jkOwnLib/gfClientLib.c File Reference

#include "common.h"
#include "hash.h"
#include "linefile.h"
#include "obscure.h"
#include "dnaseq.h"
#include "fa.h"
#include "nib.h"
#include "twoBit.h"
#include "repMask.h"
#include "gfClientLib.h"

Include dependency graph for gfClientLib.c:

Go to the source code of this file.

Functions
void	gfClientFileArray (char *fileName, char ***retFiles, int *retFileCount)
void	gfClientUnmask (struct dnaSeq *seqList)
static void	maskFromOut (struct dnaSeq *seqList, char *outFile, float minRepDivergence)
static void	maskNucSeqList (struct dnaSeq *seqList, char *seqFileName, char *maskType, boolean hardMask, float minRepDivergence)
bioSeq *	gfClientSeqList (int fileCount, char *files[], boolean isProt, boolean isTransDna, char *maskType, float minRepDivergence, boolean showStatus)

---

Function Documentation

void gfClientFileArray	(	char *	fileName,
		char ***	retFiles,
		int *	retFileCount
	)

Definition at line 16 of file gfClientLib.c.

References AllocArray, cloneString(), FALSE, mustOpen(), nibIsFile(), readAllWords(), sameString, TRUE, and twoBitIsSpec().

Referenced by blat().

{
boolean gotSingle = FALSE;
char *buf;              /* This will leak memory but won't matter. */

if (nibIsFile(fileName) || twoBitIsSpec(fileName)
    || sameString(fileName, "stdin"))
    gotSingle = TRUE;
/* Detect .fa files (where suffix is not standardized)
 * by first character being a '>'. */
else
    {
    FILE *f = mustOpen(fileName, "r");
    char c = fgetc(f);
    fclose(f);
    if (c == '>')
        gotSingle = TRUE;
    }
if (gotSingle)
    {
    char **files;
    *retFiles = AllocArray(files, 1);
    files[0] = cloneString(fileName);
    *retFileCount = 1;
    return;
    }
else
    {
    readAllWords(fileName, retFiles, retFileCount, &buf);
    }
}

Here is the call graph for this function:

Here is the caller graph for this function:

bioSeq* gfClientSeqList	(	int	fileCount,
		char *	files[],
		boolean	isProt,
		boolean	isTransDna,
		char *	maskType,
		float	minRepDivergence,
		boolean	showStatus
	)

Definition at line 157 of file gfClientLib.c.

References doMask, faReadAllMixed(), faReadAllPep(), faToDna(), faToProtein(), maskNucSeqList(), dnaSeq::next, NIB_BASE_NAME, NIB_MASK_MIXED, nibIsFile(), nibLoadAllMasked(), slCat(), twoBitIsSpec(), twoBitLoadAll(), and ZeroVar.

Referenced by blat().

{
int i;
char *fileName;
bioSeq *seqList = NULL, *seq;
boolean doMask = (maskType != NULL);

for (i=0; i<fileCount; ++i)
    {
    struct dnaSeq *list = NULL, sseq;
    ZeroVar(&sseq);
    fileName = files[i];
    if (nibIsFile(fileName))
        list = nibLoadAllMasked(NIB_MASK_MIXED|NIB_BASE_NAME, fileName);
    else if (twoBitIsSpec(fileName))
        list = twoBitLoadAll(fileName);
    else if (isProt)
      list = faReadAllPep(fileName);
    else
      list = faReadAllMixed(fileName);

    /* If necessary mask sequence from file. */
    if (doMask)
        {
        maskNucSeqList(list, fileName, maskType, isTransDna, minRepDivergence);
        }
    else
        {
        /* If not masking send everything to proper case here. */
        for (seq = list; seq != NULL; seq = seq->next)
            {
            if (isProt)
                faToProtein(seq->dna, seq->size);
            else
                faToDna(seq->dna, seq->size);
            }
        }
    /* Move local list to end of bigger list. */
    seqList = slCat(seqList, list);
    }
if (showStatus)
    {
    /* Total up size and sequence count and report. */
    int count = 0; 
    unsigned long totalSize = 0;
    for (seq = seqList; seq != NULL; seq = seq->next)
        {
        totalSize += seq->size;
        count += 1;
        }
    printf("Loaded %lu letters in %d sequences\n", totalSize, count);
    }
return seqList;
}

Here is the call graph for this function:

Here is the caller graph for this function:

void gfClientUnmask

(

struct dnaSeq *

seqList

)

Definition at line 52 of file gfClientLib.c.

References dnaSeq::dna, faToDna(), dnaSeq::next, and dnaSeq::size.

Referenced by blat(), and maskNucSeqList().

