void scanGenes(char *chromName)
{
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
struct hashEl *hel = NULL;
struct slName *snpPtr = NULL;
struct slName *snpList = NULL;
char *geneName = NULL;
sqlSafef(query, sizeof(query), "select name, txStart, txEnd from refGene where chrom = '%s'", chromName);
sr = sqlGetResult(conn, query);
while ((row = sqlNextRow(sr)) != NULL)
{
geneName = cloneString(row[0]);
snpList = readSnps(chromName, sqlUnsigned(row[1]), sqlUnsigned(row[2]));
for (snpPtr = snpList; snpPtr != NULL; snpPtr = snpPtr->next)
{
hel = hashLookup(contigHash, snpPtr->name);
if (hel == NULL)
verbose(1, "no function for %s\n", snpPtr->name);
else if (!sameString(hel->val, geneName))
verbose(1, "mismatch for %s: refGene = %s, mrna_acc = %s\n", snpPtr->name, geneName, hel->val);
}
}
sqlFreeResult(&sr);
hFreeConn(&conn);
}
int main(int argc, char *argv[])
/* Read snp126. */
/* Write coords of multiple alignments to .tab file. */
/* Write counts to .log file. */
{
if (argc != 2)
usage();
snpDb = argv[1];
hSetDb(snpDb);
outputFileHandle = mustOpen("snpMultiple.tab", "w");
logFileHandle = mustOpen("snpMultiple.log", "w");
readSnps();
writeResults();
// free hashes
carefulClose(&outputFileHandle);
carefulClose(&logFileHandle);
return 0;
}
int main(int argc, char *argv[])
/* Read snp125. */
{
if (argc != 2)
usage();
snpDb = argv[1];
hSetDb(snpDb);
readSnps();
checkExceptions();
return 0;
}
void snpValid()
/* Test snpMap --> dbSnpRs/affy for one assembly. */
{
char *Org;
char *dbSnpTbl = NULL;
struct dbSnpRs *dbSnps = NULL;
struct dbSnpRs *dbSnp = NULL;
struct affy10KDetails *affy10s = NULL;
struct affy10KDetails *affy10 = NULL;
struct affy120KDetails *affy120s = NULL;
struct affy120KDetails *affy120 = NULL;
struct axtScoreScheme *simpleDnaScheme = NULL;
int match = 0; /* good match of minimal acceptable quality */
int mismatch = 0; /* unacceptable match quality */
int missing = 0; /* unable to find rsId in dbSnpRs/affy */
int goodrc = 0; /* matches after reverse-complement */
int assemblyDash = 0; /* assembly context is just a single dash - (complex cases) */
int gapNib = 0; /* nib returns n's, we are in the gap */
int totalMatch = 0;
int totalMismatch = 0;
int totalMissing = 0;
int totalGoodrc = 0;
int totalAssemblyDash = 0;
int totalGapNib = 0;
boolean affy = FALSE;
int mode = 3;
void *next = NULL;
char *id = NULL;
char *seq = NULL;
char affy120id[12];
int matchScore = 100;
int misMatchScore = 100;
int gapOpenPenalty = 400;
int gapExtendPenalty = 50;
int noDna = 0;
int snpMapRows = 0;
/* controls whether affy120k, affy10k, or dbSnpRs is used
currently affys are human only
*/
if (!hDbIsActive(db))
{
printf("Currently no support for db %s\n", db);
return;
}
hSetDb(db);
Org = hOrganism(db);
if (sameWord(Org,"Human"))
affy = TRUE;
if (sameWord(Org,"Human"))
dbSnpTbl = "dbSnpRsHg";
else if (sameWord(Org,"Mouse"))
dbSnpTbl = "dbSnpRsMm";
else if (sameWord(Org,"Rat"))
dbSnpTbl = "dbSnpRsRn";
else
{
printf("Currently no support for Org %s\n", Org);
return;
}
simpleDnaScheme = axtScoreSchemeSimpleDna(matchScore, misMatchScore, gapOpenPenalty, gapExtendPenalty);
uglyf("dbSnp Table=%s \n",dbSnpTbl);
uglyf("Affy=%s \n", affy ? "TRUE" : "FALSE" );
