void tabPepPred(char *database, int fileCount, char *fileNames[], char *table)
/* Load a tab separated peptide file. */
{
struct hash *uniq = newHash(16);
struct lineFile *lf = lineFileOpen(fileNames[0], TRUE);
char *words[2];
if (fileCount != 1)
errAbort("Only one file allowed for tab separated peptides");
makeCustomTable(database, table, createString);
printf("Processing %s\n", fileNames[0]);
while (lineFileRow(lf, words))
{
char *upperCase;
if (hashLookupUpperCase(uniq, words[0]) != NULL)
errAbort("Duplicate (case insensitive) '%s' line %d of %s", words[0], lf->lineIx, lf->fileName);
upperCase = cloneString(words[0]);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
}
lineFileClose(&lf);
printf("Loading %s\n", fileNames[0]);
loadTableFromTabFile(database, table, fileNames[0]);
freeHash(&uniq);
}
void *hashFindValUpperCase(struct hash *hash, char *name)
/* Lookup upper cased name in hash and return val or return NULL if not found.
* (Assumes all elements of hash are themselves already in upper case.) */
{
struct hashEl *hel = hashLookupUpperCase(hash, name);
if (hel == NULL)
return NULL;
return hel->val;
}
void oneGenieFile(char *fileName, struct hash *uniq, FILE *f)
/* Process one genie peptide prediction file into known and alt tab files. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *trans;
boolean skip = FALSE;
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", fileName);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", fileName);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
trans = firstWordInLine(line+1);
if (abbr != NULL && startsWith(abbr, trans))
trans += strlen(abbr);
if (hashLookupUpperCase(uniq, trans) != NULL)
{
warn("Duplicate (case insensitive) '%s' line %d of %s. Ignoring all but first.", trans, lf->lineIx, lf->fileName);
skip = TRUE;
}
else
{
char *upperCase;
upperCase = cloneString(trans);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
fprintf(f, "%s\t", trans);
skip = FALSE;
}
}
else if (!skip)
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
void genericOne(char *fileName, struct hash *uniq, FILE *f)
/* Process one ensemble peptide prediction file into tab delimited
* output f, using uniq hash to make sure no dupes. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *trans, transBuf[128];
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", fileName);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", fileName);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
char *upperCase;
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
trans = firstWordInLine(line+1);
if (abbr != NULL && startsWith(abbr, trans))
trans += strlen(abbr);
if (suffix != NULL)
{
safef(transBuf, sizeof(transBuf), "%s%s", trans, suffix);
trans = transBuf;
}
if (hashLookupUpperCase(uniq, trans) != NULL)
errAbort("Duplicate (case insensitive) '%s' line %d of %s", trans, lf->lineIx, lf->fileName);
upperCase = cloneString(trans);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
fprintf(f, "%s\t", trans);
}
else
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
void oneEnsFile(char *ensFile, struct hash *uniq, struct hash *pToT, FILE *f)
/* Process one ensemble peptide prediction file into tab delimited
* output f, using uniq hash to make sure no dupes. */
{
struct lineFile *lf = lineFileOpen(ensFile, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *translation;
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", ensFile);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", ensFile);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
char *upperCase;
char *transcript;
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
translation = findEnsTrans(lf, line);
if (hashLookupUpperCase(uniq, translation) != NULL)
errAbort("Duplicate (case insensitive) '%s' line %d of %s", translation, lf->lineIx, lf->fileName);
upperCase = cloneString(translation);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
transcript = hashFindVal(pToT, translation);
if (transcript == NULL)
errAbort("Can't find transcript for %s", translation);
fprintf(f, "%s\t", transcript);
}
else
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
struct subjInfo *stringAdvFilter(struct column *col,
struct sqlConnection *conn, struct subjInfo *list)
/* Do advanced filter on string in main table. */
{
char *wild = advFilterVal(col, "wild");
struct hash *keyHash = keyFileHash(col);
if (keyHash != NULL)
{
struct subjInfo *newList = NULL, *next, *si;
for (si = list; si != NULL; si = next)
{
char *cell = col->cellVal(col, si, conn);
next = si->next;
if (hashLookupUpperCase(keyHash, cell))
{
slAddHead(&newList, si);
}
freez(&cell);
}
slReverse(&newList);
list = newList;
}
if (wild != NULL)
{
boolean orLogic = advFilterOrLogic(col, "logic", TRUE);
struct subjInfo *newList = NULL, *next, *si;
struct slName *wildList = stringToSlNames(wild);
for (si = list; si != NULL; si = next)
{
char *cell = col->cellVal(col, si, conn);
next = si->next;
if (wildMatchList(cell, wildList, orLogic))
{
slAddHead(&newList, si);
}
freez(&cell);
}
slReverse(&newList);
list = newList;
}
hashFree(&keyHash);
return list;
}
