void trans_data::readCovariates(string fcov) {
string buffer;vector < string > tokens;
int n_includedS = 0;
int n_excludedS = 0;
int n_missingS = 0;
int n_includedC = 0;
int n_excludedC = 0;
vector < int > mappingS;
vrb.title("Reading covariates in [" + fcov + "]");
input_file fd (fcov);
if (fd.fail()) vrb.error("Cannot open file!");
//Read samples
getline(fd, buffer);
stb.split(buffer, tokens);
for (int t = 1 ; t < tokens.size() ; t ++) {
if (filter_sample.check(tokens[t])) {
mappingS.push_back(findSample(tokens[t]));
if (mappingS.back() >= 0) n_includedS ++;
else n_missingS ++;
} else {
mappingS.push_back(-1);
n_excludedS++;
}
}
vrb.bullet(stb.str(n_includedS) + " samples included");
if (n_excludedS > 0) vrb.bullet(stb.str(n_excludedS) + " samples excluded by user");
if (n_missingS > 0) vrb.bullet(stb.str(n_missingS) + " samples without phenotype data");
if (n_includedS != sample_count) vrb.error("Cannot find covariates for " + stb.str(sample_count - n_includedS) + " samples!");
//Read covariates
while(getline(fd, buffer)) {
if (buffer.size() == 0) continue;
stb.split(buffer, tokens);
if (tokens.size() < 2) vrb.error("Incorrect number of columns!");
if (filter_covariate.check(tokens[0])) {
covariate_val.push_back(vector < string > (sample_count));
for (int t = 1 ; t < tokens.size() ; t ++) if (mappingS[t-1] >= 0) covariate_val.back()[mappingS[t-1]] = tokens[t];
n_includedC ++;
} else n_excludedC ++;
}
//Finalise
covariate_count = n_includedC;
vrb.bullet(stb.str(n_includedC) + " covariate(s) included");
if (n_excludedC > 0) vrb.bullet(stb.str(n_excludedC) + " covariate(s) excluded");
fd.close();
}
/*
* SetValue(*)
* set the value of a BaseValueType according to its type
*/
void
TypedValue::SetTo(const string &strval)
{
if (strval.size()) {
switch (type) {
case S_BOOL: val.Bool = (atoi(strval.c_str()) != 0); break;
case S_BYTE: val.byte = atoi(strval.c_str()); break;
case S_INT: val.Int = atoi(strval.c_str()); break;
case S_SHORT: val.Short = atoi(strval.c_str()); break;
case S_LONG: val.Long = atoi(strval.c_str()); break;
case S_FLOAT: val.Float = atof(strval.c_str()); break;
case S_POOL: val.pool = findPool(strval.c_str()); break;
case S_SAMPLE: val.sample = findSample(strval.c_str()); break;
// case S_STRING: val.string = strval; break;
default:
internalError("can't set value to string, type = %d", type);
}
// fprintf(stderr, "SetVal: %s %d %d\n", strval, type, val.Int);
}
}
void gwas_data::readCovariates(string fcov) {
string buffer;
vector < string > str;
int n_includedS = 0;
int n_includedC = 0;
int n_excludedC = 0;
vector < int > mappingS;
vrb.title("Reading covariates in [" + fcov + "]");
input_file fd (fcov);
if (fd.fail()) vrb.error("Cannot open file!");
//Read samples
getline(fd, buffer);
if (buffer.size() == 0) vrb.error("No header line detected!");
stb.split(buffer, str);
for (int t = 1 ; t < str.size() ; t ++) {
mappingS.push_back(findSample(str[t]));
if (mappingS.back() >= 0) n_includedS++;
}
//Read covariates
while(getline(fd, buffer)) {
stb.split(buffer, str);
if (str.size() < 2) vrb.error("Incorrect number of columns!");
if (filter_covariate.check(str[0])) {
covariate_val.push_back(vector < string > (sample_count));
assert(mappingS.size() == (str.size() - 1));
for (int t = 1 ; t < str.size() ; t ++) if (mappingS[t-1] >= 0) covariate_val.back()[mappingS[t-1]] = str[t];
n_includedC ++;
} else n_excludedC ++;
}
//Finalise
covariate_count = n_includedC;
vrb.bullet(stb.str(n_includedC) + " covariates included");
if (n_excludedC > 0) vrb.bullet(stb.str(n_excludedC) + " covariates excluded");
fd.close();
}
int SoundPool::play(int sampleID, float leftVolume, float rightVolume,
int priority, int loop, float rate)
{
ALOGV("play sampleID=%d, leftVolume=%f, rightVolume=%f, priority=%d, loop=%d, rate=%f",
sampleID, leftVolume, rightVolume, priority, loop, rate);
sp<Sample> sample;
SoundChannel* channel;
int channelID;
Mutex::Autolock lock(&mLock);
if (mQuit) {
return 0;
}
// is sample ready?
