static void getFastaOffsets(struct brokenRefPepTbl *brpTbl,
struct sqlConnection *conn,
struct extFileTbl* extFileTbl,
char *faPath)
/* parse fasta file to get offsets of proteins */
{
struct gbFa *fa = gbFaOpen(faPath, "r");
char acc[GB_ACC_BUFSZ];
struct brokenRefPep *brp;
HGID extId = extFileTblGet(extFileTbl, conn, faPath);
gbVerbMsg(5, "scanning fasta: %s", faPath);
while (gbFaReadNext(fa))
{
gbVerbMsg(5, " %s: %lld", fa->id, (long long)fa->recOff);
/* save only if same acecss, version, and file (to match mrna fa) */
short ver = gbSplitAccVer(fa->id, acc);
brp = hashFindVal(brpTbl->protAccHash, acc);
if ((brp != NULL) && (ver == brp->protVer) && sameString(faPath, brp->newFaPath))
{
gbFaGetSeq(fa); /* force read of sequence data */
brp->newFaId = extId;
brp->newFaOff = fa->recOff;
brp->newSeqSize = fa->seqLen;
brp->newRecSize = fa->off-fa->recOff;
gbVerbMsg(5, " save: %s %lld for %lld\n", fa->id, (long long)fa->recOff, (long long)fa->off);
}
}
gbFaClose(&fa);
}
static void openByAccPrefix(char* accPrefix)
/* Open up the by accession prefix */
{
char *mode;
char raPath[PATH_LEN], faPath[PATH_LEN], gbIdxPath[PATH_LEN];
carefulClose(&raFile);
gbFaClose(&faFile);
if (gbIdxName != NULL)
carefulClose(&gbIdxFile);
makeAccPrefixedFile(accPrefix, raName, raPath);
mode = isFirstOpen(raPath) ? "w" : "a";
raFile = mustOpen(raPath, mode);
makeAccPrefixedFile(accPrefix, faName, faPath);
faFile = gbFaOpen(faPath, mode);
if (gbIdxName != NULL)
{
makeAccPrefixedFile(accPrefix, gbIdxName, gbIdxPath);
gbIdxFile = mustOpen(gbIdxPath, mode);
}
strcpy(gCurAccPrefix, accPrefix);
}
void copySelectedFasta(struct gbSelect* select)
/* copy FASTA records that were selected for alignment, segregating by
* native/xeno, and partitioning large files. */
{
char inFasta[PATH_LEN];
struct gbFa* inFa;
struct outFa* nativeFa = NULL;
struct outFa* xenoFa = NULL;
if (select->orgCats & GB_NATIVE)
nativeFa = outFaNew(select, GB_NATIVE);
if (select->orgCats & GB_XENO)
xenoFa = outFaNew(select, GB_XENO);
gbProcessedGetPath(select, "fa", inFasta);
gbVerbEnter(2, "copying from %s", inFasta);
inFa = gbFaOpen(inFasta, "r");
while (copyFastaRec(select, inFa, nativeFa, xenoFa))
continue;
outFaFree(&nativeFa);
outFaFree(&xenoFa);
gbFaClose(&inFa);
gbVerbLeave(2, "copying from %s", inFasta);
}
void seqDataProcessUpdate(struct gbSelect* select)
/* Get sequences for a partition and update. Partition processed index should
* be loaded and selected versions flaged. */
{
char inFasta[PATH_LEN];
struct gbFa* inFa;
gbProcessedGetPath(select, "fa", inFasta);
inFa = gbFaOpen(inFasta, "r");
while (gbFaReadNext(inFa))
processSeq(select, inFa);
gbFaClose(&inFa);
}
int main(int argc, char *argv[])
/* Check parameters, set up, loop through each GenBank file. */
{
char *gbName;
int argi = 1;
struct hash *estAuthorHash = NULL;
char *pepFa;
optionInit(&argc, argv, optionSpecs);
if (argc < 4)
usage();
gByAccPrefixSize = optionInt("byAccPrefix", 0);
gbIdxName = optionVal("gbidx", NULL);
pepFa = optionVal("pepFa", NULL);
gbType = gbParseType(optionVal("type", "mrna,est"));
gbOrg = optionVal("org", NULL);
inclXMs = optionExists("inclXMs");
if (gByAccPrefixSize > 4) /* keep small to avoid tons of open files */
errAbort("max value of -byAccPrefix is 4");
gCurAccPrefix[0] = '\0';
faName = argv[argi++];
raName = argv[argi++];
estAuthorHash = newHash(23);
kvt = newKvt(5*1024);
gbfInit();
if (pepFa != NULL)
gPepFa = gbFaOpen(pepFa,"w");
char *blackList = optionVal("blackList", NULL);
if (blackList != NULL)
blackListRanges = genbankBlackListParse(blackList);
while (argi < argc)
{
gbName = argv[argi++];
printf("Processing %s into %s and %s\n", gbName, faName, raName);
procOneGbFile(gbName, estAuthorHash);
}
gbFaClose(&faFile);
gbFaClose(&gPepFa);
carefulClose(&raFile);
carefulClose(&gbIdxFile);
return 0;
}
void outFaOpen(struct outFa* outFa)
/* Open the fasta file */
{
char ext[64];
char path[PATH_LEN];
assert(outFa->fa == NULL);
safef(ext, sizeof(ext), "%d.fa", outFa->nextPartNum);
gbAlignedGetPath(&outFa->select, ext, workDir, path);
outFa->fa = gbFaOpen(path, "w");
outFa->numSeqs = 0;
outFa->numBases = 0;
if (createPolyASizes)
{
safef(ext, sizeof(ext), "%d.polya", outFa->nextPartNum);
gbAlignedGetPath(&outFa->select, ext, workDir, path);
outFa->polyAFh = mustOpen(path, "w");
}
outFa->nextPartNum++;
}
static void setupOutputFiles(char *acc, char *org)
/* Get the output files (in globals) for a sequence, opening as needed. */
{
if (gByAccPrefixSize > 0)
{
char accPrefix[32];
strncpy(accPrefix, acc, gByAccPrefixSize);
accPrefix[gByAccPrefixSize] = '\0';
tolowers(accPrefix);
if (!sameString(accPrefix, gCurAccPrefix))
openByAccPrefix(accPrefix);
}
else
{
/* output to a single set of files */
if (raFile == NULL)
{
raFile = mustOpen(raName, "w");
faFile = gbFaOpen(faName, "w");
if (gbIdxName != NULL)
gbIdxFile = mustOpen(gbIdxName, "w");
}
}
}
void seqDataOpen(boolean inclVersion, char *outFile)
/* open output file and set options */
{
gInclVersion = inclVersion;
gOutFa = gbFaOpen(outFile, "w");
}