Multimodal diagnostic models and subtype analysis for neoadjuvant therapy in breast cancer.

{"sci_version": "1.0", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:40170854", "kind": "science-record", "revision": 1, "content_hash": "sha256:51aab8dbddbaee79df06aa24b84ed2d5006630927189313f529353cf6d6a5b02", "created_at": "2026-11-01T10:25:50Z", "updated_at": "2026-11-01T10:25:50Z", "generator": {"name": "corpus-assay-engine", "converter": "sep_to_sci_v1", "source_package_schema": "urn:bntrg:sep:1.0.0"}, "paper": {"state": "partial", "title": "Multimodal diagnostic models and subtype analysis for neoadjuvant therapy in breast cancer.", "doi": "10.3389/fimmu.2025.1559200", "pmid": "40170854", "authors": null, "journal": "Frontiers in immunology", "year": 2025, "abstract": null, "claims": [{"id": "claim-1", "statement": "the headline numerical result the paper states for its main finding", "kind": "qualitative", "state": "blocked", "target": {"value": null, "unit": null, "tolerance": null, "source": null}, "evidence": []}], "gaps": ["abstract", "authors"]}, "experiment": {"state": "partial", "objective": "the headline numerical result the paper states for its main finding", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate and its interval from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The number a replication must land on, and the sentence it came from.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:40170854"], "outputs": ["the published estimate, its unit and its interval"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data behind the published result", "kind": "data-preparation", "state": "blocked", "blocked_on": "data:not_downloaded", "description": "The published result cannot be recomputed without the data the paper analysed.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the source data with its provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the analysis programme", "kind": "analysis", "state": "blocked", "blocked_on": "no owned computational capability is bound to this experiment", "description": "Re-run the paper's analysis on the source data and produce the registered result. | SEP input not carried into the asset list: materials:the source data behind the published result in pmid:40170854", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the registered result"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40170854' does not resolve to a declared asset"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:40170854", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.3389/fimmu.2025.1559200", "sha256": "sha256:596b9414d75f7cf6f0ed015277de6a52168671c1094ba41e292b5490fdb5cf9e", "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.3389/fimmu.2025.1559200", "version": null, "note": "Source read by the package builder: full_text.", "structure": {"access_state": "full_text"}}], "gaps": ["asset:source:pmid:40170854 not verified"]}, "analysis": {"state": "absent", "methods": [], "targets": [], "gaps": ["method", "targets"]}, "capability": {"state": "absent", "requirements": [], "gaps": ["capability"]}, "readiness": {"state": "partial", "tier": null, "blocking_axis": null, "blocked_on": null, "axes": {}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:40170854"], "steps": ["read_published_estimate"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": ["xml"], "capabilities": [], "data_gaps": ["source:pmid:40170854"]}, "gaps": ["corpus_record_absent"]}, "review": {"state": "partial", "machine_critique": {"status": "blocked", "providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "full-text-reading-v1", "prompt_hash": "13cfecace36ca5eacfc9948f7a190ef4c271bebd396078ee232c9ae21eb3a600", "source_hash": "596b9414d75f7cf6f0ed015277de6a52168671c1094ba41e292b5490fdb5cf9e", "raw_response_sha256": "sha256:6aaf36f4a26d6c9773bcee69568562cc6412bc86cd232792eceac061ffdc8279", "raw_response_chars": 4570}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "full-text-reading-v1", "prompt_hash": "64bd5d0e3ab03f6e85d803eb0d2ba1a46e298ed4ae1613eb34094ef651a2bc36", "source_hash": "596b9414d75f7cf6f0ed015277de6a52168671c1094ba41e292b5490fdb5cf9e", "raw_response_sha256": "sha256:7f046cf59f4b27f352f60f2b2b85d0def2d72e2d6dd234d219718949bb85288a", "raw_response_chars": 37187}], "target_providers": [], "agreement": false, "disagreements": {"target_result": {"minimax": "The multi-omics Ridge regression model achieved the optimal performance in predicting pathological complete response (pCR), with an AUC of 0.917.", "minimax_alt": "Our results revealed that the multi-omics Ridge regression model achieved the optimal performance in predicting pCR, with an AUC of 0.917.", "decisive": false}, "analysis_methods": {"minimax": ["Lasso regression for feature extraction and selection", "Ridge regression with L2 regularization", "Principal Component Analysis", "Uniform Manifold Approximation and Projection", "Orange3 machine learning platform"], "minimax_alt": ["Lasso regression for feature extraction and selection", "Ridge regression with L2 regularization (C=1) for multi-omics pCR prediction", "Principal Component Analysis (PCA) for multi-omics dimensionality reduction", "Machine learning models implemented using the Orange3 platform (v3.32)", "AUC as the predictive performance metric"], "decisive": false}, "statistical_method": {"minimax": "Ridge regression with L2 regularization", "minimax_alt": "Ridge regression with L2 regularization (C=1)", "decisive": true}}, "agreement_basis": {"policy": "decisive_fields", "decisive_fields": ["work_type", "number_stated", "parent_value", "unit"]}, "blinded": true}, "human_critique": {"status": "not_run", "reviewer": null, "signed_at": null, "conflict_checked": false, "commentary": null}, "stages": [{"key": "target_materials", "status": "pending", "reason": null, "evidence": null}, {"key": "independent_readings", "status": "blocked", "reason": "The independent readings are incomplete or disagree; the source evidence must resolve the difference.", "evidence": {"provider_errors": [{"provider": "minimax", "model": null, "reason": "HTTP Error 429: Too Many Requests", "raw_response": null}, {"provider": "minimax_alt", "model": null, "reason": "HTTP Error 429: Too Many Requests", "raw_response": null}]}}, {"key": "source_resolution", "status": "pending", "reason": null, "evidence": null}, {"key": "protocol_reconstruction", "status": "pending", "reason": null, "evidence": null}, {"key": "control_execution", "status": "pending", "reason": null, "evidence": null}, {"key": "control_comparison", "status": "pending", "reason": null, "evidence": null}, {"key": "publication", "status": "pending", "reason": null, "evidence": null}], "gaps": ["review.human_critique", "review.machine_critique.blocked"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "package:f80994bf9aed6d2c2dda6fe1be1d4eeea16d20fe8cf57977685447bd5079954f", "type": "sep_package", "label": "SEP v1 package for pmid:40170854"}], "activities": [{"id": "activity:conversion:pmid:40170854", "type": "conversion", "used": ["package:f80994bf9aed6d2c2dda6fe1be1d4eeea16d20fe8cf57977685447bd5079954f"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:minimax", "type": "model", "label": "minimax"}, {"id": "agent:minimax_alt", "type": "model", "label": "minimax_alt"}, {"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}], "source_run_id": "run-20261023T122449232913", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "15274d38-ae86-4036-97bd-94f0b45e0dbf", "parent_hash": "f80994bf9aed6d2c2dda6fe1be1d4eeea16d20fe8cf57977685447bd5079954f", "fork_of": null, "generation": 1}, "gaps": ["abstract", "asset:source:pmid:40170854 not verified", "authors", "capability", "corpus_record_absent", "execution.not_run", "method", "review.human_critique", "review.machine_critique.blocked", "step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40170854' does not resolve to a declared asset", "targets"]}