Frequently asked questions
What is NAP?
NAP stands for Natural Actives & Pharmacology. It is an open database of ethnomedical and ethnopharmacological information. It collects, from the published literature, which organisms and which compounds were tested, how they were tested and what came out of it.
The data comes from a subset of the Napralert database. It has been converted to RDF and is served from a SPARQL endpoint, which is why the pages you see are mostly views over graph data and why some records still carry the quirks of the original data entry system.
What is the difference between an organism and a taxon?
An organism is a specimen record. It describes the material a given publication worked with: the name the authors used, the part that was collected, the condition of that part, where it was collected and which citation it belongs to. It is a record about one batch of material, not about a species.
A taxon is a normalized name in a taxonomic hierarchy. It has a taxonomic level (class, family, genus, species, subspecies) and a link to its parent taxon. Taxa were added later to reconcile the names that had been typed record by record.
Many organism records point to the same taxon. So if you want everything that is known about a species, you want the taxon; if you want to know what exactly was tested in one paper, you want the organism.
Two practical consequences. First, an organism can be a mixture of several organisms, because that is what some publications tested. Second, some organisms are empty: the original entry system put the organism at the center of every record, so a placeholder was created even when there was no organism to describe.
Why does the search box say Taxon and not Organism?
Because that is what the name search returns. It looks for the text you typed in the organism records, then collapses the matches to one row per taxon, so you get one line per name instead of one line per specimen.
Clicking a result takes you to all the experiments recorded for that taxon, not to a single specimen page. Individual organism pages are still reachable from the experiments and from the citations they belong to.
What is a pharmacy record?
It is one experiment. The name is inherited from the original database and it is a bit misleading: a pharmacy record is the link between a biological source, a tested material, a type of work, a measured activity and a result. Most of the other entity types exist to describe one field of a pharmacy record.
What are the different entities?
| Entity | What it holds |
|---|---|
| Citation | The publication a record was extracted from: authors, title, journal, year, DOI when we have it. Everything in NAP ultimately hangs off a citation. |
| Organism | The material that was actually studied in a publication, as it was recorded: name given by the authors, collected part, condition of that part, country or geographical area of collection. |
| Taxon | A normalized name in a taxonomic hierarchy, with its level (class, family, genus, species, subspecies) and a link to its parent. Many organism records point to the same taxon. |
| Compound | A chemical entity, with its compound class, its publication name and code, its synonyms and its compound code. |
| Pharmacy | One experiment or observation. This is the central record: it ties together an organism, a compound or an extract, the type of work done, the activity measured, the animal model, the administration route, the dose and the result. |
| Pharmacology | The activity, target or assay endpoint a pharmacy record measured. |
| Worktype | The kind of work that was done, for example a chemical isolation or a biological assay. Worktypes are organized in groups. |
| Extract | The preparation that was tested, for example a specific solvent extract, rather than a pure compound. |
| Animal | The test organism or system used in the experiment. |
| Administration route | How the material was administered in the experiment. |
| Country and geographical area | Where the studied material came from. |
| Collected part and part condition | Which part of the organism was used, and in what state it was when used. |
| Compound code and compound synonym | Identifiers and alternative names attached to a compound. |
The links point at record number 1 of each type, as an example.
How do I get to the data itself?
Every entity page has a JSON version at the same address: request it with an Accept: application/json header.
For anything the pages do not cover, use the SPARQL editor. It runs against the same endpoint the site uses, and results can be exported as CSV, JSON or XML. The endpoint itself is proxied at /raw/sparql if you want to query it from a script.
Pages also show the SPARQL queries that were used to build them, at the bottom, so you can copy one and start from there.
Why do some pages show a raw list of properties?
Not every entity type has a dedicated page yet. When there is none, the address falls back to a generic view that lists the incoming and outgoing links of that record. It is less readable, but nothing is hidden from you.
The name or the classification looks wrong. Why?
Names were entered per record, over a long period, from the publications themselves. They carry the spelling and the taxonomy of their time, including typos and synonyms that have since been revised. The taxon layer reconciles what can be reconciled automatically, and the organism record keeps the original text so nothing is lost.