Best practice to concatenate non-specific SPARQL queries?

Question

I'm writing some python code to dynamically build some SPARQL query of arbitrary length.

I have a subquery that looks like this:

                ?topObj a up:Topological_Domain_Annotation;
                         rdfs:comment '{}';
                         up:range ?range.

                 ?proteins a up:Protein .
                 ?proteins up:annotation ?otherTop .
                 ?otherTop a up:Topological_Domain_Annotation;
                           rdfs:comment '{}';
                           up:range ?otherRange.""",

Where I go in and .format() the {} depending on user specified input. I want a way to stack up multiple of these subqueries that could potentially match to other objects, ie. the rdfs:comment '{}'; line. But of course the variable names that I use to traverse to my desired object are bound in the first subquery.

Would the best solution be:

• Generate random variables to store traversed nodes for each mini-query
• Use some sort of property path/ blank node trick that I don't know about?

EDIT:

Here's an example query generated from the above template concatenated together using the random variable approach.

SELECT DISTINCT ?proteins
WHERE {
    <http://purl.uniprot.org/uniprot/P04439> up:annotation ?1WGQM.
    ?1WGQM a up:Topological_Domain_Annotation;
         rdfs:comment ?topology;
         up:range ?VLIT1 .
    <http://purl.uniprot.org/uniprot/P01911> up:annotation ?FIICT.
    ?FIICT a up:Topological_Domain_Annotation;
         rdfs:comment ?topology;
         up:range ?W89B2 .
    <http://purl.uniprot.org/uniprot/P10321> up:annotation ?WU6G3.
    ?WU6G3 a up:Topological_Domain_Annotation;
         rdfs:comment ?topology;
         up:range ?ZSIQ3 .
    ?proteins a up:Protein .
    ?proteins up:annotation ?otherTop .
    ?otherTop a up:Topological_Domain_Annotation;
             rdfs:comment ?topology;
             up:range ?OTHERRANGE .
}
LIMIT 10

Answer 1

Just use BNodes instead of generating unique names for these variables, e.g. something along these lines:

SELECT DISTINCT ?proteins
WHERE {
    <http://purl.uniprot.org/uniprot/P04439> up:annotation [
        a up:Topological_Domain_Annotation;
        rdfs:comment ?topology;
        up:range [] 
    ].
    <http://purl.uniprot.org/uniprot/P01911> up:annotation [
        a up:Topological_Domain_Annotation;
        rdfs:comment ?topology;
        up:range []
    ] .
    <http://purl.uniprot.org/uniprot/P10321> up:annotation [
        a up:Topological_Domain_Annotation;
        rdfs:comment ?topology;
        up:range [] 
    ].

    ?proteins a up:Protein .
    ?proteins up:annotation [
            a up:Topological_Domain_Annotation;
            rdfs:comment ?topology;
            up:range [] 
        ].
}
LIMIT 10

HTH