void mapAltGraphXFile(struct sqlConnection *conn, char *db, char *orthoDb, char *chrom,
char *netTable, char *altGraphXFileName, char *altGraphXTableName,
FILE *agxOut, FILE *selectedOut, int *foundCount, int *notFoundCount)
/* Map over altGraphX Structures from one organism to
another. Basically create a mapping for the vertices and then reverse
them if on '-' strand.*/
{
int count =0;
struct bed *bed = NULL;
struct altGraphX *agList = NULL, *ag = NULL, *agNew = NULL;
if(altGraphXFileName != NULL)
{
warn("Loading altGraphX Records from file %s.", altGraphXFileName);
agList = altGraphXLoadAll(altGraphXFileName);
}
else if(altGraphXTableName != NULL)
{
char query[256];
warn("Reading altGraphX Records from table %s.", altGraphXTableName);
sqlSafef(query, sizeof(query), "select * from %s where tName like '%s'", altGraphXTableName, chrom);
agList = altGraphXLoadByQuery(conn, query);
}
else
errAbort("orthoMap::mapAlGraphXFile() - Need a table name or file name to load altGraphX records");
warn("Mapping altGraphX records.");
for(ag = agList; ag != NULL; ag = ag->next)
{
if(differentString(ag->tName, chrom))
continue;
occassionalDot();
agNew = mapAltGraphX(ag, conn, db, netTable);
if(agNew == NULL)
(*notFoundCount)++;
else
{
(*foundCount)++;
altGraphXTabOut(agNew, agxOut);
altGraphXFree(&agNew);
if (selectedOut != NULL)
altGraphXTabOut(ag, selectedOut);
}
count++;
}
}
int main(int argc, char *argv[])
{
struct altGraphX *agList = NULL;
int cassetteCount = 0;
float minConfidence = 0;
char *bedFileName = NULL;
char *faFile = NULL;
FILE *faOut = NULL;
FILE *bedOut = NULL;
boolean mrnaFilter = FALSE;
float estPrior = 0.0;
int minSize = 0;
if(argc < 4)
usage();
cgiSpoof(&argc, argv);
warn("Loading graphs.");
agList = altGraphXLoadAll(argv[1]);
bedFileName = cgiOptionalString("bedFile");
minConfidence = cgiDouble("minConf");
db = cgiString("db");
faFile = cgiOptionalString("faFile");
estPrior = cgiOptionalDouble("estPrior", 10);
minSize = cgiOptionalInt("minSize", 0);
mrnaFilter = cgiBoolean("mrnaFilter");
if(mrnaFilter)
loadMrnaHash();
warn("Counting cassette exons from %d clusters above confidence: %f", slCount(agList), minConfidence);
if(bedFileName != NULL)
{
bedOut = mustOpen(bedFileName, "w");
printCommandState(argc, argv, bedOut);
fprintf(bedOut, "track name=cass_conf-%4.2f_est-%3.2f description=\"spliceStats minConf=%4.2f estPrior=%3.2f minSize=%d\"\n",
minConfidence, estPrior, minConfidence, estPrior, minSize);
}
if(faFile != NULL)
faOut = mustOpen(faFile, "w");
cassetteCount = countCassetteExons(agList, minConfidence, faOut,bedOut );
carefulClose(&faOut);
carefulClose(&bedOut);
warn("%d cassette exons out of %d clusters in %s", cassetteCount, slCount(agList), argv[1]);
altGraphXFreeList(&agList);
return 0;
}
void agxStartsEnds(char *agxFile, char *softBedFile, char *hardBedFile)
/* Loop through each graph and output the exons with soft starts or ends. */
{
FILE *softOut = mustOpen(softBedFile, "w");
FILE *hardOut = mustOpen(hardBedFile, "w");
struct altGraphX *agxList = NULL, *agx = NULL;
int i;
int softFoundCount = 0;
int hardFoundCount = 0;
int totalCount = 0;
warn("Loading splice graphs");
agxList = altGraphXLoadAll(agxFile);
warn("Looking for transcription starts and ends.");
for(agx = agxList; agx != NULL; agx = agx->next)
{
int *vPos = agx->vPositions;
int *starts = agx->edgeStarts;
int *ends = agx->edgeEnds;
for(i = 0; i < agx->edgeCount; i++)
