Complete Tool Reference for Disease Information

Comprehensive reference of all ToolUniverse tools for disease information retrieval.

1. Disease Identification & Ontology

OSL_get_efo_id_by_disease_name

Purpose: Map disease name to EFO ID (primary entry point)

tu.tools.OSL_get_efo_id_by_disease_name(disease="diabetes mellitus")
# Returns: {"efo_id": "EFO:0000400", "name": "diabetes mellitus"}

ols_search_efo_terms

Purpose: Search EFO ontology for disease terms

tu.tools.ols_search_efo_terms(query="diabetes mellitus", rows=10)
# Returns: terms with iri, obo_id, label, description

ols_get_efo_term

Purpose: Get detailed EFO term information

tu.tools.ols_get_efo_term(obo_id="EFO:0000400")
# Returns: synonyms, description, has_children, is_obsolete

ols_get_efo_term_children

Purpose: Get disease subtypes/children

tu.tools.ols_get_efo_term_children(obo_id="EFO:0000400", size=20)
# Returns: child terms (disease subtypes)

OpenTargets_get_disease_id_description_by_name

Purpose: Search OpenTargets for disease by name

tu.tools.OpenTargets_get_disease_id_description_by_name(diseaseName="Diabetes Mellitus")
# Returns: id, name, description

umls_search_concepts

Purpose: Search UMLS for medical concepts

tu.tools.umls_search_concepts(query="diabetes", sabs="SNOMEDCT_US", pageSize=25)
# Returns: CUI, name, source
# Note: Requires UMLS_API_KEY

umls_get_concept_details

Purpose: Get UMLS concept details by CUI

tu.tools.umls_get_concept_details(cui="C0011849")
# Returns: definitions, semantic types

icd_search_codes

Purpose: Search ICD-10/ICD-11 codes

tu.tools.icd_search_codes(query="diabetes", version="ICD10CM")
# Returns: ICD codes with descriptions

snomed_search_concepts

Purpose: Search SNOMED CT concepts

tu.tools.snomed_search_concepts(query="diabetes mellitus")
# Returns: SNOMED concepts with codes

2. Clinical Manifestations & Phenotypes

OpenTargets_get_associated_phenotypes_by_disease_efoId

Purpose: Get HPO phenotypes for disease

tu.tools.OpenTargets_get_associated_phenotypes_by_disease_efoId(efoId="EFO_0000384")
# Returns: phenotypeHPO (id, name, description), phenotypeEFO

get_HPO_ID_by_phenotype (Monarch)

Purpose: Convert symptom name to HPO ID

tu.tools.get_HPO_ID_by_phenotype(query="seizure", limit=5)
# Returns: HPO IDs matching the phenotype

get_phenotype_by_HPO_ID (Monarch)

Purpose: Get phenotype details by HPO ID

tu.tools.get_phenotype_by_HPO_ID(id="HP:0001250")
# Returns: phenotype details

get_joint_associated_diseases_by_HPO_ID_list (Monarch)

Purpose: Find diseases from list of phenotypes (differential diagnosis)

tu.tools.get_joint_associated_diseases_by_HPO_ID_list(
    HPO_ID_list=["HP:0001250", "HP:0001251"], limit=20
)
# Returns: diseases associated with these phenotypes

MedlinePlus_search_topics_by_keyword

Purpose: Search consumer health information

tu.tools.MedlinePlus_search_topics_by_keyword(
    term="diabetes", db="healthTopics", rettype="topic"
)
# Returns: topics with title, summary, url

MedlinePlus_get_genetics_condition_by_name

Purpose: Get genetic condition information

tu.tools.MedlinePlus_get_genetics_condition_by_name(condition="alzheimer-disease")
# Returns: description, genes, synonyms

MedlinePlus_connect_lookup_by_code

Purpose: Look up by clinical code (ICD-10, LOINC)

tu.tools.MedlinePlus_connect_lookup_by_code(
    cs="2.16.840.1.113883.6.90",  # ICD-10 CM OID
    c="E11.9"  # Type 2 diabetes
)
# Returns: MedlinePlus health information