sample = findSample(sampleID);
if ((sample == 0) || (sample->state() != Sample::READY)) {
ALOGW(" sample %d not READY", sampleID);
return 0;
}
dump();
// allocate a channel
channel = allocateChannel_l(priority);
// no channel allocated - return 0
if (!channel) {
ALOGV("No channel allocated");
return 0;
}
channelID = ++mNextChannelID;
ALOGV("play channel %p state = %d", channel, channel->state());
channel->play(sample, channelID, leftVolume, rightVolume, priority, loop, rate);
return channelID;
}
void cis_data::readPhenotypes(string fbed) {
int n_includedS = 0;
int n_includedP = 0;
int n_excludedP = 0;
int n_negativeStrd = 0;
vector < int > mappingS;
//Open BED file
vrb.title("Reading phenotype data in [" + fbed + "]");
htsFile *fp = hts_open(fbed.c_str(),"r");
if (!fp) vrb.error("Cannot open file");
tbx_t *tbx = tbx_index_load(fbed.c_str());
if (!tbx) vrb.error("Cannot open index file");
kstring_t str = {0,0,0};
if (hts_getline(fp, KS_SEP_LINE, &str) <= 0 || !str.l || str.s[0] != tbx->conf.meta_char ) vrb.error("Cannot read header line!");
//Process sample names
vector < string > tokens;
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
for (int t = 6 ; t < tokens.size() ; t ++) {
mappingS.push_back(findSample(tokens[t]));
if (mappingS.back() >= 0) n_includedS++;
}
//Read phenotypes
unsigned int linecount =0;
//Read phenotypes
if (regionPhenotype.chr != "NA"){
hts_itr_t *itr = tbx_itr_querys(tbx, regionPhenotype.get().c_str());
vrb.bullet("target region [" + regionPhenotype.get() + "]");
if (!itr) vrb.error("Cannot jump to region!");
//Read data
while (tbx_itr_next(fp, tbx, itr, &str) >= 0) {
linecount ++;
if (linecount % 100000 == 0) vrb.bullet("Read " + stb.str(linecount) + " lines");
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
if ((grp_mode == GRP_NONE && filter_phenotype.check(tokens[3])) || (grp_mode != GRP_NONE && filter_phenotype.check(tokens[4]))) {
phenotype_id.push_back(tokens[3]);
phenotype_chr.push_back(tokens[0]);
phenotype_start.push_back(atoi(tokens[1].c_str()) + 1);
phenotype_end.push_back(atoi(tokens[2].c_str()));
if (grp_mode > 0 && full_test) phenotype_grp.push_back("ALL_GENES");
if (grp_mode > 0 && !full_test) phenotype_grp.push_back(tokens[4]);
phenotype_neg.push_back(tokens[5] == "-");
if (phenotype_neg.back()) n_negativeStrd ++;
phenotype_val.push_back(vector < float > (sample_count, 0.0));
for (int t = 6 ; t < tokens.size() ; t ++) {
if (mappingS[t-6] >= 0) {
if (tokens[t] == "NA") phenotype_val.back()[mappingS[t-6]] = bcf_float_missing;
else phenotype_val.back()[mappingS[t-6]] = stof(tokens[t]);
}
}
n_includedP++;
} else n_excludedP ++;
}
tbx_itr_destroy(itr);
}else{
while (hts_getline(fp, KS_SEP_LINE, &str) >= 0) {
linecount ++;
if (linecount % 100000 == 0) vrb.bullet("Read " + stb.str(linecount) + " lines");
stb.split(string(str.s), tokens);
if (str.l && str.s[0] != tbx->conf.meta_char) {
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
if ((grp_mode == GRP_NONE && filter_phenotype.check(tokens[3])) || (grp_mode != GRP_NONE && filter_phenotype.check(tokens[4]))) {
