int loadGeneFile(const char* fn, const char* gene,
OrderedMap<std::string, RangeList>* geneMap) {
std::set<std::string> geneSet;
makeSet(gene, ',', &geneSet);
OrderedMap<std::string, RangeList>& m = *geneMap;
LineReader lr(fn);
int lineNo = 0;
std::vector<std::string> fd;
while (lr.readLineBySep(&fd, "\t ")) {
++lineNo;
if (fd.size() < 6) {
logger->error(
"Skip %d line (short of columns) in gene file [ %s ], is gene file "
"format correct?",
lineNo, fn);
break;
}
std::string& geneName = fd[0];
if (geneSet.size() && geneSet.find(geneName) == geneSet.end()) continue;
std::string chr = chopChr(fd[2]);
int beg = atoi(fd[4]);
int end = atoi(fd[5]);
m[geneName].addRange(chr.c_str(), beg, end);
}
return m.size();
}
void RangeList::filterGeneName(const char* inclusionGeneFileName, const char* geneTableFileName){
// require user input gene list file
if (strlen(geneTableFileName) == 0 && strlen(inclusionGeneFileName) != 0) {
REprintf("Please provide gene list file (e.g. refFlat) until we are able to process gene\n");
return;
//exit(1);
}
// if not specify any gene, return whole range.
if (strlen(inclusionGeneFileName) == 0) {
return;
}
// store which gene do we want if specified
std::set< std::string > inclusionSet;
LineReader lr(inclusionGeneFileName);
std::string gene;
while(lr.readLine(&gene)) {
inclusionSet.insert(gene);
}
std::vector<std::string> fields;
std::string chr;
std::string geneNameTbl;
LineReader geneTable(geneTableFileName);
while (geneTable.readLineBySep(&fields, "\t ")) {
geneNameTbl = fields[0];
if (inclusionSet.find(geneNameTbl) != inclusionSet.end()){ // store gene range
chr = chopChr(fields[2].c_str());
this->rangeCollection.addRange(chr,
atoi(fields[4].c_str()), // start
atoi(fields[5].c_str())); // end
}
}
if (this->rangeCollection.size() == 0){
Rprintf("We cannot find given gene in your geneListFile, so all sites will be outputed\n");
}
}
int main(int argc, char** argv) {
time_t currentTime = time(0);
fprintf(stderr, "Analysis started at: %s", ctime(¤tTime));
PARSE_PARAMETER(argc, argv);
PARAMETER_STATUS();
if (FLAG_REMAIN_ARG.size() > 0) {
fprintf(stderr, "Unparsed arguments: ");
for (unsigned int i = 0; i < FLAG_REMAIN_ARG.size(); i++) {
fprintf(stderr, " %s", FLAG_REMAIN_ARG[i].c_str());
}
fprintf(stderr, "\n");
abort();
}
REQUIRE_STRING_PARAMETER(FLAG_inVcf,
"Please provide input file using: --inVcf");
const char* fn = FLAG_inVcf.c_str();
VCFInputFile vin(fn);
// set range filters here
// e.g.
// vin.setRangeList("1:69500-69600");
vin.setRangeList(FLAG_rangeList.c_str());
vin.setRangeFile(FLAG_rangeFile.c_str());
// set people filters here
if (FLAG_peopleIncludeID.size() || FLAG_peopleIncludeFile.size()) {
vin.excludeAllPeople();
vin.includePeople(FLAG_peopleIncludeID.c_str());
vin.includePeopleFromFile(FLAG_peopleIncludeFile.c_str());
}
vin.excludePeople(FLAG_peopleExcludeID.c_str());
vin.excludePeopleFromFile(FLAG_peopleExcludeFile.c_str());
// let's write it out.
if (FLAG_updateId != "") {
int ret = vin.updateId(FLAG_updateId.c_str());
fprintf(stdout, "%d samples have updated id.\n", ret);
}
// load gene ranges
std::map<std::string, std::string> geneRange;
if (FLAG_geneName.size()) {
if (FLAG_geneFile.size() == 0) {
fprintf(stderr, "Have to provide --geneFile to extract by gene.\n");
abort();
}
LineReader lr(FLAG_geneFile);
std::vector<std::string> fd;
while (lr.readLineBySep(&fd, "\t ")) {
if (FLAG_geneName != fd[0]) continue;
fd[2] = chopChr(fd[2]); // chop "chr1" to "1"
if (geneRange.find(fd[0]) == geneRange.end()) {
geneRange[fd[0]] = fd[2] + ":" + fd[4] + "-" + fd[5];
} else {
geneRange[fd[0]] += "," + fd[2] + ":" + fd[4] + "-" + fd[5];
}
};
}
std::string range;
for (std::map<std::string, std::string>::iterator it = geneRange.begin();
it != geneRange.end(); it++) {
if (range.size() > 0) {
range += ",";
}
range += it->second;
};
fprintf(stderr, "range = %s\n", range.c_str());
vin.setRangeList(range.c_str());
Regex regex;
if (FLAG_annoType.size()) {
regex.readPattern(FLAG_annoType);
}
// print header
std::vector<std::string> names;
vin.getVCFHeader()->getPeopleName(&names);
printf("CHROM\tPOS");
for (unsigned int i = 0; i < names.size(); i++) {
printf("\t%s", names[i].c_str());
}
printf("\n");
// real working part
int nonVariantSite = 0;
while (vin.readRecord()) {
VCFRecord& r = vin.getVCFRecord();
VCFPeople& people = r.getPeople();
VCFIndividual* indv;
if (FLAG_variantOnly) {
bool hasVariant = false;
int geno;
int GTidx = r.getFormatIndex("GT");
for (size_t i = 0; i < people.size(); i++) {
indv = people[i];
geno = indv->justGet(GTidx).getGenotype();
if (geno != 0 && geno != MISSING_GENOTYPE) hasVariant = true;
}
if (!hasVariant) {
nonVariantSite++;
continue;
}
}
if (FLAG_annoType.size()) {
bool isMissing = false;
const char* tag = r.getInfoTag("ANNO", &isMissing).toStr();
if (isMissing) continue;
// fprintf(stdout, "ANNO = %s", tag);
bool match = regex.match(tag);
// fprintf(stdout, " %s \t", match ? "match": "noMatch");
// fprintf(stdout, " %s \n", exists ? "exists": "missing");
if (!match) {
continue;
}
}
fprintf(stdout, "%s\t%s", r.getChrom(), r.getPosStr());
for (size_t i = 0; i < people.size(); i++) {
indv = people[i];
fprintf(stdout, "\t%d", indv->justGet(0).getGenotype());
}
fprintf(stdout, "\n");
};
currentTime = time(0);
fprintf(stderr, "Analysis ends at: %s", ctime(¤tTime));
return 0;
};