3. Genetic & Molecular Basis

OpenTargets_get_associated_targets_by_disease_efoId

Purpose: Get disease-gene associations with scores

tu.tools.OpenTargets_get_associated_targets_by_disease_efoId(efoId="EFO_0000384")
# Returns: target.id, target.approvedSymbol, score

OpenTargets_get_diseases_phenotypes_by_target_ensembl

Purpose: Find diseases associated with a gene (reverse lookup)

tu.tools.OpenTargets_get_diseases_phenotypes_by_target_ensembl(ensemblId="ENSG00000141510")
# Returns: diseases associated with this gene

OpenTargets_target_disease_evidence

Purpose: Get evidence for target-disease association

tu.tools.OpenTargets_target_disease_evidence(
    efoId="EFO_0000384", ensemblId="ENSG00000141510"
)
# Returns: evidence details, mutation data

clinvar_search_variants

Purpose: Search ClinVar for variants

tu.tools.clinvar_search_variants(condition="breast cancer", max_results=20)
# OR
tu.tools.clinvar_search_variants(gene="BRCA1", max_results=20)
# Returns: variant IDs, count

clinvar_get_variant_details

Purpose: Get variant details by ClinVar ID

tu.tools.clinvar_get_variant_details(variant_id="12345")
# Returns: variant information

clinvar_get_clinical_significance

Purpose: Get pathogenicity classification

tu.tools.clinvar_get_clinical_significance(variant_id="12345")
# Returns: clinical significance data

gwas_search_associations

Purpose: Search GWAS associations

tu.tools.gwas_search_associations(disease_trait="diabetes", size=20)
# Returns: associations with p_value, snp_allele, mapped_genes

gwas_get_variants_for_trait

Purpose: Get variants for specific trait

tu.tools.gwas_get_variants_for_trait(disease_trait="breast cancer", size=50)
# Returns: variants with rs_id, locations, mapped_genes

gwas_get_associations_for_trait

Purpose: Get associations sorted by significance

tu.tools.gwas_get_associations_for_trait(disease_trait="type 2 diabetes", size=20)
# Returns: associations sorted by p-value

gwas_get_studies_for_trait

Purpose: Get GWAS studies for trait

tu.tools.gwas_get_studies_for_trait(disease_trait="diabetes", size=20)
# Returns: study details, sample sizes

gwas_get_snp_by_id

Purpose: Get SNP details by rs ID

tu.tools.gwas_get_snp_by_id(rs_id="rs1234")
# Returns: SNP details, locations, alleles

gwas_get_associations_for_snp

Purpose: Get all associations for a SNP

tu.tools.gwas_get_associations_for_snp(rs_id="rs12345", size=20)
# Returns: traits associated with this SNP

gwas_get_snps_for_gene

Purpose: Get SNPs mapped to a gene

tu.tools.gwas_get_snps_for_gene(mapped_gene="BRCA1", size=20)
# Returns: SNPs in/near this gene

GWAS_search_associations_by_gene

Purpose: Search GWAS by gene name

tu.tools.GWAS_search_associations_by_gene(gene_name="TP53", size=10)
# Returns: associations for gene

gnomad_get_variant_frequency

Purpose: Get population variant frequencies

tu.tools.gnomad_get_variant_frequency(variant="1-55505647-G-T")
# Returns: population frequencies (gnomAD data)

4. Treatment Landscape

OpenTargets_get_associated_drugs_by_disease_efoId

Purpose: Get drugs for disease

tu.tools.OpenTargets_get_associated_drugs_by_disease_efoId(efoId="EFO_0000384", size=100)
# Returns: drug info, phase, status, mechanism, target

OpenTargets_get_drug_chembId_by_generic_name

Purpose: Get ChEMBL ID from drug name

tu.tools.OpenTargets_get_drug_chembId_by_generic_name(drugName="Aspirin")
# Returns: chemblId, name, description

