Datasets:

Sidhantyadav
/

OpenBioMed-Evidence

Error code:   DatasetGenerationError
Exception:    TypeError
Message:      Couldn't cast array of type
struct<assumption: string, category: string, impact_if_violated: string, provenance_tag: string, impact_if_violated: : string>
to
{'assumption': Value('string'), 'category': Value('string'), 'impact_if_violated': Value('string'), 'provenance_tag': Value('string')}
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1835, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 621, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 361, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2381, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2315, in cast_table_to_schema
                  cast_array_to_feature(
                  ~~~~~~~~~~~~~~~~~~~~~^
                      table[name] if name in table_column_names else pa.array([None] * len(table), type=schema.field(name).type),
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                      feature,
                      ^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1861, in wrapper
                  return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
                                           ~~~~^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2071, in cast_array_to_feature
                  _c(children[name] if name in array_fields else null_array, subfeature)
                  ~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1863, in wrapper
                  return func(array, *args, **kwargs)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2121, in cast_array_to_feature
                  casted_array_values = _c(array.values, feature.feature)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 1863, in wrapper
                  return func(array, *args, **kwargs)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2161, in cast_array_to_feature
                  raise TypeError(f"Couldn't cast array of type\n{_short_str(array.type)}\nto\n{_short_str(feature)}")
              TypeError: Couldn't cast array of type
              struct<assumption: string, category: string, impact_if_violated: string, provenance_tag: string, impact_if_violated: : string>
              to
              {'assumption': Value('string'), 'category': Value('string'), 'impact_if_violated': Value('string'), 'provenance_tag': Value('string')}
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1379, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 956, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1702, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