dbSnps = readDbSnps(dbSnpTbl);
printf("read hgFixed.%s \n",dbSnpTbl);
if (affy)
{
affy10s = readAffy10();
printf("read hgFixed.affy10KDetails \n");
affy120s = readAffy120();
printf("read hgFixed.affy120KDetails \n");
}
int bogus = 0;
// debug
if (0)
{
printf("rsId assembly-sequence \n");
printf("---------------------------------------------- \n");
for (dbSnp = dbSnps; dbSnp != NULL; dbSnp = dbSnp->next)
{
printf("%s %s \n",
dbSnp->rsId,
dbSnp->assembly
);
// debug: cut it short for testing only
if (++bogus > 1)
break;
}
printf("\n");
printf("\n");
}
bogus=0;
struct slName *cns = hAllChromNames();
struct slName *cn=NULL;
if (!cns)
{
printf("testDb: hAllChromNames returned empty list \n");
return;
}
if (affy)
{
mode=1; /* start on affy120 with numbers in snpMap.rsId */
}
else
{
mode=2; /* start on dbSnps with "rs*" in snpMap.rsId */
}
for (cn = cns; cn != NULL; cn = cn->next)
{
struct dnaSeq *chromSeq = NULL;
struct snpMap *snps = NULL;
struct snpMap *snp = NULL;
if (chr != NULL)
if (!sameWord(chr,cn->name))
continue;
//uglyf("testDb: beginning chrom %s \n",cn->name);
chromSeq = hLoadChrom(cn->name);
printf("testDb: chrom %s : size (%u) \n",cn->name,chromSeq->size);
snps = readSnps(cn->name);
printf("read %s.snpMap where chrom=%s \n",db,cn->name);
dbSnp = dbSnps;
affy10 = affy10s;
affy120 = affy120s;
printf("=========================================================\n");
for (snp = snps; snp != NULL; snp = snp->next)
{
int cmp = -1;
char *nibDna=NULL;
char *nibDnaRc=NULL;
++snpMapRows;
/*
printf("%s %s %u %u %s\n",
snp->name,
snp->chrom,
snp->chromStart,
snp->chromEnd,
nibDna
);
*/
while (cmp < 0)
{
while (cmp < 0)
{
switch (mode)
{
case 1:
next = affy120; break;
case 2:
next = dbSnp; break;
case 3:
next = affy10; break;
}
if (next == NULL)
{
switch (mode)
{
case 1:
++mode; break;
case 2:
++mode; break;
case 3:
cmp = 1; break;
}
}
else
{
break;
}
}
if (cmp < 0)
{
switch (mode)
{
case 1:
safef(affy120id, sizeof(affy120id), "%d", affy120->affyId); /* have int type but want string */
id = affy120id;
break;
case 2:
id = dbSnp->rsId; break;
case 3:
id = affy10->affyId; break;
}
cmp=mystrcmp(id, snp->name);
}
if (cmp < 0)
{
switch (mode)
{
case 1:
affy120 = affy120->next; break;
case 2:
dbSnp = dbSnp->next; break;
case 3:
affy10 = affy10->next; break;
}
}
}
if (cmp==0)
{
int strand=1;
char *rc = NULL;
int m = 0;
int lf = 0; /* size of left flank context (lower case dna) */
int rf = 0; /* size of right flank context (lower case dna) */
int ls = 0; /* total size of assembly dna context plus actual region in dbSnpRs/affy */
char *origSeq = NULL; /* use to display the original dnSnpRs.assembly seq */
switch (mode)
{
case 1:
seq = affy120->sequenceA; break;
case 2:
seq = dbSnp->assembly; break;
case 3:
seq = affy10->sequenceA; break;
}
if (sameString(seq,"-"))
{
++assemblyDash;
if (Verbose)
printf("(no assembly context) rsId=%s chrom=%s %u %u \n assembly=%s \n\n",
id,
snp->chrom,
snp->chromStart,
snp->chromEnd,
seq
);
continue;
}
origSeq = seq;
lf = leftFlank(origSeq);
rf = rightFlank(origSeq);
seq = cloneString(origSeq);
stripDashes(seq); /* remove dashes indicating insert to simplify and correct processing of nib data */