OpenTargets_get_drug_mechanisms_of_action_by_chemblId

Purpose: Get drug mechanism of action

tu.tools.OpenTargets_get_drug_mechanisms_of_action_by_chemblId(chemblId="CHEMBL25")
# Returns: mechanism, actionType, targets

OpenTargets_get_drug_warnings_by_chemblId

Purpose: Get drug warnings

tu.tools.OpenTargets_get_drug_warnings_by_chemblId(chemblId="CHEMBL25")
# Returns: warningType, description, toxicityClass

OpenTargets_get_drug_blackbox_status_by_chembl_ID

Purpose: Check withdrawn/blackbox status

tu.tools.OpenTargets_get_drug_blackbox_status_by_chembl_ID(chemblId="CHEMBL25")
# Returns: hasBeenWithdrawn, blackBoxWarning

search_clinical_trials

Purpose: Search ClinicalTrials.gov

tu.tools.search_clinical_trials(
    condition="lung cancer",
    intervention="pembrolizumab",
    query_term="Phase III",
    pageSize=20
)
# Returns: NCT ID, brief_title, status, phase

get_clinical_trial_descriptions

Purpose: Get trial descriptions

tu.tools.get_clinical_trial_descriptions(
    nct_ids=["NCT04852770", "NCT01728545"],
    description_type="full"
)
# Returns: detailed trial descriptions

get_clinical_trial_conditions_and_interventions

Purpose: Get conditions and interventions

tu.tools.get_clinical_trial_conditions_and_interventions(
    nct_ids=["NCT01158625"],
    condition_and_intervention=""
)
# Returns: conditions, arm_groups, interventions

get_clinical_trial_eligibility_criteria

Purpose: Get eligibility criteria

tu.tools.get_clinical_trial_eligibility_criteria(
    nct_ids=["NCT01158625"],
    eligibility_criteria=""
)
# Returns: eligibility_criteria, sex, age range

get_clinical_trial_outcome_measures

Purpose: Get outcome measures

tu.tools.get_clinical_trial_outcome_measures(
    nct_ids=["NCT01158625"],
    outcome_measures="primary"
)
# Returns: primary/secondary outcomes

extract_clinical_trial_outcomes

Purpose: Extract efficacy results

tu.tools.extract_clinical_trial_outcomes(
    nct_ids=["NCT01158625"],
    outcome_measure="overall survival"
)
# Returns: detailed outcome results

extract_clinical_trial_adverse_events

Purpose: Extract safety data

tu.tools.extract_clinical_trial_adverse_events(
    nct_ids=["NCT01158625"],
    organ_systems=["Cardiac Disorders"],
    adverse_event_type="serious"
)
# Returns: adverse event data

5. Biological Pathways & Mechanisms

Reactome_get_diseases

Purpose: Get all disease-associated pathways

tu.tools.Reactome_get_diseases()
# Returns: disease pathways with DOID annotations

Reactome_get_pathway

Purpose: Get pathway details

tu.tools.Reactome_get_pathway(stId="R-HSA-73817")
# Returns: pathway metadata, events, references

Reactome_get_pathway_reactions

Purpose: Get reactions in pathway

tu.tools.Reactome_get_pathway_reactions(stId="R-HSA-73817")
# Returns: reactions and subpathways

Reactome_map_uniprot_to_pathways

Purpose: Get pathways for protein

tu.tools.Reactome_map_uniprot_to_pathways(id="P04637")
# Returns: pathways containing this protein

Reactome_map_uniprot_to_reactions

Purpose: Get reactions for protein

tu.tools.Reactome_map_uniprot_to_reactions(id="P04637")
# Returns: reactions involving this protein

Reactome_list_top_pathways

Purpose: List top-level pathways

tu.tools.Reactome_list_top_pathways(species="Homo sapiens")
# Returns: top-level pathway hierarchy

humanbase_ppi_analysis

Purpose: Tissue-specific protein interactions

tu.tools.humanbase_ppi_analysis(
    gene_list=["TP53", "MDM2"],
    tissue="brain",
    max_node=10,
    interaction="co-expression",
    string_mode=True
)
# Returns: PPI network, GO biological processes

gtex_get_expression_by_gene

Purpose: Get tissue-specific gene expression (GTEx)

tu.tools.gtex_get_expression_by_gene(gene="BRCA1")
# Returns: expression levels across tissues

