/* hgKnownMore - Create the knownMore table from a variety of sources.. */

/* Copyright (C) 2013 The Regents of the University of California 
 * See kent/LICENSE or http://genome.ucsc.edu/license/ for licensing information. */
#include "common.h"
#include "knownMore.h"
#include "linefile.h"
#include "hash.h"
#include "jksql.h"
#include "hugoMulti.h"
#include "knownInfo.h"


void usage()
/* Explain usage and exit. */
{
errAbort(
  "hgKnownMore - Create the knownMore table from a variety of sources.\n"
  "usage:\n"
  "   hgKnownMore database loc2ref omimIds nomeids.txt\n"
  "where:\n"
  "   database - destination mysql database, often hgN for some N\n"
  "   loc2ref - downloaded from ftp://ncbi.nlm.nih.gov/refseq/LocusLink\n"
  "   mim2loc - downloaded from ftp://ncbi.nlm.nih.gov/refseq/LocusLink\n"
  "   omimIds - connects names and omimIds\n"
  "   nomeIds - lots of info from HUGO ftp site\n");
}

struct rsInfo
/* Info we store on a refSeq entry. */
    {
    struct rsInfo *next; /* Next in list. */
        /* First fields are from ra file. */
    char *locusLinkId;	/* LocusLink. */
    char *mrnaAcc;	/* mRNA accession. */
    char *pgi;		/* Protein gi number. */
    };

void readLoc2ref(char *fileName, struct hash **retPgiHash, struct hash **retLocHash)
/* Read loc2ref file.  Create hashes of rsInfo's indexed by pgi and locusLinkId. */ 
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *words[4];
int wordCount;
struct rsInfo *rs;
struct hash *locHash = newHash(0);
struct hash *pgiHash = newHash(0);
int lineCount = 0, count = 0;

while ((wordCount = lineFileChop(lf, words)) > 0)
    {
    ++lineCount;
    if (wordCount == 4)
        {
	if (hashLookup(locHash, words[0]) != NULL && startsWith("XM_", words[1]))
	    continue;
	AllocVar(rs);
	hashAddSaveName(locHash, words[0], rs, &rs->locusLinkId);
	rs->mrnaAcc = cloneString(words[1]);
	hashAddSaveName(pgiHash, words[2], rs, &rs->pgi);
	}
    }
lineFileClose(&lf);
*retLocHash = locHash;
*retPgiHash = pgiHash;
printf("Added %d locusLink ids from %s in %d lines\n", count, fileName, lineCount);
}

void readMim(char *fileName, struct hash **retMimHash)
/* Make a hash keyed by locusLinkId with omimId values. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *row[2];
struct hash *mimHash = newHash(0);

while (lineFileRow(lf, row))
    {
    if (sameString(row[1], "672")) uglyf("Got a live one %s %s\n", row[0], row[1]);
    hashAdd(mimHash, row[1], cloneString(row[0]));
    }
lineFileClose(&lf);
*retMimHash = mimHash;
}

void readHugoMultiTable(char *fileName, struct hugoMulti **retList,
	struct hash **retIdHash, struct hash **retSymbolHash)
/* Read in file into list and hashes.  Make hash keyed on omim ID
 * and on OMIM symbol.  */
{
struct hash *idHash = newHash(0);
struct hash *symbolHash = newHash(0);
struct hugoMulti *list = NULL, *el;
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *words[16];
char *line;
int lineSize, wordCount;
char *name;

while (lineFileNext(lf, &line, &lineSize))
    {
    if (line[0] == 0 || line[0] == '#')
        continue;
    wordCount = chopTabs(line, words);
    lineFileExpectWords(lf, 12, wordCount);
    el = hugoMultiLoad(words);
    slAddHead(&list, el);
    name = el->omimId;
    if (name[0] != 0)
	hashAdd(idHash, name, el);
    name = el->symbol;
    if (name[0] != 0)
	hashAdd(symbolHash, name, el);
    }
lineFileClose(&lf);
slReverse(&list);
*retList = list;
*retIdHash = idHash;
*retSymbolHash = symbolHash;
}


struct nameOmim
/* Connect name with OMIM ID */
    {
    struct nameOmim *next;	/* Next in list. */
    char *name;                 /* Symbolic name. */
    char *omimId;		/* OMIM numerical ID as a string. */
    };

void readNameOmim(char *fileName, struct nameOmim **retList, struct hash **retNameOmimHash,
	struct hash **retOmimNameHash)
/* Read in file into list and hashes.  Make hash keyed on transcriptId (txOmimHash)
 * and hash keyed on omimId (omimNameHash). */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
struct hash *nameOmimHash = newHash(0);
struct hash *omimNameHash = newHash(0);
struct nameOmim *list = NULL, *el;
char *row[2];

while (lineFileRow(lf, row))
    {
    AllocVar(el);
    slAddHead(&list, el);
    touppers(row[0]);
    hashAddSaveName(nameOmimHash, row[0], el, &el->name);
    hashAddSaveName(omimNameHash, row[1], el, &el->omimId);
    }
lineFileClose(&lf);
slReverse(&list);
*retList = list;
*retNameOmimHash = nameOmimHash;
*retOmimNameHash = omimNameHash;
}

struct knownInfo *loadKnownInfo(struct sqlConnection *conn)
/* Load up known info table into list. */
{
struct knownInfo *list = NULL, *el;
struct sqlResult *sr;
char **row;

char query[1024];
sqlSafef(query, sizeof query, "select * from knownInfo");
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
    {
    el = knownInfoLoad(row);
    slAddHead(&list, el);
    }
sqlFreeResult(&sr);
slReverse(&list);
return list;
}


