int main(int argc, char *argv[])
{
FILE *inf;
char line[1000];
int i;
char *infileName;
if (argc != 3) usage();
proteinId = argv[1];
infileName = argv[2];
for (i=0; i<8; i++) previousWord[i] = strdup("n/a");
inf = mustOpen(infileName, "r");
/* skip initial 2 header lines in .rdb format file */
mustGetLine(inf, line, sizeof(line));
mustGetLine(inf, line, sizeof(line));
/* read and process all lines one by one */
while (fgets(line, sizeof(line), inf) != NULL)
{
*(line + strlen(line) - 1) = '\0';
processOneLine(line);
}
return(0);
}
static boolean findAltSpliceRange(char *name, struct snof *snof, FILE *f,
char **retChrom, int *retStart, int *retEnd, char *retStrand)
/* Return range of chromosome covered by a gene and all of it's isoforms. */
{
char baseName[64];
char bName[64];
int snIx, maxIx;
int start = 0x7fffffff;
int end = -start;
char lineBuf[128];
char *words[3];
int wordCount;
int baseNameSize;
strcpy(baseName, name);
makeIsoformBaseName(baseName);
baseNameSize = strlen(baseName);
if (!snofFindFirstStartingWith(snof, baseName, baseNameSize, &snIx))
return FALSE;
maxIx = snofElementCount(snof);
for (;snIx < maxIx; ++snIx)
{
long offset;
char *geneName;
snofNameOffsetAtIx(snof, snIx, &geneName, &offset);
if (strncmp(geneName, baseName, baseNameSize) != 0)
break;
strcpy(bName, geneName);
makeIsoformBaseName(bName);
if (sameString(baseName, bName))
{
int s, e;
fseek(f, offset, SEEK_SET);
mustGetLine(f, lineBuf, sizeof(lineBuf));
wordCount = chopLine(lineBuf, words);
assert(wordCount == 3);
wormParseChromRange(words[0], retChrom, &s, &e);
*retStrand = words[1][0];
if (start > s)
start = s;
if (end < e)
end = e;
}
}
*retStart = start;
*retEnd = end;
return TRUE;
}
boolean wormCdnaInfo(char *name, struct wormCdnaInfo *retInfo)
/* Get info about cDNA sequence. */
{
char commentBuf[512];
char *comment;
long offset;
wormCdnaCache();
if (!snofFindOffset(cdnaSnof, name, &offset))
return FALSE;
fseek(cdnaFa, offset, SEEK_SET);
mustGetLine(cdnaFa, commentBuf, sizeof(commentBuf));
if (commentBuf[0] != '>')
errAbort("Expecting line starting with > in cDNA fa file.\nGot %s", commentBuf);
comment = cloneString(commentBuf);
wormFaCommentIntoInfo(comment, retInfo);
return TRUE;
}
int main(int argc, char *argv[])
{
struct sqlConnection *conn, *conn2;
char query2[256];
struct sqlResult *sr2;
char **row2;
char cond_str[255];
char *proteinDatabaseName;
FILE *o1, *o2, *o3;
FILE *fh[23];
char temp_str[1000];;
char *accession;
char *aaSeq;
char *chp;
int i, j, len;
int ihi, ilow;
char *answer;
char *protDisplayId;
int aaResCnt[30];
char aaAlphabet[30];
int aaResFound;
float fvalue1, fvalue2;
float p1, p2;
int icnt, jcnt;
char *taxon;
char *database;
int sortedCnt;
if (argc != 4) usage();
strcpy(aaAlphabet, "WCMHYNFIDQKRTVPGEASLXZB");
proteinDatabaseName = argv[1];
taxon = argv[2];
database = argv[3];
o2 = mustOpen("pbResAvgStd.tab", "w");
for (i=0; i<20; i++)
{
safef(temp_str, sizeof(temp_str), "%c.txt", aaAlphabet[i]);
fh[i] = mustOpen(temp_str, "w");
}
conn = hAllocConn(hDefaultDb());
conn2 = hAllocConn(hDefaultDb());
safef(query2, sizeof(query2), "select proteinID from %s.knownGene;", database);
sr2 = sqlMustGetResult(conn2, query2);
row2 = sqlNextRow(sr2);
icnt = 0;
jcnt = 0;
for (j=0; j<MAXRES; j++)
{
sumJ[j] = 0;
}
while (row2 != NULL)
{
protDisplayId = row2[0];
safef(cond_str, sizeof(cond_str), "val='%s'", protDisplayId);