record_id string	schema_version string	source_identifiers dict	abstract dict	extraction dict	cited_references list	verification_audit dict	extracted_at string
med_rec_PMC13475465	biology_schema_synth_v1.4	{ "title": "Exploring variables leading to major postmarketing label changes in leading regulatory authorities: a retrospective cohort study in Israel.", "authors": [ "Vishkautzan A", "Vexberg MH", "Berkovitch M", "Ashkenazi S", "Weiss I", "Hershkowitz R", "Gorelik E", "Maayan H", ...	{ "full_text": "Objective: The US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) are leading benchmarks for reliance approval pathways, while many regulatory authorities streamline their review processes by referencing these authorities’ evaluations and decisions. Major postmarketing (MPM)...	{ "background_and_problem_statement": { "prior_state_of_knowledge": "National health regulatory authorities such as the Israel Ministry of Health (MOH) commonly employ reliance pathways referencing benchmark agencies—specifically the US Food and Drug Administration (FDA) and European Medicines Agency (EMA)—to acc...	[ { "ref_id": "R1", "ref_number": "1", "pmid": "28342075", "pmcid": "", "doi": "", "title": "Evaluation of Pre-marketing Factors to Predict Post-marketing Boxed Warnings and Safety Withdrawals.", "authors": [ "Schick A", "Miller KL", "Lanthier M", "Dal Pan G", ...	{ "total_evidence_quotes": 21, "verified_verbatim_quotes": 21, "grounding_fidelity_percent": 100, "numerical_drift_detected": false, "unmatched_numbers": [], "cited_references_parsed_count": 20, "abstract_present": true, "ci_sign_checks_passed": true, "source_conflicts": [], "schema_version_migrated...	2026-11-10T15:38:06.973249+00:00
med_rec_PMC13598462	biology_schema_synth_v1.4	{ "title": "Global, regional, and national prevalence of second-hand smoke and attributable disease burden in 204 countries and territories, 1990-2023: a systematic analysis for the Global Burden of Disease Study 2023.", "authors": [ "GBD 2023 Second-hand smoke Collaborators." ], "affiliations": [], "pmid...	{ "full_text": "Background: Second-hand smoke (SHS) exposure remains a major source of morbidity and mortality among non-smokers. This study presents the first dedicated Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) publication to systematically quantify SHS prevalence and evaluate its effect as a...	{ "background_and_problem_statement": { "prior_state_of_knowledge": "Second-hand smoke (SHS) combines unfiltered, incompletely combusted sidestream smoke with exhaled mainstream smoke and contains higher per-unit concentrations of toxicants and carcinogens than active smoking. There is no safe level of exposure, ...	[ { "ref_id": "R1", "ref_number": "1", "pmid": "39537242", "pmcid": "", "doi": "", "title": "Association between second-hand smoke exposure and lung cancer risk in never-smokers: a systematic review and meta-analysis.", "authors": [ "Possenti I", "Romelli M", "Carreras G"...	{ "total_evidence_quotes": 17, "verified_verbatim_quotes": 17, "grounding_fidelity_percent": 100, "numerical_drift_detected": false, "unmatched_numbers": [], "cited_references_parsed_count": 42, "abstract_present": true, "ci_sign_checks_passed": true, "source_conflicts": [], "schema_version_migrated...	2026-11-10T15:38:06.994700+00:00
med_rec_PMC13575558	biology_schema_synth_v1.4	{ "title": "Biomarker Analysis from Patients with Metastatic PDAC Treated with TGFβ Antibody NIS793 plus Abraxane + Gemcitabine versus Abraxane + Gemcitabine Alone in a Phase II, Open-Label, Randomized Study.", "authors": [ "Pelletier M", "Yang J", "Joshi M", "Grauel A", "Abecunas C", "Altze...	{ "full_text": "Purpose: Transforming growth factor β (TGFβ) plays a dual role in cancer, acting as a tumor suppressor early in the disease but promoting progression and immune evasion when dysregulated. In pancreatic ductal adenocarcinoma (PDAC), TGFβ-driven desmoplasia fosters chemoresistance and immunosuppression,...	{ "background_and_problem_statement": { "prior_state_of_knowledge": "Transforming growth factor β (TGFβ; isoforms TGFβ1–3) acts early as an epithelial tumor suppressor but later drives desmoplasia, cancer-associated fibroblast (CAF) activation, extracellular matrix (ECM) deposition, angiogenesis, and T-cell exclu...	[ { "ref_id": "R1", "ref_number": "1", "pmid": "23396013", "pmcid": "", "doi": "", "title": "Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples.", "authors": [ "Cibulskis K", "Lawrence MS", "Carter SL", "Sivachenko A", "J...	{ "total_evidence_quotes": 19, "verified_verbatim_quotes": 19, "grounding_fidelity_percent": 100, "numerical_drift_detected": false, "unmatched_numbers": [], "cited_references_parsed_count": 43, "abstract_present": true, "ci_sign_checks_passed": true, "source_conflicts": [], "schema_version_migrated...	2026-11-10T15:38:07.015296+00:00