ls = strlen(seq); /* used to be: lengthOneDash(seq); */
//debug
//uglyf("about to call checkandFetchNib origSeq=%s lf=%d, rf=%d ls=%d \n", origSeq, lf, rf, ls);
nibDna = checkAndFetchNib(chromSeq, snp, lf, ls);
if (nibDna==NULL)
{
++noDna;
printf("no dna for %s %s %u %u \n",
snp->name,
snp->chrom,
snp->chromStart,
snp->chromEnd
);
continue;
}
//debug
//uglyf("got past checkandFetchNib call: \n nibDna=%s \n",nibDna);
if (allNs(nibDna))
{
++gapNib;
++mismatch;
if (Verbose)
printf("(nib gap) rsId=%s chrom=%s %u %u \n assembly=%s \n snpMap=%s \n\n",
id,
snp->chrom,
snp->chromStart,
snp->chromEnd,
seq,
nibDna
);
continue;
}
m = misses(seq,nibDna);
if (m > 1)
{
//debug
//uglyf("rc: about to call checkandFetchNib \n");
rc = checkAndFetchNib(chromSeq, snp, rf, ls);
if (rc==NULL)
{
++noDna;
printf("no dna for %s %s %u %u \n",
snp->name,
snp->chrom,
snp->chromStart,
snp->chromEnd
);
continue;
}
//debug
//uglyf("rc: got past checkandFetchNib call: \n rc Dna=%s \n",rc);
reverseComplement(rc,strlen(rc));
int n = misses(seq, rc);
if (n < m)
{
strand=-1;
m = n;
}
}
if (m <= 1)
{
++match;
if (strand < 1)
++goodrc;
}
else
{
struct dnaSeq query, target;
struct axt *axtAln = NULL;
int bestScore = 0;
ZeroVar(&query);
query.dna = seq;
query.size = strlen(query.dna);
ZeroVar(&target);
target.dna = nibDna;
target.size = strlen(target.dna);
axtAln = axtAffine(&query, &target, simpleDnaScheme);
strand = 1;
if (axtAln)
{
bestScore = axtAln->score / ls;
}
axtFree(&axtAln);
if (bestScore < threshold)
{
ZeroVar(&target);
target.dna = rc;
target.size = strlen(target.dna);
axtAln = axtAffine(&query, &target, simpleDnaScheme);
if ((axtAln) && (bestScore < (axtAln->score / ls)))
{
strand = -1;
bestScore = axtAln->score / ls;
}
axtFree(&axtAln);
}
if (bestScore >= threshold)
{
++match;
if (strand < 1)
++goodrc;
}
else
{
++mismatch;
}
if ((bestScore < threshold) || Verbose)
{
printf(
"score=%d misses=%u strand=%d rsId=%s chrom=%s %u %u lf=%d ls=%d \n"
" assembly=%s \n"
" snpMap=%s \n"
"rc snpMap=%s \n"
"\n",
bestScore,
m,
strand,
id,
snp->chrom,
snp->chromStart,
snp->chromEnd,
lf,
ls,
seq,
nibDna,
rc
);
}
}
freez(&rc);
freez(&seq);
}
else
{
char snpLkup[10] = "";
/* this id is missing from dbSnpRs/affy! */
++missing;
switch (mode)
{
case 1:
safef(snpLkup,sizeof(snpLkup),"%s","affy120"); break;
case 2:
safef(snpLkup,sizeof(snpLkup),"%s",dbSnpTbl); break;
case 3:
safef(snpLkup,sizeof(snpLkup),"%s","affy10"); break;
}
if (Verbose)
printf("snpMap.name=%s is missing from %s (now at %s) \n\n",snp->name,snpLkup,id);
}
freez(&nibDna);
// debug: cut it short for testing only
//break;
}
snpMapFreeList(&snps);
dnaSeqFree(&chromSeq);
printf("\n\n\n Total matches for chrom %s:\n ",cn->name);
printf(" matches: %u \n ",match);
printf(" mismatches: %u \n",mismatch);
printf("missing from dbSnpRs: %u \n",missing);
printf(" rev compl matches: %u \n",goodrc);
printf(" assembly = -: %u \n",assemblyDash);
printf(" nib in gap : %u \n",gapNib);
printf("\n\n=========================================\n");
totalMatch += match;
totalMismatch += mismatch;
totalMissing += missing;
totalGoodrc += goodrc;
totalAssemblyDash += assemblyDash;
totalGapNib += gapNib;
match = 0;
mismatch = 0;
