Some Examples

Links below use the API v1 (deprecated) and HTTP GET (query parameters are URL-encoded)

1. Find something by keyword:"FGFR2"
2. Find pathways by keyword:"CALCRL"
3. Find protein references by xref id
4. Pagination: find everything, the second result page
5. Find Control interactions that contain "binding" but not "transcription" word
6. Find the interactions which pathway contains a keyword:"immune*" ('*' means any ending)
7. Find all Reactome or PID pathways
8. Get a ProteinReference (Q06609) by absolute URI and convert to JSON-LD format
9. Find, fetch by bio identifier (FGFR2) and convert to GSEA format (it first performs a full-text search 'xrefid:FGFR2' to find bioentity URIs, then extracts the sub-model and converts to GSEA GMT format
10. Get 'Signaling by BMP' Pathway (format: BioPAX; see also: R-HSA-201451)
11. Nearest neighborhood of the protein reference bioregistry.io/uniprot:P20908 (COL5A1), all reactions where the corresponding protein forms/states take part, converted to SIF format
12. Nearest neighborhood query using the gene symbol COL5A1 (P20908) instead of absolute URI (performs full-text search and id-mapping internally; such query potentially returns a larger sub-network as it might start from more matching entities rather than from one ProteinReference; One can mix URIs along with UniProt, NCBI Gene, ChEBI IDs in a single graph or get/fetch query; other identifiers might also work; see: about URIs and IDs)
13. Traverse to find the display name of the organism of the ProteinReference specified by the URI
14. Traverse to find the URI of the organism for each of the Protein References
15. Traverse to list the names of all states of RAD51 protein (by its ProteinReference URI, using property path="ProteinReference/entityReferenceOf:Protein/name")
16. Traverse to list the URIs of states of BRCA1_HUMAN
17. Traverse to list the names of several different objects (using abstract type 'Named' from Paxtools API)
18. Find the top-level pathways related to 'TP53'
19. Find the top pathways from Reactome, matching 'insulin'; request JSON format

Parameter Values

About URIs or IDs

Query parameters: source, uri, target accept an array of URIs of the BioPAX elements, which look like bioregistry.io/<prefix>:<identifier> (reference bio entities and vocabularies) or pc14:<unique_id> (bio processes and participants). BioPAX URIs are not to guess; instead, they should be discovered with /search or /top_pathways queries. However, IDs from the following gene/protein or small molecule collections are also acceptable in place of URIs in the fetch and graph queries:

• HUGO Gene Symbol
• NCBI Gene ID
• Uniprot AC
• RefSeq
• Ensembl
• ChEBI
• ChEMBL
• KEGG Compound
• DrugBank
• PharmGKB Drug
• PubChem Compound/Substance (must prefix with CID: or SID:, respectively, to distinguish from each other and Gene ID)

As a rule, using full URIs makes a precise query, whereas using the IDs makes a more exploratory one, which also depends on the full-text search (index) and id-mapping.

We integrated multiple data sources and consistently normalized Xref, EntityReference, Provenance, BioSource, and ControlledVocabulary objects. We did not merge physical entities (states) and processes from different sources automatically, as accurately matching and aligning pathways at that level is still an open research problem.

About Enumerations

Note that enumeration parameter values, such as those for Type, Format, Direction, Limit, Pattern, are case-insensitive and can also use e.g. "interacts-with" instead of "INTERACTS_WITH" (both '-' and '_' are acceptable), "jsonld" instead "JSONLD", "neighborhood" instead of "NEIGHBORHOOD", etc.