{
struct dnaSeq *seq;
for (seq = seqList; seq != NULL; seq = seq->next)
    faToDna(seq->dna, seq->size);
}

Here is the call graph for this function:

Here is the caller graph for this function:

static void maskFromOut	(	struct dnaSeq *	seqList,
		char *	outFile,
		float	minRepDivergence
	)			[static]

Definition at line 60 of file gfClientLib.c.

References chopLine, dnaSeq::dna, errAbort(), lineFile::fileName, freeHash(), hashAdd(), hashFindVal(), lineFileClose(), lineFileNext(), lineFileOpen(), dnaSeq::name, newHash(), dnaSeq::next, repeatMaskOut::percDel, repeatMaskOut::percDiv, repeatMaskOut::percInc, repeatMaskOut::qEnd, repeatMaskOut::qName, repeatMaskOut::qStart, repeatMaskOutStaticLoad(), dnaSeq::size, startsWith(), toUpperN(), TRUE, and warn().

Referenced by maskNucSeqList().

{
struct lineFile *lf = lineFileOpen(outFile, TRUE);
struct hash *hash = newHash(0);
struct dnaSeq *seq;
char *line;

for (seq = seqList; seq != NULL; seq = seq->next)
    hashAdd(hash, seq->name, seq);
if (!lineFileNext(lf, &line, NULL))
    errAbort("Empty mask file %s\n", lf->fileName);
if (!startsWith("There were no", line)) /* No repeats is ok. Not much work. */
    {
    if (!startsWith("   SW", line))
        errAbort("%s isn't a RepeatMasker .out file.", lf->fileName);
    if (!lineFileNext(lf, &line, NULL) || !startsWith("score", line))
        errAbort("%s isn't a RepeatMasker .out file.", lf->fileName);
    lineFileNext(lf, &line, NULL);  /* Blank line. */
    while (lineFileNext(lf, &line, NULL))
        {
        char *words[32];
        struct repeatMaskOut rmo;
        int wordCount;
        int seqSize;
        int repSize;
        wordCount = chopLine(line, words);
        if (wordCount < 14)
            errAbort("%s line %d - error in repeat mask .out file\n", lf->fileName, lf->lineIx);
        repeatMaskOutStaticLoad(words, &rmo);
        /* If repeat is more than 15% divergent don't worry about it. */
        if (rmo.percDiv + rmo.percDel + rmo.percInc <= minRepDivergence)
            {
            if((seq = hashFindVal(hash, rmo.qName)) == NULL)
                errAbort("%s is in %s but not corresponding sequence file, files out of sync?\n", 
                        rmo.qName, lf->fileName);
            seqSize = seq->size;
            if (rmo.qStart <= 0 || rmo.qStart > seqSize || rmo.qEnd <= 0 
                || rmo.qEnd > seqSize || rmo.qStart > rmo.qEnd)
                {
                warn("Repeat mask sequence out of range (%d-%d of %d in %s)\n",
                    rmo.qStart, rmo.qEnd, seqSize, rmo.qName);
                if (rmo.qStart <= 0)
                    rmo.qStart = 1;
                if (rmo.qEnd > seqSize)
                    rmo.qEnd = seqSize;
                }
            repSize = rmo.qEnd - rmo.qStart + 1;
            if (repSize > 0)
                toUpperN(seq->dna + rmo.qStart - 1, repSize);
            }
        }
    }
freeHash(&hash);
lineFileClose(&lf);
}

Here is the call graph for this function:

Here is the caller graph for this function:

static void maskNucSeqList	(	struct dnaSeq *	seqList,
		char *	seqFileName,
		char *	maskType,
		boolean	hardMask,
		float	minRepDivergence
	)			[static]

Definition at line 118 of file gfClientLib.c.

References dnaSeq::dna, gfClientUnmask(), maskFromOut(), dnaSeq::next, sameWord, dnaSeq::size, toggleCase(), and upperToN().

Referenced by gfClientSeqList().

{
struct dnaSeq *seq;
char *outFile = NULL, outNameBuf[512];

if (sameWord(maskType, "upper"))
    {
    /* Already has dna to be masked in upper case. */
    }
else if (sameWord(maskType, "lower"))
    {
    for (seq = seqList; seq != NULL; seq = seq->next)
        toggleCase(seq->dna, seq->size);
    }
else
    {
    /* Masking from a RepeatMasker .out file. */
    if (sameWord(maskType, "out"))
        {
        sprintf(outNameBuf, "%s.out", seqFileName);
        outFile = outNameBuf;
        }
    else
        {
        outFile = maskType;
        }
    gfClientUnmask(seqList);
    maskFromOut(seqList, outFile, minRepDivergence);
    }
if (hardMask)
    {
    for (seq = seqList; seq != NULL; seq = seq->next)
        upperToN(seq->dna, seq->size);
    }
}

Here is the call graph for this function:

Here is the caller graph for this function:

---