accession = sqlGetField(proteinDatabaseName, "displayId", "acc", cond_str);
if (accession == NULL)
{
safef(cond_str, sizeof(cond_str), "acc='%s'", protDisplayId);
accession = sqlGetField(proteinDatabaseName, "displayId", "acc", cond_str);
if (accession == NULL)
{
verbose(2, "'%s' not found.\n", protDisplayId);
goto skip;
}
}
safef(cond_str, sizeof(cond_str), "accession='%s'", accession);
answer = sqlGetField("proteins040115", "spXref2", "biodatabaseID", cond_str);
if (answer == NULL)
{
/* this protein might be a variant splice protein, and then it won't be in spXref2 */
goto skip;
}
if (answer[0] != '1')
{
/* printf("%s not in SWISS-PROT\n", protDisplayId);fflush(stdout); */
goto skip;
}
safef(cond_str, sizeof(cond_str), "acc='%s'", accession);
aaSeq = sqlGetField(proteinDatabaseName, "protein", "val", cond_str);
if (aaSeq == NULL)
{
printf("Can't find peptide sequence for %s, exiting ...\n", protDisplayId);
fflush(stdout);
exit(1);
}
len = strlen(aaSeq);
if (len < 100) goto skip;
lenDouble = (double)len;
for (j=0; j<MAXRES; j++)
{
aaResCnt[j] = 0;
}
chp = aaSeq;
for (i=0; i<len; i++)
{
aaResFound = 0;
for (j=0; j<MAXRES; j++)
{
if (*chp == aaAlphabet[j])
{
aaResFound = 1;
aaResCnt[j] ++;
}
}
if (!aaResFound)
{
fprintf(stderr, "%c %d not a valid AA residue.\n", *chp, *chp);
}
chp++;
}
for (j=0; j<MAXRES; j++)
{
freq[icnt][j] = (double)aaResCnt[j]/lenDouble;
sumJ[j] = sumJ[j] + freq[icnt][j];
}
for (j=0; j<20; j++)
{
fprintf(fh[j], "%15.7f\t%s\n", freq[icnt][j], accession);
fflush(fh[j]);
}
icnt++;
if (icnt >= MAXN)
errAbort("Too many proteins - please set MAXN to be more than %d\n", MAXN);
skip:
row2 = sqlNextRow(sr2);
}
recordCnt = icnt;
recordCntDouble = (double)recordCnt;
for (j=0; j<20; j++)
{
carefulClose(&(fh[j]));
}
sqlFreeResult(&sr2);
hFreeConn(&conn);
hFreeConn(&conn2);
for (j=0; j<MAXRES; j++)
{
avg[j] = sumJ[j]/recordCntDouble;
}
for (j=0; j<20; j++)
{
sum = 0.0;
for (i=0; i<recordCnt; i++)
{
sum = sum + (freq[i][j] - avg[j]) * (freq[i][j] - avg[j]);
}
sigma[j] = sqrt(sum/(double)(recordCnt-1));
fprintf(o2, "%c\t%f\t%f\n", aaAlphabet[j], avg[j], sigma[j]);
}
carefulClose(&o2);
o1 = mustOpen("pbAnomLimit.tab", "w");
for (j=0; j<20; j++)
{
safef(temp_str, sizeof(temp_str), "cat %c.txt|sort|uniq > %c.srt", aaAlphabet[j], aaAlphabet[j]);
mustSystem(temp_str);
/* figure out how many unique entries */
safef(temp_str, sizeof(temp_str), "wc %c.srt > %c.tmp", aaAlphabet[j], aaAlphabet[j]);
mustSystem(temp_str);
safef(temp_str, sizeof(temp_str), "%c.tmp", aaAlphabet[j]);
o3 = mustOpen(temp_str, "r");
mustGetLine(o3, temp_str, 1000);
chp = temp_str;
while (*chp == ' ') chp++;
while (*chp != ' ') chp++;
*chp = '\0';
sscanf(temp_str, "%d", &sortedCnt);
safef(temp_str, sizeof(temp_str), "rm %c.tmp", aaAlphabet[j]);
mustSystem(temp_str);
/* cal hi and low cutoff threshold */
ilow = (int)((float)sortedCnt * 0.025);
ihi = (int)((float)sortedCnt * 0.975);
safef(temp_str, sizeof(temp_str), "%c.srt", aaAlphabet[j]);
o2 = mustOpen(temp_str, "r");
i=0;
for (i=0; i<ilow; i++)
{
mustGetLine(o2, temp_str, 1000);
}
sscanf(temp_str, "%f", &fvalue1);
mustGetLine(o2, temp_str, 1000);
sscanf(temp_str, "%f", &fvalue2);
p1 = (fvalue1 + fvalue2)/2.0;
for (i=ilow+1; i<ihi; i++)
{
mustGetLine(o2, temp_str, 1000);
}
sscanf(temp_str, "%f", &fvalue1);
mustGetLine(o2, temp_str, 1000);
sscanf(temp_str, "%f", &fvalue2);
