{"id":"AIBD-1227","title":"Anthropic says that Claude helped its biology lab find, isolate and map the first 3D structure of a vault particle from prokaryotes and cut a three-day purification to about 90 minutes","summary":"Anthropic says that Claude helped its lab solve the first 3D structure of a vault particle from prokaryotes and cut a three-day purification to about 90 minutes.","date_observed":"2026-11-04","date_basis":"source publication","date_precision":"day","benefit_status":"reported","account":"Vaults are large, hollow barrels of protein found in cells across the tree of life. Their function is still unknown. Anthropic's biology research lab used Claude at each step of a study of vaults in bacteria and in the viruses that infect them. Under the team's direction, Claude searched protein databases, designed the DNA constructs, optimized the isolation method and processed the microscopy data. The team mapped the first 3D structure of a vault particle found in prokaryotic cells. The particle came from a virus that infects bacteria. Anthropic says that a vault purification that conventionally takes three days took about 90 minutes. A first structural model was ready 34 hours later, which the team estimates would normally have taken weeks. Claude also made errors. A protocol that it wrote called for a growth medium with an additive that keeps DNA out of cells, which set the project back by two weeks. Early in the cryo-EM processing it skipped a standard step and produced a flawed map blended with its mirror image. Anthropic announced the work on 4 November 2026 and submitted a preprint to bioRxiv. The Neuron reported it on 5 November. The time savings are Anthropic's own estimates.","category":"benefit created","unexpected":false,"purpose":"Anthropic's biology research lab builds Claude into every step of laboratory research to speed up work in fundamental biology.","system":"Claude","model_version":null,"developer":"Anthropic","deployer":"Anthropic (biology research lab)","location":null,"domain":"Structural biology","segment":"Research","registry_ids":{},"ratings":[{"group":"organizational","depth":1,"reason":"Anthropic's lab reports that a three-day purification took about 90 minutes and that a first model came in 34 hours instead of weeks."},{"group":"societal","depth":1,"reason":"One new structure, of a vault particle from prokaryotes, adds a clue to the open question of what vaults do."}],"ai_role":{"status":"unassessed","statement":"Anthropic describes Claude's part at each step, under the team's direction. The faster purification came from a size-exclusion method that the team chose after Claude first argued against it, and Claude then helped make it work. The time savings are the makers' estimates, and no outside group has checked them."},"evidence":[{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"As a result, a vault purification that conventionally takes three days took us about 90 minutes, and a first structural model was ready 34 hours later, which we estimate would normally have taken us weeks."},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On the finding.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"the first 3D structure of a vault particle found in prokaryotic cells"},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On Claude's errors.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"A protocol it wrote called for a growth medium containing an additive that interferes with getting DNA into cells, which set the project back by two weeks.","supports_benefit":false},{"kind":"news report","citation":"The Neuron, 5 November 2026.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"In a November 4 announcement, Anthropic revealed that its biology team used Claude to reconstruct the first 3D structure of a vault particle derived from a virus that infects bacteria."},{"kind":"news report","citation":"The Neuron, 5 November 2026. On the errors.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"But the AI also made mistakes, including one that delayed the experiment by two weeks.","supports_benefit":false}],"replication":{"produced_by":["A language model given database searches, DNA design, method optimization and cryo-EM processing under a research team's direction","A wet lab that produced the proteins in bacteria and imaged them by cryo-electron microscopy"],"must_hold":["Researchers must check each step, as the protocol error and the flawed map show","The lab needs large protein databases and the computing to process cryo-EM data"],"echoes":[]},"paired_incident":null,"headline":"*Claude* helps Anthropic map the first vault particle from prokaryotes","metric":"a three-day purification to about 90 minutes","comparator":{"against":"the conventional vault purification","before":"about three days","after":"about 90 minutes","basis":"assumed","quote":"As a result, a vault purification that conventionally takes three days took us about 90 minutes, and a first structural model was ready 34 hours later, which we estimate would normally have taken us weeks.","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude"},"later":{"status":"unknown","date":null,"note":"The benefit was first reported on 4 November 2026, and no later report was found.","quote":null,"url":null},"learns_from_data":{"value":"shown","note":"Claude is a family of large language models made by Anthropic.","quote":"Claude is a series of large language models (LLMs) developed by American software company Anthropic.","url":"https://en.wikipedia.org/wiki/Claude_(language_model)"},"benefit_categories":["knowledge_discovery","operational_improvement"],"size":{"reach":{"band":1,"fact":"One laboratory, Anthropic's biology research lab, gained the faster purification.","quote":"As a result, a vault purification that conventionally takes three days took us about 90 minutes, and a first structural model was ready 34 hours later, which we estimate would normally have taken us weeks.","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","reason":null},"duration":{"band":5,"fact":"A solved structure is a finding that stands.","quote":"the first 3D structure of a vault particle found in prokaryotic cells","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","reason":null},"distribution":{"band":null,"fact":null,"quote":null,"url":null,"reason":"The sources compare no groups."