an aggresome. This is the strongest evidence that factories of the large cytoplasmic DNA viruses co-opt a pre-existing host response rather than building a structure from scratch.
Structural cage enclosing the factory. PMID:11331297 shows that ASFV factories and aggresomes share the same vimentin rearrangement, that both assemble near the microtubule organizing center, and that both require an intact microtubule network - but it establishes the cage correlatively and does not test whether vimentin is required for factory integrity. The source review also reports a vimentin requirement for Ebola virus factories; that claim is not repeated here because no reference in this module supports it.
Mitochondria migrate en masse along microtubules to the factory and cluster at its periphery, adopting the ultrastructure of actively respiring organelles. The recruitment is selective for factories containing assembling particles rather than precursor membranes, consistent with supplying the ATP that morphogenesis consumes.
Places an ATP source immediately outside the factory boundary during the assembly phase.
Picornaviral and some flaviviral replication organelles are not merely shaped host membrane but chemically distinct from it: viral proteins redirect Arf1/GBF1 effector recruitment to enrich the organelle in phosphatidylinositol-4-phosphate, which the viral RNA polymerase binds directly.
Host enzyme recruited to the nascent replication organelle; its product PI4P both defines the organelle's lipid identity and is the direct binding partner of the viral RNA-dependent RNA polymerase, coupling membrane composition to polymerase localisation.
Marseilleviridae were assumed to be fully cytoplasmic on gene content alone, but their virions do not package a transcription apparatus. Instead the factory transiently draws host nuclear transcription machinery out to the cytoplasm, the host nucleus becoming briefly leaky and then recovering. A striking case of a factory importing a function rather than encoding it.
Supplies the transcription activity the virion does not carry, making an otherwise cytoplasmic replication cycle nucleus-dependent.
The bacterial counterpart of cytoskeletal recruitment, except that the cytoskeleton is virus-encoded: a tubulin-like protein builds a bipolar spindle that pushes the phage nucleus to midcell and holds it there. Capsids are trafficked to the shell surface for packaging, but no reference cited in this module attributes that trafficking to the spindle, so it is modelled on the assembly part rather than here.
Positions the compartment at midcell by pushing against it with dynamically unstable bipolar filaments. Capsids do migrate from the membrane to the compartment surface for packaging (PMID:28082593, and modelled on the assembly part of this module), but neither reference cited here attributes that trafficking to PhuZ, so no such role is asserted for it.
Transcription occurs at defined sites within the factory, and its products leave. Where the geometry has been resolved - mimivirus and the phage nucleus - transcription is physically separated from translation, which occurs in a ring or zone immediately outside the compartment. That separation is a defining property of the factory rather than an incidental one, and it is the observation the viral eukaryogenesis argument rests on.
Cytoplasmic DNA viruses (poxviruses, Nucleocytoviricota) carry their own multisubunit RNA polymerase into the factory, which is what allows a DNA virus to transcribe outside the nucleus at all. Marseilleviridae are the exception that proves the point: lacking a packaged polymerase, they must recruit the host's.
Transcribes the viral genome at defined foci inside the factory. Left as an abstract selector: the giant-virus polymerases are poorly represented in reviewed UniProt entries.
In Mononegavirales the same L polymerase both transcribes and replicates, switching mode according to the N/P context it meets - which is precisely the context the condensate concentrates.
Recruited into the condensate by the phosphoprotein, where it clusters into foci whose spacing changes when replication is switched on.
Genome synthesis is the function the compartment exists to accelerate and to hide. Its position within the factory is the best-established fact about coronaviral and flaviviral replication organelles, and the step that phage nucleus formation gates.
Replicates the genome within the factory. In vaccinia the ER-derived membranes surrounding early factories coincide with active DNA replication and disperse as the factory shifts to virion assembly, so the boundary and this step are coupled in time.
Synthesises genomic RNA inside the compartment; in membrane-bounded factories it is anchored at the spherule neck or DMV interior, in condensate factories it is concentrated by the scaffold proteins.
The factory's late programme. Where it has been resolved, assembly is spatially ordered: in giant viruses morphogenesis initiates centrally and matures toward the factory periphery; in jumbo phages capsids assemble on the cytoplasmic membrane and dock on the shell surface to be filled. Either way, packaging happens at the boundary, which is why assembly and the boundary cannot be separated into independent modules.
Concentrated into the factory to near-exclusivity, which is the original observation that motivated the aggresome comparison.
Concrete grounding for the abstract capsid role: supplies roughly a third of virion mass and is almost exclusively localised to the perinuclear factory during infection.
In the phage nucleus the geometry is explicit: capsids assemble outside the compartment, are trafficked to its surface, and are filled there, so the shell is both the container of the genome and the docking surface for packaging.
Compartmentalisation is simultaneously a concentration device and a concealment device, and GO's definition of a viral factory names both. The shielding is a consequence of the boundary rather than a separate machine, which is why this part carries a CAUSES edge from boundary formation rather than sitting in the temporal chain.
The cleanest demonstration that compartmentalisation alone confers immunity: the phage shell blocks CRISPR-Cas and restriction enzymes without any anti-CRISPR chemistry, and the protection is lost when the nuclease is engineered to enter. The complementary result - sensitivity to RNA-targeting Cas13a, whose substrate leaves the compartment - shows the barrier is spatial rather than biochemical.
Physically segregates the phage genome from host immunity nucleases for the whole infection. The exclusion is steric, not inhibitory: relocalising EcoRI inside the compartment restores targeting, and RNA-targeting Cas13a remains effective because its substrate leaves the shell.
In membrane-bounded factories the double-stranded replication intermediate, the ligand for cytosolic pattern-recognition receptors, sits inside a sealed lumen. This is the most widely invoked - and the least directly demonstrated - shielding mechanism: the dsRNA has been localised inside the vesicles, but the causal link to reduced sensing is usually inferred.
Keeps the dsRNA replication intermediate out of the cytosol where it would be sensed.
Optional because many infections end in lysis. Where non-lytic egress has been resolved, the route runs through host organelles recruited to the factory: reovirus virions pass from the inclusion into modified, de-acidified lysosome-derived sorting organelles that bud membranous carriers to the plasma membrane. Exocytosis-like release is also inferred for pandoraviruses and cedratviruses from vesicle accumulation in the absence of lysis.
Receives mature virions from the inclusion through membrane-fusion points and delivers them to the plasma membrane in membranous carriers, without lysing the cell.
Two unrelated classes of antagonist converge on the factory as a unit rather than on any individual step: host restriction proteases that accumulate in replication compartments, and virophages that parasitise the giant-virus factory as a replication niche. Grouped as one optional part because both presuppose a fully formed factory.
Host range restriction factor counteracted by the SV40 large T antigen C-terminal region; its proteolytic activity is the presumed but not demonstrated basis of the restriction.
The paralog that actually restricts murine cytomegalovirus in primate cells, and the clearest case of a host factor localising to the factory itself. Present in primates and absent in rodents, which is why MCMV has no countermeasure for it.
Cannot replicate in the amoeba alone and depends on the giant-virus factory as its replication niche, reducing the giant virus's infectious progeny. Evidence that the factory is a resource exploitable by a third party - the strongest functional analogy between a factory and a cellular organelle.