p2 = (fvalue1 + fvalue2)/2.0;
carefulClose(&o2);
fprintf(o1, "%c\t%f\t%f\n", aaAlphabet[j], p1, p2);
fflush(stdout);
for (i=0; i<recordCnt; i++)
{
measure[i] = freq[i][j];
}
safef(temp_str, sizeof(temp_str), "pbAaDist%c.tab", aaAlphabet[j]);
calDist(measure, recordCnt, 51, 0.0, 0.005, temp_str);
}
carefulClose(&o1);
return(0);
}
int main(int argc, char *argv[])
{
char *id;
char *ox;
char *chp;
char *infName, *outfName;
char line[2000];
FILE *inf, *outf;
if (argc!=3)
{
usage();
}
infName = argv[1];
outfName = argv[2];
if ((inf = fopen(infName, "r")) == NULL)
{
fprintf(stderr, "Can't open file %s.\n", infName);
exit(8);
}
outf = fopen(outfName, "w");
while (fgets(line, 1000, inf) != NULL)
{
chp = strstr(line, "ID ");
if (chp != line)
{
fprintf(stderr, "expected ID line, but got: %s\n", line);
exit(1);
}
chp = chp + strlen("ID ");
id = chp;
chp = strstr(id, " ");
*chp = '\0';
id = strdup(id);
again:
if (fgets(line, 1000, inf) == NULL) break;
/* "//" is the end of record line */
if ((line[0] == '/') && (line[1] == '/')) goto one_done;
chp = strstr(line, "OX ");
if (chp != NULL)
{
chp = strstr(line, "NCBI_TaxID=");
ox = chp + strlen("NCBI_TaxID=");
again1:
chp = strstr(ox, ",");
if (chp != NULL)
{
*chp='\0';
while (*ox == ' ') ox++;
fprintf(outf, "%s\t%s\n", id, ox);
chp++;
ox = chp;
if (*ox == '\n')
{
mustGetLine(inf, line, sizeof(line));
chp = strstr(line, "OX ");
if (chp == NULL)
{
fprintf(stderr, "no OX line after OX continuation line!\n");
exit(1);
}
ox = line + strlen("OX ");
goto again1;
}
}
else
{
chp = strstr(ox, ";");
if (chp != NULL)
{
*chp = '\0';
while (*ox == ' ') ox++;
fprintf(outf, "%s\t%s\n", id, ox);
ox = NULL;
}
}
if (ox != NULL) goto again1;
}
goto again;
one_done: id = id;
}
fclose(outf);
return 0;
}
boolean wormGeneRange(char *name, char **retChrom, char *retStrand, int *retStart, int *retEnd)
/* Return chromosome position of a chrom range, gene, ORF, cosmid, or nameless cluster. */
{
static struct snof *c2gSnof = NULL, *c2cSnof = NULL;
static FILE *c2gFile = NULL, *c2cFile = NULL;
long offset;
char fileName[512];
struct slName *syn = NULL;
boolean ok;
if (wormIsChromRange(name))
{
*retStrand = '.';
return wormParseChromRange(name, retChrom, retStart, retEnd);
}
getDirs();
/* Translate biologist type name to cosmid.N name */
if (wormIsGeneName(name))
{
syn = wormGeneToOrfNames(name);
if (syn != NULL)
{
name = syn->name;
}
}
if (wormFixupOrfName(name)) /* See if ORF, and if so make nice. */
{
if (c2gSnof == NULL)
{
sprintf(fileName, "%sc2g", featDir);
c2gSnof = snofMustOpen(fileName);
sprintf(fileName, "%sc2g", featDir);
c2gFile = mustOpen(fileName, "rb");
}
ok = findAltSpliceRange(name, c2gSnof, c2gFile, retChrom, retStart, retEnd, retStrand);
}
else /* Lets say it's a cosmid. */
{
char lineBuf[128];
char *words[3];
int wordCount;
touppers(name);
if (c2cSnof == NULL)
{
sprintf(fileName, "%sc2c", featDir);
c2cSnof = snofMustOpen(fileName);
sprintf(fileName, "%sc2c", featDir);
c2cFile = mustOpen(fileName, "rb");
}
if (!snofFindOffset(c2cSnof, name, &offset) )
return FALSE;
fseek(c2cFile, offset, SEEK_SET);
mustGetLine(c2cFile, lineBuf, sizeof(lineBuf));
wordCount = chopLine(lineBuf, words);
assert(wordCount == 3);
assert(strcmp(words[2], name) == 0);
assert(wormIsChromRange(words[0]));
*retStrand = words[1][0];
ok = wormParseChromRange(words[0], retChrom, retStart, retEnd);
}
slFreeList(&syn);
return ok;
}