},"frequency":{"band":1,"fact":"One structure was solved, once.","quote":"the first 3D structure of a vault particle found in prokaryotic cells","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","reason":null}},"value":null,"developer_country":"United States","deployer_country":"United States","unrated_because":null,"brief_notes":null,"brief":null,"confidence":"claimed","shade":"low","claim_status":"The makers' claim","claim_basis":"Sources: the makers' report and the makers' news report. For the benefit itself, the cited sources include no independent source and no finding by an external body.","case_jurisdictions":[],"design":"no comparison","certainty":"very low","certainty_reasons":["Starts very low because no comparison was measured."],"profile":{"case_id":"AIBD-1227","source_ids":["AIBD-1227-source-0001","AIBD-1227-source-0002","AIBD-1227-source-0003","AIBD-1227-source-0004","AIBD-1227-source-0005"],"families":{"strategic assessments":{"state":"unresolved","record_ids":[]},"evaluations":{"state":"unresolved","record_ids":[]},"monetary observations":{"state":"unresolved","record_ids":[]},"institutional decisions":{"state":"unresolved","record_ids":[]}},"fields":{"actors[]":{"state":"value_present","value":{"kind":"record_fields","fields":{"developer":"Anthropic","deployer":"Anthropic (biology research lab)"}}},"actors[].ownership_control":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"ai_contribution":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"ai_relationship":{"state":"value_present","value":{"kind":"record_fields","fields":{"category":"benefit created","ai_role":{"status":"unassessed","statement":"Anthropic describes Claude's part at each step, under the team's direction. The faster purification came from a size-exclusion method that the team chose after Claude first argued against it, and Claude then helped make it work. The time savings are the makers' estimates, and no outside group has checked them."}}}},"assessments[].alternatives":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"assessments[].assessor_date":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"assessments[].depth":{"state":"value_present","value":{"kind":"record_fields","fields":{"ratings":[{"group":"organizational","depth":1,"reason":"Anthropic's lab reports that a three-day purification took about 90 minutes and that a first model came in 34 hours instead of weeks."},{"group":"societal","depth":1,"reason":"One new structure, of a vault particle from prokaryotes, adds a clue to the open question of what vaults do."}]}}},"assessments[].display_confidence":{"state":"value_present","value":{"kind":"structured","data":{"confidence":"low","evidence_word":"claimed","claim_status":"The makers' claim","basis":"Sources: the makers' report and the makers' news report. For the benefit itself, the cited sources include no independent source and no finding by an external body.","interpretation":"Existing display rule preserved; the richer outcome-specific confidence question still needs review."}}},"assessments[].rationale":{"state":"value_present","value":{"kind":"record_fields","fields":{"ratings":[{"group":"organizational","depth":1,"reason":"Anthropic's lab reports that a three-day purification took about 90 minutes and that a first model came in 34 hours instead of weeks."},{"group":"societal","depth":1,"reason":"One new structure, of a vault particle from prokaryotes, adds a clue to the open question of what vaults do."}]}}},"assessments[].rubric_version":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"assessments[].status_basis":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"assessments[].target":{"state":"value_present","value":{"kind":"record_fields","fields":{"ratings":[{"group":"organizational","depth":1,"reason":"Anthropic's lab reports that a three-day purification took about 90 minutes and that a first model came in 34 hours instead of weeks."},{"group":"societal","depth":1,"reason":"One new structure, of a vault particle from prokaryotes, adds a clue to the open question of what vaults do."}]}}},"availability":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"brief":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"case_kind":{"state":"value_present","value":{"kind":"structured","data":{"classification":"program or activity","answer":"A program or activity. A research program in Anthropic's own laboratory.","scope":"The period and activity described in the case."