Static data API
There is no server behind raremap. Everything the tools use is a JSON file under /data/, and you can fetch the same files. No key, no rate limit beyond the web server's, no tracking.
1
Files
sizes from the current build| Path | Size | Transfer |
|---|---|---|
| /data/index.json | 1.18 MB | 0.30 MB gzip |
| /data/xref.json | 1.59 MB | 0.24 MB gzip |
| /data/genes.json | 0.92 MB | 0.26 MB gzip |
| /data/hpo/terms.json | 1.92 MB | 0.48 MB gzip |
| /data/hpo/annotations.json | 0.97 MB | 0.29 MB gzip |
| /data/raremap-data.json.gz | 17.04 MB | 17.04 MB gzip |
| /data/names/en.json | 1.14 MB | 0.29 MB gzip |
| /data/names/fr.json | 1.23 MB | 0.30 MB gzip |
| /data/names/de.json | 1.13 MB | 0.31 MB gzip |
| /data/names/es.json | 1.23 MB | 0.30 MB gzip |
| /data/names/it.json | 1.14 MB | 0.29 MB gzip |
| /data/names/nl.json | 1.28 MB | 0.32 MB gzip |
| /data/names/pt.json | 0.92 MB | 0.24 MB gzip |
| /data/names/pl.json | 1.22 MB | 0.34 MB gzip |
| /data/names/cs.json | 1.27 MB | 0.34 MB gzip |
| /data/names/uk.json | 1.91 MB | 0.37 MB gzip |
| /data/names/zh.json | 0.70 MB | 0.21 MB gzip |
| /data/names/tr.json | 1.00 MB | 0.27 MB gzip |
| /data/d/<orpha>.json | 11,645 files | 72.5 MB total, median 6.0 kB |
| /data/d/ORPHA-<n>/index.html | 11,645 files | 137.3 MB total, median 11.2 kB |
2
One disorder
GET /data/d/<orpha>.json — the full record for one ORPHAcode, e.g. /data/d/558.json. Fields:
| orpha, name, type, group, status, flags | ORPHAcode (integer), English preferred name, Orphanet disorder type and group; status is active, inactive, obsolete, deprecated or historical (from Orphanet's flags). |
|---|---|
| synonyms, definition, note | English synonyms and definition; note is Orphanet's automatic text for moved entries. |
| names | {lang: {name, synonyms, definition}} for fr de es it nl pt pl cs uk zh tr, where Orphadata ships them. |
| xrefs | [{source, code, qualifier, icd_relation, validated, icd11_uri?}]. source: OMIM, ICD-10, ICD-11, MONDO, MeSH, UMLS, GARD, MedDRA. qualifier: E, NTBT, BTNT, ND (W if Orphanet ever ships it). icd_relation: specific, attributed, index, inclusion. |
| associations, incoming_associations | Orphanet's "Moved to" / "Referred to" links from and to this code. |
| preferential_parent, parents, children | Linearisation parent (product 7) and classification parents/children (product 3), parents with the classification ids they come from (names in meta.json). |
| genes, genes_via_subtypes | [{symbol, name, gene_type, association, status, source, hgnc, ensembl, omim, loci}]. genes_via_subtypes lists genes Orphanet records on this disorder's subtypes, each with via. |
| inheritance, onset | Lists of Orphanet labels (product 9). |
| prevalence | [{type, qualification, class, value, area, status, source}]; value is Orphadata's ValMoy. |
| phenotypes | {validation_status, validation_date, source, terms: [{hpo, label, frequency, criteria}]} (product 4). |
3
Bulk files
/data/index.json—{types, statuses, d: [[orpha, name, type_index, status_index, synonyms?], …]}, every ORPHAcode./data/xref.json—{sources, icd_relations, d: {orpha: [[[source_index, code, qualifier, icd_relation_index, validated], …], [[assoc_type, target_orpha], …], group]}}./data/names/<lang>.json—{lang, d: [[orpha, name, synonyms?], …]}for en fr de es it nl pt pl cs uk zh tr. Note the dates inmeta.json: Orphadata's uk, zh and tr files are older than the others./data/genes.json—{association_types, g: {SYMBOL: {name, hgnc, ensembl, omim, synonyms, d: [[orpha, association_index, assessed], …]}}}./data/hpo/terms.json— HPO{version, t: [[id, label, synonyms, is_a_parents], …], alt_ids, replaced_by}; integer ids meanHP:%07d. Labels, synonyms and is_a edges as released by the HPO; obsolete terms are left out./data/hpo/annotations.json—{frequencies, weights, n_disorders, d: [[orpha, [hpo…], [frequency_index…]], …], ic: [[hpo, information_content], …]}./data/raremap-data.json.gz— everything above in one gzip file (what the Python package reads)./data/meta.json— source URLs, HTTP dates, SHA-256 of every downloaded file, counts, file sizes.
4
Examples
curl -s https://raremap.alftech.space/data/d/558.json | jq '.xrefs[] | select(.source=="OMIM")'
# every ORPHAcode that Orphanet maps to ICD-10 Q87.4, with its qualifier
curl -s https://raremap.alftech.space/data/xref.json | jq -r '
.sources as $s | .d | to_entries[] | .key as $o | .value[0][]
| select($s[.[0]]=="ICD-10" and .[1]=="Q87.4") | "ORPHA:\($o)\t\(.[2])"'
// JavaScript, in a page
const rec = await (await fetch('https://raremap.alftech.space/data/d/586.json')).json();
console.log(rec.name, rec.genes.map(g => g.symbol));
5
Python package
pure Python 3.8+, standard library onlyThe wheel bundles the same data, so it works offline after installing.
pip install https://raremap.alftech.space/dl/raremap-2026.6.0-py3-none-any.whl
import raremap
rows = raremap.crosswalk(["OMIM:154700", "Q87.4", "Mukoviszidose"]) # list of dicts, one per (input, ORPHAcode)
rec = raremap.disease("ORPHA:558") # the same record as /data/d/558.json
res = raremap.differential(["HP:0001166", "HP:0001083", "HP:0002616", "HP:0000098"])
for r in res["results"][:3]:
print(r["orpha"], r["name"], r["score"], r["explanation"])
print(raremap.info()) # data versions
raremap crosswalk codes.txt > out.csv # one identifier per line; --format tsv|excel|json
raremap disease ORPHA:586 # JSON record
raremap differential HP:0001166 HP:0001083 -n 5
raremap search marfan
raremap hpo arachnodactyly
raremap info
Data lookup order: $RAREMAP_DATA, the copy inside the wheel, ~/.cache/raremap/, then a one-time download of /data/raremap-data.json.gz.
6
Terms of use
Orphanet data: CC BY 4.0; credit "Orphanet (Orphadata Science), INSERM" and give the version from meta.json. HPO: use with credit and the version shown; do not modify its terms or relationships (licence). raremap is not an Orphanet product and was not validated by Orphanet/INSERM. Nothing here is medical advice.