test query with Phenopackets

[kingmanzhang]

@pnrobinson
Could you add me to your phenopacket repo?

Now we can call Phenomiser with multiple query lists. The app optimizes the query sequence so to avoid repeated IO. Entry point is:

Phenomiser/phenomiser-core/src/main/java/org/jax/Phenomiser.java

Line 114 in 27c8307

public static List<List<Item2PValueAndSimilarity<TermId>>> batchQuery(List<List<TermId>> queries, List<DiseaseDB> dbs) {

It can also output the rank of the target disease if you provide one:

Phenomiser/phenomiser-core/src/main/java/org/jax/Phenomiser.java

Lines 145 to 146 in 27c8307

	public static int[] batchFindRank(List<List<TermId>> queries,
	List<TermId> targetDiseases, List<DiseaseDB> dbs){

[kingmanzhang]

@pnrobinson Where are disease labels for those phenopackets?
https://github.com/monarch-initiative/hrmd/tree/master/data/casereports

[pnrobinson]

Take a look at the LIRICAL SimulateVcf class, e;g;

Phenopacket pp = ...//
if (pp.getDiseasesCount() != 1) {
    System.err.println("[ERROR] to run this simulation a phenoopacket must have exactly one disease diagnosis");
    System.err.println("[ERROR]  " + pp.getSubject().getId() + " had " + pp.getDiseasesCount());
    return; // skip to next Phenopacket
       }
Disease diagnosis = pp.getDiseases(0);
simulatedDisease = diagnosis.getTerm().getId(); // should be an ID such as OMIM:600102

    Disease diagnosis = pp.getDiseases(0);

[kingmanzhang]

oh, I see. It's already in the phenopacket.