}},"display":{"answer":"A program or activity. A research program in Anthropic's own laboratory.","inferred":false},"source_refs":["AIBD-1227-source-0001"]},"case_links[]":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"claims[].evidence_links":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"claims[].evidence_strength":{"state":"value_present","value":{"kind":"structured","data":{"answer":"The makers' claim. Sources: the makers' report and the makers' news report. For the benefit itself, the cited sources include no independent source and no finding by an external body.","evidence_word":"claimed","claim_status":"The makers' claim"}},"display":{"answer":"The makers' claim. Sources: the makers' report and the makers' news report. For the benefit itself, the cited sources include no independent source and no finding by an external body.","inferred":false},"source_refs":["AIBD-1227-source-0001","AIBD-1227-source-0002","AIBD-1227-source-0003","AIBD-1227-source-0004","AIBD-1227-source-0005"]},"claims[].missingness":{"state":"value_present","value":{"kind":"text","text":"The time savings are Anthropic's own estimates against a conventional method. The preprint had been submitted to bioRxiv but had not yet appeared, so no outside group had checked the work."},"display":{"answer":"The time savings are Anthropic's own estimates against a conventional method. The preprint had been submitted to bioRxiv but had not yet appeared, so no outside group had checked the work.","inferred":false},"source_refs":["AIBD-1227-source-0001"]},"claims[].proposition":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"cross_border_flows[]":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"decisions[].action_outcome":{"state":"unresolved","family":"institutional decisions"},"decisions[].grounds":{"state":"unresolved","family":"institutional decisions"},"decisions[].institution_task":{"state":"unresolved","family":"institutional decisions"},"decisions[].success_criterion":{"state":"unresolved","family":"institutional decisions"},"deployment_stage":{"state":"value_present","value":{"kind":"structured","data":{"classification":"routine use","answer":"The system was in routine use. Anthropic's lab builds Claude into every step of its research.","scope":"The period and activity described in the case."}},"display":{"answer":"The system was in routine use. Anthropic's lab builds Claude into every step of its research.","inferred":false},"source_refs":["AIBD-1227-source-0001"]},"economics[].amount":{"state":"unresolved","family":"monetary observations"},"economics[].costs[]":{"state":"unresolved","family":"monetary observations"},"economics[].counterfactual":{"state":"unresolved","family":"monetary observations"},"economics[].currency_price_basis":{"state":"unresolved","family":"monetary observations"},"economics[].gross_net":{"state":"unresolved","family":"monetary observations"},"economics[].measure_type":{"state":"unresolved","family":"monetary observations"},"economics[].period_unit":{"state":"unresolved","family":"monetary observations"},"economics[].realization":{"state":"unresolved","family":"monetary observations"},"economics[].recipient_payer":{"state":"unresolved","family":"monetary observations"},"economics[].return_payback":{"state":"unresolved","family":"monetary observations"},"economics[].valuation_method":{"state":"unresolved","family":"monetary observations"},"evaluations[].ai_uplift":{"state":"unresolved","family":"evaluations"},"evaluations[].comparator":{"state":"unresolved","family":"evaluations"},"evaluations[].design":{"state":"unresolved","family":"evaluations"},"evaluations[].effect":{"state":"unresolved","family":"evaluations"},"evaluations[].exposure":{"state":"unresolved","family":"evaluations"},"evaluations[].id":{"state":"unresolved","family":"evaluations"},"evaluations[].measurement_method":{"state":"unresolved","family":"evaluations"},"evaluations[].metric":{"state":"unresolved","family":"evaluations"},"evaluations[].reproduction_materials":{"state":"unresolved","family":"evaluations"},"evaluations[].setting_sample":{"state":"unresolved","family":"evaluations"},"evaluations[].uncertainty_limitations":{"state":"unresolved","family":"evaluations"},"external_references[]":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"follow_up[]":{"state":"value_present","value":{"kind":"record_fields","fields":{"later":{"status":"unknown","note":"The benefit was first reported on 4 November 2026, and no later report was found."}}}},"human_oversight":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"id":{"state":"value_present","value":{"kind":"record_fields","fields":{"id":"AIBD-1227"}}},"infrastructure_dependencies[]":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].beneficiaries":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].beneficiary_count":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].claim":{"state":"value_present","value":{"kind":"record_fields","fields":{"account":"Vaults are large, hollow barrels of protein found in cells across the tree of life. Their function is still unknown. Anthropic's biology research lab used Claude at each step of a study of vaults in bacteria and in the viruses that infect them. Under the team's direction, Claude searched protein databases, designed the DNA constructs, optimized the isolation method and processed the microscopy data. The team mapped the first 3D structure of a vault particle found in prokaryotic cells. The particle came from a virus that infects bacteria. Anthropic says that a vault purification that conventionally takes three days took about 90 minutes. A first structural model was ready 34 hours later, which the team estimates would normally have taken weeks. Claude also made errors. A protocol that it wrote called for a growth medium with an additive that keeps DNA out of cells, which set the project back by two weeks. Early in the cryo-EM processing it skipped a standard step and produced a flawed map blended with its mirror image. Anthropic announced the work on 4 November 2026 and submitted a preprint to bioRxiv. The Neuron reported it on 5 November. The time savings are Anthropic's own estimates."