void hgKnownMore(char *database, char *loc2ref, char *mim2loc, char *omimIds, char *nomeIds)
/* hgKnownMore - Create the knownMore table from a variety of sources.. */
{
struct hash *pgiHash = NULL;	/* Hash of rsInfo indexed by gi. */
struct hash *locHash = NULL;	/* Hash of rsInfo indexed by locusLink IDs. */
struct rsInfo *rs;
struct hash *hmOmimHash = NULL, *hmSymbolHash = NULL;
struct hash *mimHash = NULL;
struct hugoMulti *hmList = NULL, *hm;
struct hash *nameOmimHash = NULL, *omimNameHash = NULL;
struct nameOmim *nameOmimList = NULL, *nameOmim;
struct knownInfo *kiList = NULL, *ki;
struct knownMore km;
struct sqlConnection *conn;
char *tabName = "knownMore.tab";
FILE *f = NULL;
char *omimIdString = NULL;
char query[256];


readLoc2ref(loc2ref, &pgiHash, &locHash);
readMim(mim2loc, &mimHash);
readHugoMultiTable(nomeIds, &hmList, &hmOmimHash, &hmSymbolHash);
printf("Read %d elements in %s\n", slCount(hmList), nomeIds);
readNameOmim(omimIds, &nameOmimList, &nameOmimHash, &omimNameHash);
printf("Read %d elements in %s\n", slCount(nameOmimList), omimIds);

conn = sqlConnect(database);
kiList = loadKnownInfo(conn);
printf("Read %d elements from knownInfo table in %s\n", 
	slCount(kiList), database);

printf("Writing %s\n", tabName);
f = mustOpen(tabName, "w");
for (ki = kiList; ki != NULL; ki = ki->next)
    {
    /* Fill out a knownMore data structure.  Start with all zero
     * just to avoid garbage. */
    zeroBytes(&km, sizeof(km));

    /* First fields come from knownInfo generally. */
    km.name = ki->name;	/* The name displayed in the browser: OMIM, gbGeneName, or transId */
    km.transId = ki->transId;	/* Transcript id. Genie generated ID. */
    km.geneId = ki->geneId;	/* Gene (not transcript) Genie ID */
    km.gbGeneName = ki->geneName;	/* Connect to geneName table. Genbank gene name */
    km.gbProductName = ki->productName;	/* Connects to productName table. Genbank product name */
    km.gbProteinAcc = ki->proteinId;	/* Genbank accession of protein */
    km.gbNgi = ki->ngi;	/* Genbank gi of nucleotide seq. */
    km.gbPgi = ki->pgi;	/* Genbank gi of protein seq. */

    /* Fill in rest with acceptable values for no-data-present. */
    km.omimId = 0;	/* OMIM ID or 0 if none */
    km.omimName = "";	/* OMIM primary name */
    km.hugoId = 0;	/* HUGO Nomeclature Committee ID or 0 if none */
    km.hugoSymbol = "";	/* HUGO short name */
    km.hugoName = "";	/* HUGO descriptive name */
    km.hugoMap = "";	/* HUGO Map position */
    km.pmId1 = 0;	/* I have no idea - grabbed from a HUGO nomeids.txt */
    km.pmId2 = 0;	/* Likewise, I have no idea */
    km.refSeqAcc = "";	/* Accession of RefSeq mRNA */
    km.aliases = "";	/* Aliases if any.  Comma and space separated list */
    km.locusLinkId = 0;	/* Locus link ID */
    km.gdbId = "";	/* NCBI GDB database ID */

    /* See if it's a disease gene with extra info. */
    omimIdString = NULL;
    rs = hashFindVal(pgiHash, km.gbPgi);
    if (rs != NULL && rs->locusLinkId != NULL)
	{
	km.locusLinkId = atoi(rs->locusLinkId);
	omimIdString = hashFindVal(mimHash, rs->locusLinkId);
	}
    if (rs != NULL && rs->mrnaAcc != NULL)
	km.refSeqAcc = rs->mrnaAcc;
    if (omimIdString != NULL)
	{
	km.omimId = atoi(omimIdString);	/* OMIM ID or 0 if none */
	nameOmim = hashFindVal(omimNameHash, omimIdString);
	if (nameOmim != NULL)
	    {
	    km.name = km.omimName = nameOmim->name;
	    }
	hm = hashFindVal(hmOmimHash, omimIdString);
	if (hm != NULL)
	    {
	    km.hugoId = hm->hgnc;	/* HUGO Nomeclature Committee ID or 0 if none */
	    km.name = km.hugoSymbol = hm->symbol;	/* HUGO short name */
	    km.hugoName = hm->name;	/* HUGO descriptive name */
	    km.hugoMap = hm->map;	/* HUGO Map position */
	    km.pmId1 = hm->pmId1;	/* I have no idea - grabbed from a HUGO nomeids.txt */
	    km.pmId2 = hm->pmId2;	/* Likewise, I have no idea */
	    km.refSeqAcc = hm->refSeqAcc;	/* Accession of RefSeq mRNA */
	    km.aliases = hm->aliases;	/* Aliases if any.  Comma and space separated list */
	    km.locusLinkId = hm->locusLinkId;	/* Locus link ID */
	    km.gdbId = hm->gdbId;	/* NCBI GDB database ID */
	    }
	}
    knownMoreTabOut(&km, f);
    }
carefulClose(&f);

printf("Loading database %s\n", database);
sqlSafef(query, sizeof query, "delete from knownMore");
sqlUpdate(conn, query);
sqlSafef(query, sizeof query, "load data local infile '%s' into table knownMore", tabName);
sqlUpdate(conn, query);

sqlDisconnect(&conn);
}


int main(int argc, char *argv[])
/* Process command line. */
{
if (argc != 6)
    usage();
hgKnownMore(argv[1], argv[2], argv[3], argv[4], argv[5]);
return 0;
}