int main(int argc, char** argv){
time_t currentTime = time(0);
fprintf(stderr, "Analysis started at: %s", ctime(¤tTime));
////////////////////////////////////////////////
BEGIN_PARAMETER_LIST(pl)
ADD_PARAMETER_GROUP(pl, "Input/Output")
ADD_STRING_PARAMETER(pl, inVcf, "--inVcf", "input VCF File")
ADD_STRING_PARAMETER(pl, snp, "--snp", "input dbSNP File (.rod)")
ADD_STRING_PARAMETER(pl, hapmap, "--hapmap", "input HapMap File (.bim)")
ADD_PARAMETER_GROUP(pl, "Site Filter")
ADD_STRING_PARAMETER(pl, rangeList, "--rangeList", "Specify some ranges to use, please use chr:begin-end format.")
ADD_STRING_PARAMETER(pl, rangeFile, "--rangeFile", "Specify the file containing ranges, please use chr:begin-end format.")
END_PARAMETER_LIST(pl)
;
pl.Read(argc, argv);
pl.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 defaultDbSnp[] = "/net/fantasia/home/zhanxw/amd/data/umake-resources/dbSNP/dbsnp_129_b37.rod.map";
if (FLAG_snp == "") {
FLAG_snp = defaultDbSnp;
fprintf(stderr, "Use default dbsnp: [ %s ]\n", defaultDbSnp);
}
SiteSet snpSet;
snpSet.loadRodFile(FLAG_snp);
fprintf(stderr, "%zu dbSNP sites loaded.\n", snpSet.getTotalSite());
const char defaultHM3[] = "/net/fantasia/home/zhanxw/amd/data/umake-resources/HapMap3/hapmap3_r3_b37_fwd.consensus.qc.poly.bim";
if (FLAG_hapmap == "") {
FLAG_hapmap = defaultHM3;
fprintf(stderr, "Use default HapMap: [ %s ]\n", defaultHM3);
}
SiteSet hmSet;
hmSet.loadBimFile(FLAG_hapmap);
fprintf(stderr, "%zu Hapmap sites loaded.\n", hmSet.getTotalSite());
const char* fn = FLAG_inVcf.c_str();
LineReader lr(fn);
// // set range filters here
// // e.g.
// // vin.setRangeList("1:69500-69600");
// vin.setRangeList(FLAG_rangeList.c_str());
// vin.setRangeFile(FLAG_rangeFile.c_str());
std::map<std::string, Variant> freq;
std::string chrom;
int pos;
// std::string filt;
std::string anno;
char ref, alt;
bool inDbSnp;
bool inHapmap;
int lineNo = 0;
std::vector<std::string> fd;
while(lr.readLineBySep(&fd, " \t")){
lineNo ++;
if (fd[0][0] == '#') continue; // skip header
chrom = fd[0]; // ref is on column 0 (0-based)
pos = atoi(fd[1]); // ref is on column 1 (0-based)
ref = fd[3][0]; // ref is on column 3 (0-based)
alt = fd[4][0]; // ref is on column 4 (0-based)
// filt = fd[6]; // filt is on column 6 (0-based)
anno = extractAnno(fd[7]); // info is on column 7 (0-based), we will extract ANNO=
inDbSnp = snpSet.isIncluded(chrom.c_str(), pos);
inHapmap = hmSet.isIncluded(chrom.c_str(), pos);
Variant& v = freq[anno];
v.total++;
if ( isTs(ref, alt) ) {
v.ts ++;
if (inDbSnp) {
v.tsInDbSnp ++;
v.dbSnp ++;
}
} else if (isTv(ref, alt)) {
v.tv ++;
if (inDbSnp) {
v.tvInDbSnp ++;
v.dbSnp ++;
}
};
if (inHapmap)
v.hapmap ++;
if (lineNo % 10000 == 0) {
fprintf(stderr, "\rProcessed %d lines...\r", lineNo);
}
};
fprintf(stdout, "Total %d VCF records have been read successfully\n", lineNo);
//////////////////////////////////////////////////////////////////////
std::string title = "Summarize per annotation type";
int pad = (170 - title.size() ) /2 ;
std::string outTitle = std::string(pad, '-') + title + std::string(pad, '-');
puts(outTitle.c_str());
printf("%40s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\n",
"Filter", "#SNPs", "#dbSNP", "%dbSNP", "Known Ts/Tv", "Novel Ts/Tv", "Overall", "%TotalHM3", "%HMCalled");
std::map<std::string, Variant> indvFreq;
Variant total;
for (std::map<std::string, Variant>::iterator i = freq.begin() ; i != freq.end(); ++i ){
i->second.print(i->first, hmSet);
total += i->second;
};
total.print("TOTAL", hmSet);
currentTime = time(0);
fprintf(stderr, "Analysis end at: %s", ctime(¤tTime));
return 0;
};
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 defaultDbSnp[] =
"/net/fantasia/home/zhanxw/amd/data/umake-resources/dbSNP/"
"dbsnp_129_b37.rod.map";
if (FLAG_snp == "") {
FLAG_snp = defaultDbSnp;
fprintf(stderr, "Use default dbsnp: [ %s ]\n", defaultDbSnp);
}
SiteSet snpSet;
snpSet.loadRodFile(FLAG_snp);
fprintf(stderr, "%zu dbSNP sites loaded.\n", snpSet.getTotalSite());
const char defaultHM3[] =
"/net/fantasia/home/zhanxw/amd/data/umake-resources/HapMap3/"
"hapmap3_r3_b37_fwd.consensus.qc.poly.bim";
if (FLAG_hapmap == "") {
FLAG_hapmap = defaultHM3;
fprintf(stderr, "Use default HapMap: [ %s ]\n", defaultHM3);
}
SiteSet hmSet;
hmSet.loadBimFile(FLAG_hapmap);
fprintf(stderr, "%zu Hapmap sites loaded.\n", hmSet.getTotalSite());
const char* fn = FLAG_inVcf.c_str();
LineReader lr(fn);
// // set range filters here
// // e.g.