}}},"outcomes[].distribution":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].duration":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].id":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"outcomes[].stakeholder_group":{"state":"value_present","value":{"kind":"record_fields","fields":{"ratings":[{"group":"organizational","depth":1,"reason":"Anthropic's lab reports that a three-day purification took about 90 minutes and that a first model came in 34 hours instead of weeks."},{"group":"societal","depth":1,"reason":"One new structure, of a vault particle from prokaryotes, adds a clue to the open question of what vaults do."}]}}},"outcomes[].status":{"state":"value_present","value":{"kind":"record_fields","fields":{"benefit_status":"reported"}}},"outcomes[].tradeoffs":{"state":"value_present","value":{"kind":"text","text":"A protocol that Claude wrote set the project back by two weeks, and Claude produced a flawed cryo-EM map before the team stepped in."},"display":{"answer":"A protocol that Claude wrote set the project back by two weeks, and Claude produced a flawed cryo-EM map before the team stepped in.","inferred":false},"source_refs":["AIBD-1227-source-0001"]},"outcomes[].unexpected":{"state":"value_present","value":{"kind":"record_fields","fields":{"unexpected":false,"purpose":"Anthropic's biology research lab builds Claude into every step of laboratory research to speed up work in fundamental biology."}}},"places[]":{"state":"value_present","value":{"kind":"record_fields","fields":{"developer_country":"United States","deployer_country":"United States"}}},"purpose":{"state":"value_present","value":{"kind":"record_fields","fields":{"purpose":"Anthropic's biology research lab builds Claude into every step of laboratory research to speed up work in fundamental biology."}}},"record_dates":{"state":"value_present","value":{"kind":"record_fields","fields":{"date_observed":"2026-11-04","date_basis":"source publication","date_precision":"day"}}},"replication.attempts[]":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"replication.conditions":{"state":"value_present","value":{"kind":"record_fields","fields":{"replication":{"produced_by":["A language model given database searches, DNA design, method optimization and cryo-EM processing under a research team's direction","A wet lab that produced the proteins in bacteria and imaged them by cryo-electron microscopy"],"must_hold":["Researchers must check each step, as the protocol error and the flawed map show","The lab needs large protein databases and the computing to process cryo-EM data"],"echoes":[]}}}},"sector":{"state":"value_present","value":{"kind":"record_fields","fields":{"domain":"Structural biology","segment":"Research"}}},"sources[].access_version":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"sources[].citation_locator":{"state":"value_present","value":{"kind":"record_fields","fields":{"evidence":[{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"As a result, a vault purification that conventionally takes three days took us about 90 minutes, and a first structural model was ready 34 hours later, which we estimate would normally have taken us weeks."},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On the finding.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"the first 3D structure of a vault particle found in prokaryotic cells"},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On Claude's errors.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"A protocol it wrote called for a growth medium containing an additive that interferes with getting DNA into cells, which set the project back by two weeks.","supports_benefit":false},{"kind":"news report","citation":"The Neuron, 5 November 2026.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"In a November 4 announcement, Anthropic revealed that its biology team used Claude to reconstruct the first 3D structure of a vault particle derived from a virus that infects bacteria."},{"kind":"news report","citation":"The Neuron, 5 November 2026. On the errors.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"But the AI also made mistakes, including one that delayed the experiment by two weeks.","supports_benefit":false}]}}},"sources[].id":{"state":"value_present","value":{"kind":"identifiers","identifiers":["AIBD-1227-source-0001","AIBD-1227-source-0002","AIBD-1227-source-0003","AIBD-1227-source-0004","AIBD-1227-source-0005"],"role":"Existing evidence items, in their original order"}},"sources[].language":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"sources[].relationship":{"state":"value_present","value":{"kind":"record_fields","fields":{"evidence":[{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"As a result, a vault purification that conventionally takes three days took us about 90 minutes, and a first structural model was ready 34 hours later, which we estimate would normally have taken us weeks."},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On the finding.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"the first 3D structure of a vault particle found in prokaryotic cells"},{"kind":"report","citation":"Anthropic, Investigating vault particles with Claude, 4 November 2026. On Claude's errors.","title":"Investigating vault particles with Claude","url":"https://www.anthropic.com/research/investigating-vault-particles-with-claude","date":"2026-11-04","written_by":"makers","quote":"A protocol it wrote called for a growth medium containing an additive that interferes with getting DNA into cells, which set the project back by two weeks.","supports_benefit":false},{"kind":"news report","citation":"The Neuron, 5 November 2026.