HPA_get_protein_expression

Purpose: Get protein expression from Human Protein Atlas

tu.tools.HPA_get_protein_expression(gene="TP53")
# Returns: protein expression by tissue, subcellular localization

geo_search_datasets

Purpose: Search GEO for gene expression datasets

tu.tools.geo_search_datasets(query="Alzheimer disease", max_results=20)
# Returns: GEO dataset accessions, descriptions

6. Literature & Research

PubMed_search_articles

Purpose: Search biomedical literature

tu.tools.PubMed_search_articles(
    query='"Alzheimer disease" AND biomarker',
    limit=50
)
# Returns: PMIDs

PubMed_get_article

Purpose: Get article metadata

tu.tools.PubMed_get_article(pmid="12345678")
# Returns: title, abstract, authors, journal

PubMed_get_related

Purpose: Get related articles

tu.tools.PubMed_get_related(pmid="20210808", limit=20)
# Returns: related PMIDs

PubMed_get_cited_by

Purpose: Get citing articles

tu.tools.PubMed_get_cited_by(pmid="20210808", limit=20)
# Returns: PMIDs of citing articles

OpenTargets_get_publications_by_disease_efoId

Purpose: Get publications for disease

tu.tools.OpenTargets_get_publications_by_disease_efoId(efoId="EFO_0000384")
# Returns: disease-related publications

OpenTargets_get_publications_by_target_ensemblID

Purpose: Get publications for target

tu.tools.OpenTargets_get_publications_by_target_ensemblID(ensemblId="ENSG00000141510")
# Returns: target-related publications

openalex_search_works

Purpose: Search OpenAlex for works with institutional data

tu.tools.openalex_search_works(query="Alzheimer disease biomarker", limit=50)
# Returns: works with authors, institutions, citations, topics

europe_pmc_search_abstracts

Purpose: Search Europe PMC literature

tu.tools.europe_pmc_search_abstracts(query="Parkinson disease mechanism", limit=50)
# Returns: abstracts from Europe PMC

semantic_scholar_search_papers

Purpose: Search Semantic Scholar with citation networks

tu.tools.semantic_scholar_search_papers(query="cancer immunotherapy", limit=50)
# Returns: papers with citation counts, influential citations

7. Similar Diseases

OpenTargets_get_similar_entities_by_disease_efoId

Purpose: Find similar diseases, targets, drugs

tu.tools.OpenTargets_get_similar_entities_by_disease_efoId(
    efoId="EFO_0000249",
    threshold=0.5,
    size=20
)
# Returns: similar entities with scores

8. Cancer-Specific (CIViC)

civic_search_diseases

Purpose: Search cancer diseases

tu.tools.civic_search_diseases(limit=50)
# Returns: cancer diseases in CIViC

civic_search_genes

Purpose: Search cancer genes

tu.tools.civic_search_genes(query="BRAF", limit=10)
# Returns: gene id, name, description

civic_get_variants_by_gene

Purpose: Get variants for gene

tu.tools.civic_get_variants_by_gene(gene_id=5, limit=50)
# Returns: variants for gene

civic_get_variant

Purpose: Get variant details

tu.tools.civic_get_variant(variant_id=4170)
# Returns: variant details

civic_get_evidence_item

Purpose: Get clinical evidence

tu.tools.civic_get_evidence_item(evidence_id=116)
# Returns: evidence description, level, type

civic_search_therapies

Purpose: Search cancer therapies

tu.tools.civic_search_therapies(limit=50)
# Returns: therapy list

civic_search_molecular_profiles

Purpose: Search biomarker profiles

tu.tools.civic_search_molecular_profiles(limit=50)
# Returns: molecular profiles

9. Pharmacology (GtoPdb)

GtoPdb_list_diseases

Purpose: Search diseases

tu.tools.GtoPdb_list_diseases(name="diabetes", limit=20)
# Returns: diseases with IDs, OMIM, DOID

