int main(int argc, char *argv[])
/* Process command line. */
/* argv[1] is database */
/* argv[2] is chrom */
/* argv[3] is nib file */
/* argv[4] is output file */
{
database = argv[1];
if (!hDbIsActive(database))
{
printf("Currently no support for %s\n", database);
return -1;
}
hSetDb(database);
chromName = argv[2];
// testFB();
snpMask(argv[3], argv[4]);
return 0;
}
char *checkParams(char *database, char *prefix, char *type)
/* If we don't have valid CGI parameters, quit with a Bad Request HTTP response. */
{
pushWarnHandler(htmlVaBadRequestAbort);
pushAbortHandler(htmlVaBadRequestAbort);
if(prefix == NULL || database == NULL)
errAbort("%s", "Missing prefix and/or db CGI parameter");
if (! hDbIsActive(database))
errAbort("'%s' is not a valid, active database", htmlEncode(database));
if (isNotEmpty(type) && differentString(type, ALT_OR_PATCH))
errAbort("'%s' is not a valid type", type);
char *table = NULL;
if (! sameOk(type, ALT_OR_PATCH))
{
struct sqlConnection *conn = hAllocConn(database);
table = connGeneSuggestTable(conn);
hFreeConn(&conn);
if(table == NULL)
errAbort("gene autosuggest is not supported for db '%s'", database);
}
popWarnHandler();
popAbortHandler();
return table;
}
void verifyGreatAssemblies()
{
// First read in the assembly name and description information into name lists
struct slName* supportedAssemblies = NULL;
struct lineFile *lf = lineFileOpen(greatData, TRUE);
int fieldCount = 1;
char* row[fieldCount];
int wordCount;
while ((wordCount = lineFileChopTab(lf, row)) != 0)
{
if (wordCount != fieldCount)
errAbort("The %s file is not properly formatted.\n", greatData);
slNameAddHead(&supportedAssemblies, row[0]);
}
lineFileClose(&lf);
boolean invalidAssembly = TRUE;
struct slName* currAssembly;
for (currAssembly = supportedAssemblies; currAssembly != NULL; currAssembly = currAssembly->next)
{
if (!hDbIsActive(currAssembly->name))
{
errAbort("Assembly %s in supported assembly file is not an active assembly.\n", currAssembly->name);
}
if (sameOk(database, currAssembly->name))
{
invalidAssembly = FALSE;
break;
}
}
if (invalidAssembly)
{
slReverse(&supportedAssemblies);
currAssembly = supportedAssemblies;
struct dyString* dy = dyStringNew(0);
addAssemblyToSupportedList(dy, currAssembly->name);
currAssembly = currAssembly->next;
while (currAssembly != NULL)
{
dyStringAppend(dy, ", ");
if (currAssembly->next == NULL)
dyStringAppend(dy, "and ");
addAssemblyToSupportedList(dy, currAssembly->name);
currAssembly = currAssembly->next;
}
hPrintf("<script type='text/javascript'>\n");
hPrintf("function logSpecies() {\n");
hPrintf("try {\n");
hPrintf("var r = new XMLHttpRequest();\n");
hPrintf("r.open('GET', 'http://great.stanford.edu/public/cgi-bin/logSpecies.php?species=%s');\n", database);
hPrintf("r.send(null);\n");
hPrintf("} catch (err) { }\n");
hPrintf("}\n");
hPrintf("window.onload = logSpecies;\n");
hPrintf("</script>\n");
errAbort("GREAT only supports the %s assemblies."
