====== Python utilities to handle NUVA ======
A [[https://github.com/fkaag71/nuva-utils/tree/master/NUVA%20Utils|library available on GitHub]] allows to retrieve and explore NUVA. It is a work in progress, that will be progressively enriched to provide metrics on code systems based upon their mapping to NUVA codes.

The supported functions are so far:
<code python>
get_nuva_version()
</code>
Returns the version index for the last publication of NUVA.

<code python>
get_nuva(version)
</code>
Uploads in the current directory the referenced version in RDF/XML format as **nuva_ans.rdf**, and creates a rebased version **nuva_ivci.rdf**.

<code python>
split_nuva()
</code>
From the uploaded **nuva_ivci.rdf** file, creates a split version as a collection of files in RDF/Turtle format:
  * **nuva_core.ttl** including the concepts for vaccines, valences, target diseases and their labels in English
  * **nuva_lang_XX.ttl** includes all translations for language XX
  * **nuva_refcode_YYY.ttl** includes the concepts and the NUVA alignments for code system YYY

<code python>
refturtle_to_map(code)
</code>
Starting from the **nuva_refcode_YYY.ttl** file for the given code, creates a simple CSV file **nuva_refcode_YYY.csv** with alignments between the given code and NUVA.

<code python>
map_to_turtle(code)
</code>
Assuming that the **nuva_refcode_YYY.csv** file has been copied to work file **nuva_code_YYY.csv**, then edited for enhancing the alignments, creates a Turtle work file **nuva_code_YYY.ttl** for further processing.

Note that the refcode file contains the NUVA English labels of vaccines for convenience, but these are not required nor processed from the work code file.

<code python>
query_core(q)
</code>
Runs a SPARQL query q against the core graph loaded from **nuva_core.ttl**

<code python>
query_code(q,code)
</code>
Runs a SPARQL query q against a graph formed by merging **nuva_core.ttl** and the work file **nuva_code_YYY.ttl**, thus allowing to run checks and measures on the alignment.


<code>
eval_code(code)
</code>
Produces the metrics for a code system, given a **nuva_code_YYY.csv** file for alignments.

Subproducts are:
  * **nuva_reverse_YYY.csv** : file with all NUVA codes matching a given external code
  * **nuva_best_YYY.csv**: file with the best possible external code for a given NUVA code

An example use sequence is included in the file:
<code python>
# Here the main program - Adapt the work directory to your environment

os.chdir(str(Path.home())+"/Documents/NUVA")
get_nuva(get_nuva_version())
split_nuva()
refturtle_to_map("CVX")
shutil.copyfile("nuva_refcode_CVX.csv","nuva_code_CVX.csv")
map_to_turtle("CVX")

q = """ 
   # All vaccines against smallpox
    SELECT ?vcode ?vl WHERE { 
    ?dis rdfs:subClassOf nuva:Disease .
    ?dis rdfs:label "Smallpox-Monkeypox"@en .
    ?vac rdfs:subClassOf nuva:Vaccine .
    ?vac rdfs:label ?vl . 
    ?vac skos:notation ?vcode .
    ?vac nuvs:containsValence ?val . 
    ?val nuvs:prevents ?dis 
 }
"""
res = query_core(q)
for row in res:
     print (f"{row.vcode} - {row.vl}")

res = eval_code("CVX")
print ("Completeness {:.1%} ".format(res['Completeness']))
print ("Precision {:.1%} ".format(res['Precision']))

</code>