med_rec_PMC13424262	biology_schema_synth_v1.4	{"title":"CYP2C19 Genotyping to Guide Clopidogrel Prescribing for Neurovascular Indications: A Surve(...TRUNCATED)	{"full_text":"Dual antiplatelet therapy (DAPT) with aspirin and clopidogrel is commonly used for sec(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Dual antiplatelet therapy (DAPT) wi(...TRUNCATED)	[{"ref_id":"R1","ref_number":"1","pmid":"19106084","pmcid":"","doi":"","title":"Cytochrome p-450 pol(...TRUNCATED)	{"total_evidence_quotes":18,"verified_verbatim_quotes":18,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.027819+00:00
med_rec_PMC13468982	biology_schema_synth_v1.4	{"title":"Medication-related examination questions in clinical courses within undergraduate medical (...TRUNCATED)	{"full_text":"Purpose: Pharmacotherapy is a core competence for physicians and students’ learning (...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Pharmacotherapy is a core clinical (...TRUNCATED)	[{"ref_id":"R1","ref_number":"1","pmid":"31454058","pmcid":"","doi":"","title":"[A new international(...TRUNCATED)	{"total_evidence_quotes":22,"verified_verbatim_quotes":22,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.044571+00:00
med_rec_PMC13689281	biology_schema_synth_v1.4	{"title":"Unintended Pregnancies and Breastfeeding Studies during Central Nervous System Drug Develo(...TRUNCATED)	{"full_text":"Pregnant and breastfeeding individuals are routinely excluded from clinical drug devel(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Pregnant and breastfeeding individu(...TRUNCATED)	[{"ref_id":"R1","ref_number":"1","pmid":"26488521","pmcid":"","doi":"","title":"ACOG Committee Opini(...TRUNCATED)	{"total_evidence_quotes":21,"verified_verbatim_quotes":21,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.067932+00:00
med_rec_PMC13574604	biology_schema_synth_v1.4	{"title":"Predicting clinically significant medication errors: Development and validation of a risk (...TRUNCATED)	{"full_text":"Background: Medication error reporting systems capture large volumes of incidents, yet(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Medication errors drive roughly $42(...TRUNCATED)	[{"ref_id":"B1","ref_number":"1","pmid":"","pmcid":"","doi":"","title":"World Health Organization Me(...TRUNCATED)	{"total_evidence_quotes":23,"verified_verbatim_quotes":23,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.090576+00:00
med_rec_PMC13520735	biology_schema_synth_v1.4	{"title":"Effect of Evolocumab on Grafts Post-Coronary Artery Bypass Surgery: Post-Hoc Analysis of N(...TRUNCATED)	{"full_text":"Background: Graft failure remains a significant clinical problem.\nObjectives: This st(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Coronary artery bypass graft (CABG)(...TRUNCATED)	[{"ref_id":"R1","ref_number":"1","pmid":"41211925","pmcid":"","doi":"","title":"Evolocumab in Patien(...TRUNCATED)	{"total_evidence_quotes":24,"verified_verbatim_quotes":24,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.110235+00:00
med_rec_PMC13541161	biology_schema_synth_v1.4	{"title":"Patterns in biomedical research funding in India: An analysis of ICMR small and intermedia(...TRUNCATED)	{"full_text":"Background and objectives: The Indian Council of Medical Research (ICMR) funds small a(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"India faces a distinctive epidemiol(...TRUNCATED)	[{"ref_id":"ref1","ref_number":"1","pmid":"25664236","pmcid":"PMC4318649","doi":"10.1016/j.gheart.20(...TRUNCATED)	{"total_evidence_quotes":18,"verified_verbatim_quotes":18,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.123973+00:00
med_rec_PMC13708324	biology_schema_synth_v1.4	{"title":"MDL-001: An Oral, Direct-Acting Broad-Spectrum Antiviral for Influenza-like Illness (ILI) (...TRUNCATED)	{"full_text":"Background/Objectives: Endemic viral respiratory illnesses caused by influenza, RSV, a(...TRUNCATED)	{"background_and_problem_statement":{"prior_state_of_knowledge":"Viral respiratory illnesses (VRIs)(...TRUNCATED)	[{"ref_id":"B1-pharmaceuticals-19-01639","ref_number":"1","pmid":"40003946","pmcid":"PMC11855616","d(...TRUNCATED)	{"total_evidence_quotes":26,"verified_verbatim_quotes":26,"grounding_fidelity_percent":100.0,"numeri(...TRUNCATED)	2026-11-10T15:38:07.149458+00:00