{
if(isSoftExon(agx, i))
{
fprintf(softOut, "%s\t%d\t%d\t%s.%d\t0\t%s\n",
agx->tName, vPos[starts[i]], vPos[ends[i]], agx->name, totalCount++, agx->strand);
softFoundCount++;
}
else if(getSpliceEdgeType(agx,i) == ggExon)
{
fprintf(hardOut, "%s\t%d\t%d\t%s.%d\t0\t%s\n",
agx->tName, vPos[starts[i]], vPos[ends[i]], agx->name, totalCount++, agx->strand);
hardFoundCount++;
}
}
}
warn("Found %d exons that start or end a transcript in %d total graphs. %.2f per graph",
softFoundCount, slCount(agxList), (double)softFoundCount/slCount(agxList));
warn("Found %d internal exon in %d total graphs. %.2f per graph",
hardFoundCount, slCount(agxList), (double)hardFoundCount/slCount(agxList));
altGraphXFreeList(&agxList);
carefulClose(&softOut);
carefulClose(&hardOut);
}
void createIntronBeds(char *agxFile, char *bedFile)
/* Make intron beds for evaluation. */
{
struct altGraphX *ag=NULL, *agList = NULL;
struct bed *bed=NULL, *bedList=NULL;
FILE *bedOut = NULL;
int count;
warn("Rading AltGraphX list.");
agList = altGraphXLoadAll(agxFile);
warn("Converting to intron beds.");
bedOut = mustOpen(bedFile, "w");
for(ag = agList; ag != NULL; ag = ag->next)
{
occassionalDot();
bedList = bedIntronsFromAgx(ag);
for(bed=bedList; bed != NULL; bed=bed->next)
{
bedTabOutN(bed, 12, bedOut);
}
bedFreeList(&bedList);
}
altGraphXFreeList(&agList);
}
void altSummary(char *db, char *agxFileName, char *summaryOutName, char *htmlOutName, char *htmlFramesOutName)
/* Look through a bunch of splice sites and output some statistics and links. */
{
struct altGraphX *agList = NULL, *ag = NULL;
struct altSpliceSite *aSpliceList = NULL, *aSplice=NULL;
char *RDataName = optionVal("RData", NULL);
char *bedName = optionVal("bedName", NULL);
FILE *htmlOut = NULL;
FILE *htmlFramesOut = NULL;
FILE *summaryOut = NULL;
int altSpliceSites = 0, altSpliceLoci = 0, totalSpliceSites = 0;
warn("Loading splicing graphs.");
agList = altGraphXLoadAll(agxFileName);
htmlFramesOut = mustOpen(htmlFramesOutName, "w");
htmlOut = mustOpen(htmlOutName, "w");
summaryOut = mustOpen(summaryOutName, "w");
if(RDataName != NULL)
{
char buff[256];
safef(buff, sizeof(buff), "%s.control", RDataName);
RDataCont = mustOpen(buff, "w");
outputRHeader(RDataCont);
safef(buff, sizeof(buff), "%s.alt", RDataName);
RData = mustOpen(buff, "w");
outputRHeader(RData);
}
if(bedName != NULL)
{
openBedFiles(bedName);
}
writeOutFrames(htmlFramesOut, htmlOutName, db);
carefulClose(&htmlFramesOut);
warn("Examining splicing graphs.");
fprintf(htmlOut, "<html>\n<body bgcolor=\"#FFF9D2\"><b>Alt-Splice List</b>\n"
"<table border=1><tr><th>Name (count)</th><th>Type</th><th>Size</th></tr>\n");
for(ag=agList; ag != NULL; ag=ag->next)
{
lookForAltSplicing(db, ag, &aSpliceList, &altSpliceSites, &altSpliceLoci, &totalSpliceSites);
for(aSplice=aSpliceList; aSplice != NULL; aSplice= aSplice->next)
{
altSpliceSiteOutput(aSplice, summaryOut, '\t', '\n');
htmlLinkOut(db, aSplice, htmlOut);
if(bedViewOutFile != NULL)
bedViewOut(aSplice, bedViewOutFile);
}
altSpliceSiteFreeList(&aSpliceList);
}
warn("\nDone.");
fprintf(htmlOut,"</body></html>\n");
warn("%d altSpliced sites in %d alt-spliced loci out of %d total loci.",
altSpliceSites, altSpliceLoci, slCount(agList));
printSpliceTypeInfo(altSpliceLoci);
altGraphXFreeList(&agList);
if(RData != NULL)
{
carefulClose(&RData);
carefulClose(&RDataCont);
}
carefulClose(&htmlOut);
carefulClose(&summaryOut);
}
