b7c2dcc865207dd23298429f93057c9068acef68
angie
  Wed Dec 9 15:48:34 2020 -0800
hgPhyloPlace: Add config option treeChoices: tab-sep file of {protobuf, metadata, source, description} so we can offer the user a choice between GISAID and public sequence trees (possibly different releases for reproducibility).

diff --git src/hg/hgPhyloPlace/phyloPlace.h src/hg/hgPhyloPlace/phyloPlace.h
index 9408a7f..531ae88 100644
--- src/hg/hgPhyloPlace/phyloPlace.h
+++ src/hg/hgPhyloPlace/phyloPlace.h
@@ -3,30 +3,40 @@
 #ifndef _PHYLO_PLACE_H_
 #define _PHYLO_PLACE_H_
 
 #include "common.h"
 #include "dnaseq.h"
 #include "hash.h"
 #include "linefile.h"
 #include "parsimonyProto.h"
 #include "phyloTree.h"
 #include "trashDir.h"
 
 #define PHYLOPLACE_DATA_DIR "hgPhyloPlaceData"
 
 #define NEXTSTRAIN_DRAG_DROP_DOC "https://docs.nextstrain.org/projects/auspice/en/latest/advanced-functionality/drag-drop-csv-tsv.html"
 
+struct treeChoices
+/* Phylogenetic tree versions for the user to choose from. */
+{
+    char **protobufFiles;      // Mutation annotated tree files in protobuf format for UShER
+    char **metadataFiles;      // Sample metadata a la GISAID's nextmeta download option
+    char **sources;            // GISAID or public
+    char **descriptions;       // Menu labels to describe the options to the user
+    int count;                 // Number of choices (and size of each array)
+};
+
 struct seqInfo
 /* User sequences, alignments and statistics */
 {
     struct seqInfo *next;
     struct dnaSeq *seq;
     struct psl *psl;
     struct singleNucChange *sncList;
     struct singleNucChange *maskedSncList;
     struct slRef *maskedReasonsList;
     uint nCountStart;
     uint nCountMiddle;
     uint nCountEnd;
     uint ambigCount;
 };
 
@@ -111,51 +121,55 @@
                               struct slName **retSampleIds, struct seqInfo **retSeqInfo,
                               struct slPair **retFailedSeqs, struct slPair **retFailedPsls,
                               int *pStartTime);
 /* Read in FASTA from lf and make sure each item has a reasonable size and not too high
  * percentage of N's.  Align to reference, extract SNVs from alignment, and save as VCF
  * with sample genotype columns. */
 
 struct usherResults *runUsher(char *usherPath, char *usherAssignmentsPath, char *vcfFile,
                               int subtreeSize, struct slName *userSampleIds,
                               struct hash *condensedNodes, int *pStartTime);
 /* Open a pipe from Yatish Turakhia's usher program, save resulting big trees and
  * subtrees to trash files, return list of slRef to struct tempName for the trash files
  * and parse other results out of stderr output. */
 
 void treeToAuspiceJson(struct subtreeInfo *sti, char *db, struct dnaSeq *ref,
-                       char *bigGenePredFile, struct hash *sampleMetadata, char *jsonFile);
+                       char *bigGenePredFile, struct hash *sampleMetadata, char *jsonFile,
+                       char *source);
 /* Write JSON for tree in Nextstrain's Augur/Auspice V2 JSON format
  * (https://github.com/nextstrain/augur/blob/master/augur/data/schema-export-v2.json). */
 
 struct tempName *writeCustomTracks(struct tempName *vcfTn, struct usherResults *ur,
                                    struct slName *sampleIds, struct phyloTree *bigTree,
                                    int fontHeight, int *pStartTime);
 /* Write one custom track per subtree, and one custom track with just the user's uploaded samples. */
 
 
 struct sampleMetadata *metadataForSample(struct hash *sampleMetadata, char *sampleId);
 /* Look up sampleId in sampleMetadata, by accession if sampleId seems to include an accession. */
 
 struct phyloTree *phyloPruneToIds(struct phyloTree *node, struct slName *sampleIds);
 /* Prune all descendants of node that have no leaf descendants in sampleIds. */
 
 char *phyloPlaceDbSetting(char *db, char *settingName);
 /* Return a setting from hgPhyloPlaceData/<db>/config.ra or NULL if not found. */
 
 char *phyloPlaceDbSettingPath(char *db, char *settingName);
 /* Return path to a file named by a setting from hgPhyloPlaceData/<db>/config.ra,
  * or NULL if not found.  (Append hgPhyloPlaceData/<db>/ to the beginning of relative path) */
 
+struct treeChoices *loadTreeChoices(char *db);
+/* If <db>/config.ra specifies a treeChoices file, load it up, else return NULL. */
+
 void reportTiming(int *pStartTime, char *message);
 /* Print out a report to stderr of how much time something took. */
 
 boolean hgPhyloPlaceEnabled();
 /* Return TRUE if hgPhyloPlace is enabled in hg.conf and db wuhCor1 exists. */
 
-char *phyloPlaceSamples(struct lineFile *lf, char *db, boolean doMeasureTiming, int subtreeSize,
-                        int fontHeight);
-/* Given a lineFile that contains either FASTA or VCF, prepare VCF for add_missing_samples;
- * if that goes well then run add_missing_samples, report results, make custom track files
+char *phyloPlaceSamples(struct lineFile *lf, char *db, char *defaultProtobuf,
+                        boolean doMeasureTiming, int subtreeSize, int fontHeight);
+/* Given a lineFile that contains either FASTA or VCF, prepare VCF for usher;
+ * if that goes well then run usher, report results, make custom track files
  * and return the top-level custom track file; otherwise return NULL. */
 
 #endif //_PHYLO_PLACE_H_