End of preview.

OpenBioMed-Evidence

OpenBioMed-Evidence is a verbatim-grounded biomedical and clinical research evidence corpus structured for scientific literature analysis, quantitative evidence synthesis, and domain-specific reasoning.

Each record represents a single peer-reviewed empirical research publication decomposed into canonical bibliographic metadata, structured abstracts, study methodology, quantitative outcomes, mechanistic pathways, explicit assumptions, cited-literature synthesis, and source quotations subjected to automated matching and numerical consistency checks.

Release Note: Dataset records are curated incrementally in validated collection cycles; additional records are appended as curation progresses.

Domain Scope

Coverage spans primary empirical research across:

Clinical Pharmacology & Therapeutics: Dosing regimens, PK/PD endpoints, adverse events, drug–drug precautions, and postmarketing pharmacovigilance.
Oncology, Immunology & Translational Medicine: Clinical trials, biomarker stratification, signaling pathways, and molecular assays.
Epidemiology, Neurology, Cardiology & Toxicology: Cohort/comparative-risk studies, regression modeling, effect sizes, and safety thresholds.

Record Structure & Schema (`biology_schema_synth_v1.3`)

Every JSONL record conforms to the formal JSON Schema defined in schema.json (biology_schema_synth_v1.3) across five primary sections:

source_identifiers: Canonical pmid, pmcid, doi, title, authors, affiliations, journal, publication_date, download_url, pdf_url, license, cited_by_count, grants_and_funding, trial_and_data_accessions (e.g., NCT, GSE), keywords_and_mesh, and domain_category.
abstract: Complete full_text abstract alongside original section splits (sections).
extraction:
- background_and_problem_statement — Prior state of knowledge, identified gap, formal hypothesis, scientific rationale, and exact verbatim_quote.
- methodology_and_study_design — Study classification, population/model criteria, sample size & arm allocation, interventions/dosing/comparators, assay protocols, statistical software/models, and verbatim_quotes.
- quantitative_measurements_and_statistics — List of measured metrics with arm/subgroup comparators, exact value_with_units, baseline/reference values, statistical_significance (p-values, CIs, HRs/ORs/RRs), interpretation, and verbatim_quote.
- key_results_and_resolution — Primary outcomes, secondary/subgroup findings, negative/null results, mechanistic pathway, translational implications, and verbatim_quotes.
- precautions_limitations_and_safety — Graded adverse events/toxicities, contraindications/precautions, study limitations, confounders/biases, and verbatim_quote.
- underlying_assumptions — Structured assumptions categorized by domain (biological_surrogate, model_translatability, statistical_model, measurement_validity, clinical_generalizability) paired with impact_if_violated.
- referenced_papers_and_literature_synthesis — Contextual links keyed by ref_id into cited_references, specifying relationship_to_current_study, detailed_synthesis, and supporting verbatim_quote.
- biomedical_entities — Normalized lists of drugs/compounds, genes/proteins/pathways, diseases/phenotypes, and organisms/models/cell lines.
cited_references: Complete structured bibliography (ref_id, ref_number, pmid, pmcid, doi, title, authors, journal, year, volume, pages, raw_citation).
verification_audit: Quantitative integrity metrics per record (verified_verbatim_quotes, total_evidence_quotes, grounding_fidelity_percent, numerical_drift_detected, unmatched_numbers, cited_references_parsed_count, abstract_present).

Data Quality Notes

Initial manual audits identified limitations that automated verification may not detect, including lost negative signs in uncertainty intervals, conflicting values between abstracts and main-text tables, and causal wording that overstates observational findings. Therefore, 100% verbatim quote matching does not guarantee numerical-semantic accuracy or scientific validity. Records should be checked against their original publications before downstream use.

Licensing

Because OpenBioMed-Evidence aggregates open-access publications from multiple publishers and appends new records incrementally, no single blanket license covers all underlying source publications. See LICENSE for full details.

Each record preserves its authoritative source publication license in source_identifiers.license (such as cc by, cc by-nc, cc by-nc-sa, and cc by-nc-nd) alongside canonical attribution (title, authors, journal, doi, pmid, pmcid, download_url).

Downstream users should respect each record's license terms and filter on source_identifiers.license if a uniform subset is needed:

# Example: filter to commercially permissive CC BY records only
cc_by_records = [r for r in records if r["source_identifiers"]["license"].lower() == "cc by"]

Downloads last month: 124

Total file size:

69.7 MB

OpenBioMed-Evidence

Domain Scope

Record Structure & Schema (biology_schema_synth_v1.3)

Data Quality Notes

Licensing

Record Structure & Schema (`biology_schema_synth_v1.3`)