The Open PHACTS API returns several URLs, but not all of them resolve to a useful page. Why is this the case?
We use these URLs (or URIs, uniform resource identifiers) as identifiers for concepts in the data. Some of these can directly be used to access the data of the original datasource (e.g. http://purl.uniprot.org/uniprot/Q9Y5Y9), but others will return data in rdf format (e.g. http://rdf.ebi.ac.uk/resource/chembl/target/CHEMBL5451). You can still access a readable page by replacing the link in many cases (e.g. https://www.ebi.ac.uk/chembl/target/inspect/CHEMBL5451 for ChEMBL targets). Or you can use these URIs directly as input for other API calls.
Also some information is given in the format of a URI. The provenance of the data for example is returned in such a way with the "inDatasource" property. This can, again, give the link of the actual data provider (e.g. http://www.ebi.ac.uk/chembl), or it links to metadata about the data set (e.g. http://rdf.imim.es/disgenet-void.ttl#gene).