missing = 0;
goodrc = 0;
assemblyDash = 0;
gapNib = 0;
// debug: cut it to just one or two chrom for testing
//if (++bogus > 1)
// break;
printf("\n");
printf("\n");
}
slFreeList(&cns);
dbSnpRsFreeList(&dbSnps);
if (affy)
{
affy10KDetailsFreeList(&affy10s);
affy120KDetailsFreeList(&affy120s);
}
axtScoreSchemeFree(&simpleDnaScheme);
printf("\n\n\n Grand Totals: \n ");
printf(" matches: %u \n ",totalMatch);
printf(" mismatches: %u \n",totalMismatch);
printf("missing from dbSnpRs: %u \n",totalMissing);
printf(" rev compl matches: %u \n",totalGoodrc);
printf(" assembly = -: %u \n",totalAssemblyDash);
printf(" nib in gap : %u \n",totalGapNib);
printf("\n Total rows in snpMap: %u \n ",snpMapRows);
printf("\n # no dna found for : %u \n ",noDna);
printf("\n\n=========================================\n");
}
void affyCheck()
/* affyCheck - read in all Affy SNPs, compare to lookupTable. */
{
struct snpSimple *snps = NULL;
struct snpSimple *snp = NULL;
struct snpSimple *newSnp = NULL;
char query[512];
struct sqlConnection *conn = hAllocConn();
struct sqlResult *sr;
char **row;
int count = 0;
int obsLen = 0;
snps = readSnps();
verbose(1, "checking....\n");
for (snp = snps; snp != NULL; snp = snp->next)
{
count++;
verbose(2, "----------------------\n");
verbose(2, "%d: %s:%d-%d\n", count, snp->chrom, snp->chromStart, snp->chromEnd);
sqlSafef(query, sizeof(query), "select name, chrom, chromStart, chromEnd, strand, observed from %s"
" where chrom = '%s' and chromStart = %d and chromEnd = %d", lookupTable, snp->chrom, snp->chromStart, snp->chromEnd);
sr = sqlGetResult(conn, query);
row = sqlNextRow(sr);
if (row == NULL)
{
verbose(2, "no matches for %s %s\n", affyTable, snp->name);
continue;
}
newSnp = snpSimpleLoad(row);
verbose(2, "comparing %s %s to %s %s\n", affyTable, snp->name, lookupTable, newSnp->name);
if (sameString(newSnp->observed, "n/a"))
{
while ((row = sqlNextRow(sr)) != NULL) { }
continue;
}
if (sameString(newSnp->observed, "t/n"))
{
while ((row = sqlNextRow(sr)) != NULL) { }
continue;
}
if (sameString(snp->observed, newSnp->observed))
{
while ((row = sqlNextRow(sr)) != NULL) { }
continue;
}
obsLen = strlen(newSnp->observed);
if (obsLen > 3)
{
verbose(2, "%s is not bi-allelic (%s)\n", newSnp->name, newSnp->observed);
while ((row = sqlNextRow(sr)) != NULL)
{
}
continue;
}
verbose(1, "----------------------\n");
verbose(1, "%d: %s:%d-%d\n", count, snp->chrom, snp->chromStart, snp->chromEnd);
verbose(1, "comparing %s %s to %s %s\n", affyTable, snp->name, lookupTable, newSnp->name);
verbose(1, "observed difference\n");
verbose(1, "%s observed = %s, %s observed = %s\n", affyTable, snp->observed, lookupTable, newSnp->observed);
if (snp->strand == '-' && newSnp->strand == '+')
{
verbose(1, "strand difference\n");
verbose(1, "%s strand = %c, %s strand = %c\n", affyTable, snp->strand, lookupTable, newSnp->strand);
}
if (snp->strand == '+' && newSnp->strand == '-')
{
verbose(1, "strand difference\n");
verbose(1, "%s strand = %c, %s strand = %c\n", affyTable, snp->strand, lookupTable, newSnp->strand);
}
/* check here for multiple matches */
while ((row = sqlNextRow(sr)) != NULL)
{
}
/* free newSnp */
}
snpSimpleFreeList(&snps);
}