// // vin.setRangeList("1:69500-69600");
// vin.setRangeList(FLAG_rangeList.c_str());
// vin.setRangeFile(FLAG_rangeFile.c_str());
std::map<std::string, Variant> freq;
std::string chrom;
int pos;
std::string filt;
char ref, alt;
bool inDbSnp;
bool inHapmap;
int lineNo = 0;
std::vector<std::string> fd;
while (lr.readLineBySep(&fd, " \t")) {
lineNo++;
if (fd[0][0] == '#') continue; // skip header
chrom = fd[0]; // ref is on column 0 (0-based)
pos = atoi(fd[1]); // ref is on column 1 (0-based)
ref = fd[3][0]; // ref is on column 3 (0-based)
alt = fd[4][0]; // ref is on column 4 (0-based)
filt = fd[6]; // filt is on column 6 (0-based)
inDbSnp = snpSet.isIncluded(chrom.c_str(), pos);
inHapmap = hmSet.isIncluded(chrom.c_str(), pos);
Variant& v = freq[filt];
v.total++;
if (isTs(ref, alt)) {
v.ts++;
if (inDbSnp) {
v.tsInDbSnp++;
v.dbSnp++;
}
} else if (isTv(ref, alt)) {
v.tv++;
if (inDbSnp) {
v.tvInDbSnp++;
v.dbSnp++;
}
};
if (inHapmap) v.hapmap++;
};
fprintf(stdout, "Total %d VCF records have converted successfully\n", lineNo);
//////////////////////////////////////////////////////////////////////
std::string title = "Summarize per combined filter";
int pad = (170 - title.size()) / 2;
std::string outTitle = std::string(pad, '-') + title + std::string(pad, '-');
puts(outTitle.c_str());
printf("%40s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\n", "Filter",
"#SNPs", "#dbSNP", "%dbSNP", "Known Ts/Tv", "Novel Ts/Tv", "Overall",
"%TotalHM3", "%HMCalled");
std::map<std::string, Variant> indvFreq;
Variant pass;
Variant fail;
Variant total;
std::vector<std::string> filters; // individual filter
for (std::map<std::string, Variant>::iterator i = freq.begin();
i != freq.end(); ++i) {
const std::string& filt = i->first;
const Variant& v = i->second;
v.print(filt, hmSet);
// calculate indvFreq, pass, fail and total
stringTokenize(filt, ';', &filters);
for (unsigned int j = 0; j < filters.size(); j++) {
const std::string& filt = filters[j];
indvFreq[filt] += v;
}
if (filt == "PASS")
pass += v;
else
fail += v;
total += v;
};
//////////////////////////////////////////////////////////////////////
title = "Summarize per individual filter";
pad = (170 - title.size()) / 2;
outTitle = std::string(pad, '-') + title + std::string(pad, '-');
puts(outTitle.c_str());
printf("%40s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\n", "Filter",
"#SNPs", "#dbSNP", "%dbSNP", "Known Ts/Tv", "Novel Ts/Tv", "Overall",
"%TotalHM3", "%HMCalled");
for (std::map<std::string, Variant>::iterator i = indvFreq.begin();
i != indvFreq.end(); ++i) {
const std::string& filt = i->first;
const Variant& v = i->second;
v.print(filt, hmSet);
}
//////////////////////////////////////////////////////////////////////
title = "Summarize per pass/fail filter";
pad = (170 - title.size()) / 2;
outTitle = std::string(pad, '-') + title + std::string(pad, '-');
puts(outTitle.c_str());
printf("%40s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\n", "Filter",
"#SNPs", "#dbSNP", "%dbSNP", "Known Ts/Tv", "Novel Ts/Tv", "Overall",
"%TotalHM3", "%HMCalled");
pass.print("PASS", hmSet);
fail.print("FAIL", hmSet);
total.print("TOTAL", hmSet);
currentTime = time(0);
fprintf(stderr, "Analysis end at: %s", ctime(¤tTime));
return 0;
};
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 defaultDbSnp[] =