"\nPlease go back and ensure that one of those assemblies is chosen.",
dyStringContents(dy));
htmlClose();
dyStringFree(&dy);
}
slNameFreeList(&supportedAssemblies);
}
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 mafPrettyOut(FILE *f, struct mafAli *maf, int lineSize,
boolean onlyDiff, int blockNo)
{
int ii, ch;
int srcChars = 0;
struct mafComp *mc;
int lineStart, lineEnd;
char *summaryLine = needMem(lineSize+1);
char *referenceText;
int startChars, sizeChars, srcSizeChars;
boolean haveInserts = FALSE;
struct mafComp *masterMc = maf->components;
startChars = sizeChars = srcSizeChars = 0;
for (mc = maf->components; mc != NULL; mc = mc->next)
{
/* Figure out length of source (species) field. */
/*if (mc->size != 0)*/
{
char dbOnly[128];
int len;
char *org;
memset(dbOnly, 0, sizeof(dbOnly));
safef(dbOnly, sizeof(dbOnly), "%s", mc->src);
chopPrefix(dbOnly);
if ((org = hOrganism(dbOnly)) == NULL)
len = strlen(dbOnly);
else
len = strlen(org);
if (srcChars < len)
srcChars = len;
len = digitsBaseTen(mc->start);
if (startChars < len)
startChars = len;
len = digitsBaseTen(mc->size);
if (sizeChars < len)
sizeChars = len;
len = digitsBaseTen(mc->srcSize);
if (srcSizeChars < len)
srcSizeChars = len;
if (mc->text && (mc->rightStatus == MAF_INSERT_STATUS) && (masterMc->start + masterMc->size < winEnd))
haveInserts = TRUE;
#ifdef REVERSESTRAND
/* complement bases if hgTracks is on reverse strand */
if (mc->size && cartCgiUsualBoolean(cart, COMPLEMENT_BASES_VAR, FALSE))
complement(mc->text, maf->textSize);
#endif
}
}
/* first sequence in the alignment */
referenceText = maf->components->text;
for (lineStart = 0; lineStart < maf->textSize; lineStart = lineEnd)
{
int size;
lineEnd = lineStart + lineSize;
if (lineEnd >= maf->textSize)
lineEnd = maf->textSize;
size = lineEnd - lineStart;
initSummaryLine(summaryLine, size, '*');
for (mc = maf->components; mc != NULL; mc = mc->next)
{
char dbOnly[128], *chrom;
int s = mc->start;
int e = s + mc->size;
char *org;
char *revComp = "";
char strand = mc->strand;
struct dyString *dy = newDyString(512);
#ifdef REVERSESTRAND
if (cartCgiUsualBoolean(cart, COMPLEMENT_BASES_VAR, FALSE))
strand = (strand == '+') ? '-' : '+';
#endif
if (strand == '-') revComp = "&hgSeq.revComp=on";
dyStringClear(dy);
safef(dbOnly, sizeof(dbOnly), "%s", mc->src);
chrom = chopPrefix(dbOnly);
if ((org = hOrganism(dbOnly)) == NULL)
org = dbOnly;
if (mc->strand == '-')
reverseIntRange(&s, &e, mc->srcSize);
if (mc->text != NULL)
{
if (lineStart == 0)
{
if (hDbIsActive(dbOnly))
{
dyStringPrintf(dy, "%s Browser %s:%d-%d %c %*dbps",hOrganism(dbOnly),chrom, s+1, e, mc->strand,sizeChars, mc->size);
linkToOtherBrowserTitle(dbOnly, chrom, s, e, dy->string);
dyStringClear(dy);
fprintf(f, "B</A> ");
}
else
fprintf(f, " ");
if (hDbExists(dbOnly))
{
dyStringPrintf(dy, "Get %s DNA %s:%d-%d %c %*dbps",hOrganism(dbOnly),chrom, s+1, e, mc->strand,sizeChars, mc->size);
printf("<A TITLE=\"%s\" TARGET=\"_blank\" HREF=\"%s?o=%d&g=getDna&i=%s&c=%s&l=%d&r=%d&db=%s%s\">D</A> ", dy->string,hgcName(),
s, cgiEncode(chrom),
chrom, s, e, dbOnly, revComp);
}
else
fprintf(f, " ");
}
else
{
fprintf(f, " ");
}
dyStringClear(dy);
dyStringPrintf(dy, "%s:%d-%d %c %*dbps",chrom, s+1, e, mc->strand,sizeChars, mc->size);