GtoPdb_get_disease

Purpose: Get disease details

tu.tools.GtoPdb_get_disease(disease_id=652)
# Returns: targets, ligands, description

GtoPdb_get_targets

Purpose: Get pharmacological targets

tu.tools.GtoPdb_get_targets(target_type="GPCR", limit=20)
# Returns: targets with drugs, ligands

GtoPdb_get_target

Purpose: Get target details

tu.tools.GtoPdb_get_target(target_id=290)
# Returns: detailed target info

GtoPdb_get_target_interactions

Purpose: Get target-ligand interactions

tu.tools.GtoPdb_get_target_interactions(
    target_id=290,
    action_type="Agonist"
)
# Returns: interactions with affinity

GtoPdb_search_interactions

Purpose: Search drug-target interactions

tu.tools.GtoPdb_search_interactions(
    approved_only=True,
    limit=100
)
# Returns: interaction data

GtoPdb_list_ligands

Purpose: Search ligands/drugs

tu.tools.GtoPdb_list_ligands(ligand_type="Approved", limit=20)
# Returns: ligands with properties

GtoPdb_get_ligand

Purpose: Get ligand details

tu.tools.GtoPdb_get_ligand(ligand_id=1016)
# Returns: SMILES, properties, targets

10. Protein Information (UniProt)

UniProt_get_disease_variants_by_accession

Purpose: Get disease-associated variants

tu.tools.UniProt_get_disease_variants_by_accession(accession="P05067")
# Returns: disease variants for protein

UniProt_get_function_by_accession

Purpose: Get protein function

tu.tools.UniProt_get_function_by_accession(accession="P05067")
# Returns: protein function description

UniProt_get_subcellular_location_by_accession

Purpose: Get protein localization

tu.tools.UniProt_get_subcellular_location_by_accession(accession="P05067")
# Returns: cellular location

11. Adverse Events

AdverseEventPredictionQuestionGenerator

Purpose: Generate safety questions

tu.tools.AdverseEventPredictionQuestionGenerator(
    disease_name="Alzheimer's disease",
    drug_name="Kisunla"
)
# Returns: safety prediction questions

AdverseEventICDMapper

Purpose: Map adverse events to ICD codes

tu.tools.AdverseEventICDMapper(
    source_text="Patient experienced headache and nausea"
)
# Returns: ICD-10 codes for adverse events

FAERS_count_reactions_by_drug_event

Purpose: Get FDA adverse event reports count

tu.tools.FAERS_count_reactions_by_drug_event(drug="metformin", event="nausea")
# Returns: count of adverse event reports from FAERS

ID Mapping Summary

From	To	Tool
Disease name	EFO ID	`OSL_get_efo_id_by_disease_name`
Disease name	EFO ID	`OpenTargets_get_disease_id_description_by_name`
Drug name	ChEMBL ID	`OpenTargets_get_drug_chembId_by_generic_name`
Gene symbol	Ensembl ID	Use OpenTargets search
UniProt ID	Pathways	`Reactome_map_uniprot_to_pathways`
Symptom	HPO ID	`get_HPO_ID_by_phenotype`
HPO IDs	Diseases	`get_joint_associated_diseases_by_HPO_ID_list`
Gene	Diseases	`OpenTargets_get_diseases_phenotypes_by_target_ensembl`
SNP rs ID	Diseases	`gwas_get_associations_for_snp`

Query Construction Tips

PubMed Queries

Good query construction:

# Specific disease + topic
query = '"Alzheimer disease" AND mechanism'

# Multiple terms with OR
query = '"Parkinson disease" OR "Parkinson\'s disease" AND therapy'

# Exclude terms
query = '"diabetes" NOT "gestational diabetes" AND treatment'

# Recent papers only
query = '"cancer" AND immunotherapy'
arguments = {'query': query, 'years': 2}  # Last 2 years

Field-specific searches:

# Title only
query = 'Alzheimer[Title] AND biomarker[Title]'

# MeSH terms
query = '"Alzheimer Disease"[MeSH] AND "Drug Therapy"[MeSH]'

# Publication types
query = '"diabetes" AND systematic review[Publication Type]'

OpenTargets Queries

Disease ID formats:

EFO IDs: EFO_0000249 (Alzheimer's)
Orphanet: Orphanet_558 (rare diseases)
MONDO: MONDO_0008199

Finding disease IDs:

# Search by name
result = tu.tools.OSL_get_efo_id_by_disease_name(disease='Alzheimer disease')
efo_id = result.get('efo_id')  # Get EFO ID

Clinical Trials Queries

Effective search strategies:

# By condition
{'condition': 'Alzheimer Disease'}

# By intervention
{'condition': 'cancer', 'intervention': 'pembrolizumab'}

# By phase
{'condition': 'diabetes', 'query_term': 'Phase 3'}

# By status
{'condition': 'depression', 'status': 'Recruiting'}

Common Issues & Solutions

Issue: Disease name vs EFO ID mismatch

Solution: Always try to get both

if disease_name and not disease_id:
    # Get EFO ID from name
    result = tu.tools.OSL_get_efo_id_by_disease_name(disease=disease_name)
    disease_id = result.get('efo_id')
elif disease_id and not disease_name:
    # Get name from EFO ID
    result = tu.tools.OpenTargets_get_disease_id_description_by_name(efoId=disease_id)
    disease_name = result.get('name')

Issue: Empty results from a tool

Solution: Try alternative tools or queries

targets = tu.tools.OpenTargets_get_associated_targets_by_disease_efoId(efoId=disease_id)
if not targets.get('data'):
    # Try PubMed text mining as fallback
    pmids = tu.tools.PubMed_search_articles(query=f'"{disease_name}" AND gene')

Issue: Timeout on slow queries

Solution: Set appropriate timeouts and handle gracefully

try:
    result = future.result(timeout=120)  # 2 minutes
except TimeoutError:
    result = {'status': 'timeout', 'message': 'Query too slow'}

Issue: Rate limiting

Solution: Add delays or use caching

import time
from functools import lru_cache

@lru_cache(maxsize=100)
def cached_query(tool_name, args_json):
    # Cache results to avoid repeated queries
    import json
    return tu.run({'name': tool_name, 'arguments': json.loads(args_json)})

Performance Optimization

Parallel Execution Best Practices

# Good: Independent paths in parallel
with ThreadPoolExecutor(max_workers=5) as executor:
    futures = {
        'path1': executor.submit(path1_func),
        'path2': executor.submit(path2_func),
        # All independent
    }

# Bad: Dependent queries in parallel
# Don't parallelize if path2 needs path1 results

Result Limiting

# Limit results to avoid overwhelming output
top_targets = targets['data'][:10]  # Top 10 only
top_pathways = pathways['data'][:5]  # Top 5 only
top_drugs = drugs['data'][:5]  # Top 5 only

Caching Strategy

# Cache expensive queries
cache = {}

def get_gene_info(gene_id):
    if gene_id in cache:
        return cache[gene_id]
    
    result = tu.tools.UniProt_get_entry_by_accession(accession=gene_id)
    cache[gene_id] = result
    return result

Data Quality Indicators

Track data quality in your synthesis:

quality_metrics = {
    'sources_queried': 15,  # How many tools used
    'sources_successful': 12,  # How many returned data
    'completeness_score': 0.80,  # 80% of paths succeeded
    'data_recency': {
        'publications': '2024',  # Most recent paper
        'trials': '2024',  # Most recent trial
        'approval': '2023'  # Most recent drug approval
    }
}

Include in report:

Data Quality: ⭐⭐⭐⭐ (80% complete, 12/15 sources)
Most recent data: 2024

FilesExpand file tree

TOOLS_REFERENCE.md

Latest commit

History