"/net/fantasia/home/zhanxw/amd/data/umake-resources/dbSNP/"
"dbsnp_129_b37.rod.map";
if (FLAG_snp == "") {
FLAG_snp = defaultDbSnp;
fprintf(stderr, "Use default dbsnp: [ %s ]\n", defaultDbSnp);
}
SiteSet snpSet;
snpSet.loadRodFile(FLAG_snp);
fprintf(stderr, "%zu dbSNP sites loaded.\n", snpSet.getTotalSite());
const char defaultHM3[] =
"/net/fantasia/home/zhanxw/amd/data/umake-resources/HapMap3/"
"hapmap3_r3_b37_fwd.consensus.qc.poly.bim";
if (FLAG_hapmap == "") {
FLAG_hapmap = defaultHM3;
fprintf(stderr, "Use default HapMap: [ %s ]\n", defaultHM3);
}
SiteSet hmSet;
hmSet.loadBimFile(FLAG_hapmap);
fprintf(stderr, "%zu Hapmap sites loaded.\n", hmSet.getTotalSite());
const char* fn = FLAG_inVcf.c_str();
LineReader lr(fn);
// // set range filters here
// // e.g.
// // vin.setRangeList("1:69500-69600");
// vin.setRangeList(FLAG_rangeList.c_str());
// vin.setRangeFile(FLAG_rangeFile.c_str());
std::map<std::string, Variant> freq;
std::string chrom;
int pos;
// std::string filt;
// std::string anno;
std::string numVariant;
char ref, alt;
bool inDbSnp;
bool inHapmap;
int lineNo = 0;
std::vector<std::string> fd;
while (lr.readLineBySep(&fd, " \t")) {
lineNo++;
if (fd[0][0] == '#') continue; // skip header
chrom = fd[0]; // ref is on column 0 (0-based)
pos = atoi(fd[1]); // ref is on column 1 (0-based)
ref = fd[3][0]; // ref is on column 3 (0-based)
alt = fd[4][0]; // ref is on column 4 (0-based)
// filt = fd[6]; // filt is on column 6 (0-based)
// anno = extractAnno(fd[7]); // info is on column 7 (0-based), we will
// extract ANNO=
// obtain number of variants
if (fd.size() <= 9) { // first 9 columns are not individuals
numVariant = toString(0);
} else {
int numVar = 0;
for (size_t i = 9; i < fd.size(); ++i) {
int varCount = countVariant(fd[i]);
if (varCount > 0) numVar += varCount;
}
numVariant = toString(numVar);
}
inDbSnp = snpSet.isIncluded(chrom.c_str(), pos);
inHapmap = hmSet.isIncluded(chrom.c_str(), pos);
Variant& v = freq[numVariant];
v.total++;
if (isTs(ref, alt)) {
v.ts++;
if (inDbSnp) {
v.tsInDbSnp++;
v.dbSnp++;
}
} else if (isTv(ref, alt)) {
v.tv++;
if (inDbSnp) {
v.tvInDbSnp++;
v.dbSnp++;
}
};
if (inHapmap) v.hapmap++;
if (lineNo % 10000 == 0) {
fprintf(stderr, "\rProcessed %d lines...\r", lineNo);
}
};
fprintf(stdout, "Total %d VCF records have been read successfully\n", lineNo);
//////////////////////////////////////////////////////////////////////
std::string title = "Summarize per annotation type";
int pad = (170 - title.size()) / 2;
std::string outTitle = std::string(pad, '-') + title + std::string(pad, '-');
puts(outTitle.c_str());
printf("%40s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\t%10s\n", "Filter",
"#SNPs", "#dbSNP", "%dbSNP", "Known Ts/Tv", "Novel Ts/Tv", "Overall",
"%TotalHM3", "%HMCalled");
std::map<std::string, Variant> indvFreq;
Variant total;
// to sort variants by its integer order, we use a temporary map
std::map<int, Variant> tmp;
for (std::map<std::string, Variant>::iterator i = freq.begin();
i != freq.end(); ++i) {
tmp[atoi(i->first)] = i->second;
};
for (std::map<int, Variant>::iterator i = tmp.begin(); i != tmp.end(); ++i) {
i->second.print(toString(i->first), hmSet);
total += i->second;
};
total.print("TOTAL", hmSet);
currentTime = time(0);
fprintf(stderr, "Analysis end at: %s", ctime(¤tTime));
return 0;
};