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"In a November 4 announcement, Anthropic revealed that its biology team used Claude to reconstruct the first 3D structure of a vault particle derived from a virus that infects bacteria."},{"kind":"news report","citation":"The Neuron, 5 November 2026. On the errors.","title":"Claude Helped Scientists Investigate a 40-Year-Old Cell Mystery. Here's What It Found","url":"https://www.theneuron.ai/news/claude-anthropic-40-year-old-cell-mystery/","date":"2026-11-05","written_by":"makers","quote":"But the AI also made mistakes, including one that delayed the experiment by two weeks.","supports_benefit":false}]}}},"sources[].upstream_sources":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"strategic_assessments[]":{"state":"unresolved","family":"strategic assessments"},"summary":{"state":"value_present","value":{"kind":"record_fields","fields":{"summary":"Anthropic says that Claude helped its lab solve the first 3D structure of a vault particle from prokaryotes and cut a three-day purification to about 90 minutes.","account":"Vaults are large, hollow barrels of protein found in cells across the tree of life. Their function is still unknown. Anthropic's biology research lab used Claude at each step of a study of vaults in bacteria and in the viruses that infect them. Under the team's direction, Claude searched protein databases, designed the DNA constructs, optimized the isolation method and processed the microscopy data. The team mapped the first 3D structure of a vault particle found in prokaryotic cells. The particle came from a virus that infects bacteria. Anthropic says that a vault purification that conventionally takes three days took about 90 minutes. A first structural model was ready 34 hours later, which the team estimates would normally have taken weeks. Claude also made errors. A protocol that it wrote called for a growth medium with an additive that keeps DNA out of cells, which set the project back by two weeks. Early in the cryo-EM processing it skipped a standard step and produced a flawed map blended with its mirror image. Anthropic announced the work on 4 November 2026 and submitted a preprint to bioRxiv. The Neuron reported it on 5 November. The time savings are Anthropic's own estimates."}}},"systems[].autonomy_permissions":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].component_role":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].deployment_configuration":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].inputs_outputs":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].model_name":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].model_version":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"systems[].product_name":{"state":"value_present","value":{"kind":"record_fields","fields":{"system":"Claude"}}},"systems[].provider":{"state":"unresolved","reason":"This is an explicit unresolved assignment: the sources on record do not settle this question."},"title":{"state":"value_present","value":{"kind":"record_fields","fields":{"title":"Anthropic says that Claude helped its biology lab find, isolate and map the first 3D structure of a vault particle from prokaryotes and cut a three-day purification to about 90 minutes","headline":"*Claude* helps Anthropic map the first vault particle from prokaryotes"}}},"version":{"state":"value_present","value":{"kind":"identifiers","identifiers":["1687c8bbcd00b79b083d4a1f9f763636c61e3c426a826955fba1a4897fc9726a"],"role":"SHA-256 of the complete public record and its assessments"},"evidence_note":"SHA-256 of the public record, its assessments, and its displayed answers and notes, excluding its own identifier and the derived coverage summary."},"workflow":{"state":"value_present","value":{"kind":"record_fields","fields":{"account":"Vaults are large, hollow barrels of protein found in cells across the tree of life. Their function is still unknown. Anthropic's biology research lab used Claude at each step of a study of vaults in bacteria and in the viruses that infect them. Under the team's direction, Claude searched protein databases, designed the DNA constructs, optimized the isolation method and processed the microscopy data. The team mapped the first 3D structure of a vault particle found in prokaryotic cells. The particle came from a virus that infects bacteria. Anthropic says that a vault purification that conventionally takes three days took about 90 minutes. A first structural model was ready 34 hours later, which the team estimates would normally have taken weeks. Claude also made errors. A protocol that it wrote called for a growth medium with an additive that keeps DNA out of cells, which set the project back by two weeks. Early in the cryo-EM processing it skipped a standard step and produced a flawed map blended with its mirror image. Anthropic announced the work on 4 November 2026 and submitted a preprint to bioRxiv. The Neuron reported it on 5 November. The time savings are Anthropic's own estimates.","replication":{"produced_by":["A language model given database searches, DNA design, method optimization and cryo-EM processing under a research team's direction","A wet lab that produced the proteins in bacteria and imaged them by cryo-electron microscopy"],"must_hold":["Researchers must check each step, as the protocol error and the flawed map show","The lab needs large protein databases and the computing to process cryo-EM data"],"echoes":[]}}}}}}}