fprintf(f, "<A TITLE=\"%s\"> %*s </A> ", dy->string, srcChars, org);
updateSummaryLine(summaryLine, referenceText + lineStart, mc->text + lineStart, size);
blueCapWrite(f, mc->text + lineStart, size,
(onlyDiff && mc != maf->components) ? referenceText + lineStart : NULL);
fprintf(f, "\n");
}
else
{
if (((mc->leftStatus == MAF_CONTIG_STATUS) && (mc->rightStatus == MAF_CONTIG_STATUS) )
|| ((mc->leftStatus == MAF_TANDEM_STATUS) && (mc->rightStatus == MAF_TANDEM_STATUS) )
|| ((mc->leftStatus == MAF_INSERT_STATUS) && (mc->rightStatus == MAF_INSERT_STATUS) )
|| ((mc->leftStatus == MAF_MISSING_STATUS) && (mc->rightStatus == MAF_MISSING_STATUS) ))
{
if (lineStart == 0)
{
int s = mc->start;
int e = s + mc->rightLen;
struct dyString *dy = newDyString(512);
if (mc->strand == '-')
reverseIntRange(&s, &e, mc->srcSize);
if ( hDbIsActive(dbOnly))
{
dyStringPrintf(dy, "%s Browser %s:%d-%d %c %d bps Unaligned",hOrganism(dbOnly),chrom, s+1, e, mc->strand, e-s);
linkToOtherBrowserTitle(dbOnly, chrom, s, e, dy->string);
fprintf(f,"B</A> ");
dyStringClear(dy);
}
else
fprintf(f," ");
if (hDbExists(dbOnly))
{
dyStringPrintf(dy, "Get %s DNA %s:%d-%d %c %d bps Unaligned",hOrganism(dbOnly),chrom, s+1, e, mc->strand, e-s);
printf("<A TITLE=\"%s\" TARGET=\"_blank\" HREF=\"%s?o=%d&g=getDna&i=%s&c=%s&l=%d&r=%d&db=%s%s\">D</A> ", dy->string, hgcName(),
s, cgiEncode(chrom),
chrom, s, e, dbOnly,revComp);
}
else
fprintf(f, " ");
}
else
fprintf(f, " ");
initSummaryLine(summaryLine, size, ' ');
dyStringClear(dy);
dyStringPrintf(dy, "%s:%d-%d %c %*dbps",chrom, s+1, e, mc->strand,sizeChars, mc->size);
fprintf(f, "<A TITLE=\"%s\">%*s</A> ", dy->string, srcChars, org);
ch = '-';
switch(mc->rightStatus)
{
case MAF_INSERT_STATUS:
ch = '=';
break;
case MAF_MISSING_STATUS:
ch = 'N';
break;
case MAF_TANDEM_STATUS:
case MAF_CONTIG_STATUS:
ch = '-';
break;
}
for(ii=lineStart; ii < lineEnd ; ii++)
fputc(ch,f);
fprintf(f,"\n");
}
}
}
#ifdef ADDMATCHLINE
if (lineStart == 0)
fprintf(f, " %-*s %s\n", srcChars, "", summaryLine);
else
fprintf(f, "%-*s %s\n", srcChars, "", summaryLine);
#else
fprintf(f, "\n");
#endif
}
if (haveInserts)
{
fprintf(f, "<B>Inserts between block %d and %d in window</B>\n",blockNo, blockNo+1);
for (mc = maf->components; mc != NULL; mc = mc->next)
{
char dbOnly[128], *chrom;
int s = mc->start + mc->size;
int e = s + mc->rightLen;
char *org;
if (mc->text == NULL)
continue;
if (mc->strand == '-')
reverseIntRange(&s, &e, mc->srcSize);
safef(dbOnly, sizeof(dbOnly), "%s", mc->src);
chrom = chopPrefix(dbOnly);
if ((org = hOrganism(dbOnly)) == NULL)
org = dbOnly;
if (mc->rightStatus == MAF_INSERT_STATUS)
{
char *revComp = "";
if (hDbIsActive(dbOnly))
{
char strand = mc->strand;
#ifdef REVERSESTRAND
if (cartCgiUsualBoolean(cart, COMPLEMENT_BASES_VAR, FALSE))
strand = (strand == '+') ? '-' : '+';
#endif
if (strand == '-') revComp = "&hgSeq.revComp=on";
linkToOtherBrowser(dbOnly, chrom, s, e);
fprintf(f,"B");
fprintf(f, "</A>");
fprintf(f, " ");
}
else
fprintf(f, " ");
if (hDbExists(dbOnly))
{
printf("<A TARGET=\"_blank\" HREF=\"%s?o=%d&g=getDna&i=%s&c=%s&l=%d&r=%d"
"&db=%s%s\">D</A> ", hgcName(), s, cgiEncode(chrom), chrom, s,
e, dbOnly,revComp);
}
else
fprintf(f, " ");
fprintf(f, "%*s %dbp\n", srcChars, org,mc->rightLen);
}
}
fprintf(f, "\n");
}
freeMem(summaryLine);
}
