be4311c07e14feb728abc6425ee606ffaa611a58 markd Fri Jan 22 06:46:58 2021 -0800 merge with master diff --git src/hg/hgPhyloPlace/phyloPlace.h src/hg/hgPhyloPlace/phyloPlace.h index 9408a7f..1a22d68 100644 --- src/hg/hgPhyloPlace/phyloPlace.h +++ src/hg/hgPhyloPlace/phyloPlace.h @@ -1,161 +1,175 @@ /* Place SARS-CoV-2 sequences in phylogenetic tree using add_missing_samples program. */ #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; }; struct variantPathNode /* A list of these gives a path from phylo tree root to some node (usually user seq leaf). */ { struct variantPathNode *next; char *nodeName; // Either a numeric internal node ID or the user seq name struct singleNucChange *sncList; // One or more single nucleotide changes associated with node }; struct bestNodeInfo { struct bestNodeInfo *next; char *name; // Node name struct variantPathNode *variantPath; // Mutations assigned to nodes along path from root->node boolean isSibling; // Placement would be as sibling of node (not child) }; struct baseVal /* List of imputed base positions and values */ { struct baseVal *next; int chromStart; // 0-based position char *val; // nucleotide base(s) }; struct placementInfo /* Info about a sample's mutations and placement in the phylo tree */ { char *sampleId; // Sample name from FASTA or VCF header struct slName *sampleMuts; // Differences with the reference genome struct variantPathNode *variantPath; // Mutations assigned to nodes along path from root struct bestNodeInfo *bestNodes; // Other nodes identified as equally parsimonious struct baseVal *imputedBases; // Ambiguous bases imputed to ref/alt [ACGT] int parsimonyScore; // Parsimony cost of placing sample int bestNodeCount; // Number of equally parsimonious placements }; struct subtreeInfo /* Parsed subtree from usher and derivative products. */ { struct subtreeInfo *next; struct tempName *subtreeTn; // Newick file from usher (may have condensed nodes) struct phyloTree *subtree; // Parsed subtree (#*** with annotated muts? shoudl we?_ struct hash *subtreeIdToIx; // Map of subtree nodes to VCF output sample order struct slName *subtreeUserSampleIds; // List of user-uploaded samples in this subtree struct slName *subtreeNameList; // List of leaf names with nicer names for cond. nodes }; struct usherResults /* Tree+samples download file, sample placements, and subtrees parsed from usher output. */ { struct tempName *bigTreePlusTn; // Newick file: original tree plus user's samples struct hash *samplePlacements; // Info about each sample's placement in the tree struct subtreeInfo *subtreeInfoList; // For each subtree: tree, file, node info etc. }; struct sampleMetadata /* Information about a virus sample. */ { char *strain; // Strain name, usually of the form Country/ArbitraryId/YYYY-MM-DD char *epiId; // GISAID EPI_ISL_[0-9]+ ID char *gbAcc; // GenBank accession char *date; // Sample collection date char *author; // Author(s) to credit char *nClade; // Nextstrain year-letter clade char *gClade; // GISAID amino acid change clade char *lineage; // Pangolin lineage char *country; // Country in which sample was collected char *division; // Administrative division in which sample was collected (country or state) char *location; // Location in which sample was collected (city) char *countryExp; // Country in which host was exposed to virus char *divExp; // Administrative division in which host was exposed to virus char *origLab; // Originating lab char *subLab; // Submitting lab char *region; // Continent on which sample was collected }; struct tempName *vcfFromFasta(struct lineFile *lf, char *db, struct dnaSeq *refGenome, boolean *informativeBases, struct slName **maskSites, 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); + char *source, 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_