phenotype_id.push_back(tokens[3]);
phenotype_chr.push_back(tokens[0]);
phenotype_start.push_back(atoi(tokens[1].c_str()) + 1);
phenotype_end.push_back(atoi(tokens[2].c_str()));
if (grp_mode > 0 && full_test) phenotype_grp.push_back("ALL_GENES");
if (grp_mode > 0 && !full_test) phenotype_grp.push_back(tokens[4]);
phenotype_neg.push_back(tokens[5] == "-");
if (phenotype_neg.back()) n_negativeStrd ++;
phenotype_val.push_back(vector < float > (sample_count, 0.0));
for (int t = 6 ; t < tokens.size() ; t ++) {
if (mappingS[t-6] >= 0) {
if (tokens[t] == "NA") phenotype_val.back()[mappingS[t-6]] = bcf_float_missing;
else phenotype_val.back()[mappingS[t-6]] = stof(tokens[t]);
}
}
n_includedP++;
} else n_excludedP ++;
}
}
}
//Finalize & verbose
tbx_destroy(tbx);
if (hts_close(fp)) vrb.error("Cannot properly close file");
phenotype_count = phenotype_id.size();
vrb.bullet(stb.str(n_includedP) + " phenotypes included");
if (n_excludedP > 0) vrb.bullet(stb.str(n_excludedP) + " phenotypes excluded by user");
if (n_negativeStrd > 0 ) vrb.bullet(stb.str(n_negativeStrd) + " phenotypes are on the negative strand");
if (phenotype_count == 0) vrb.leave("Cannot find phenotypes in target region!");
}
void union_data::readGenotypesVCF(string fvcf,string region) {
int n_includedG = 0;
int n_excludedG_mult = 0;
int n_excludedG_void = 0;
int n_excludedG_user = 0;
int n_includedS = 0;
vector < int > mappingS;
genotype_id.clear();
genotype_chr.clear();
genotype_start.clear();
genotype_end.clear();
genotype_val.clear();
genotype_count=0;
genotype_id_to_idx.clear();
//Opening files
bcf_srs_t * sr = bcf_sr_init();
//vrb.bullet("target region [" + regionGenotype.get() + "]");
//if (bcf_sr_set_regions(sr, regionGenotype.get().c_str(), 0) == -1) vrb.error("Cannot jump to region!");
bcf_sr_set_regions(sr, region.c_str(), 0);
if(!(bcf_sr_add_reader (sr, fvcf.c_str()))) {
switch (sr->errnum) {
case not_bgzf: vrb.error("File not compressed with bgzip!");
case idx_load_failed: vrb.error("Impossible to load index file!");
case file_type_error: vrb.error("File format not detected by htslib!");
default : vrb.error("Unknown error!");
}
}
//Sample processing
int n_samples = bcf_hdr_nsamples(sr->readers[0].header);
for (int i0 = 0 ; i0 < n_samples ; i0 ++) {
mappingS.push_back(findSample(string(sr->readers[0].header->samples[i0])));
if (mappingS.back() >= 0) n_includedS++;
}
//Read genotype data
int ngt, ngt_arr = 0, nds, nds_arr = 0, * gt_arr = NULL, nsl, nsl_arr = 0, * sl_arr = NULL;
float * ds_arr = NULL;
bcf1_t * line;
unsigned int linecount = 0;
while(bcf_sr_next_line (sr)) {
linecount ++;
if (linecount % 100000 == 0) vrb.bullet("Read " + stb.str(linecount) + " lines");
line = bcf_sr_get_line(sr, 0);
if (line->n_allele == 2) {
ngt = bcf_get_genotypes(sr->readers[0].header, line, >_arr, &ngt_arr);
nds = bcf_get_format_float(sr->readers[0].header, line,"DS", &ds_arr, &nds_arr);
if (nds == n_samples || ngt == 2*n_samples) {
bcf_unpack(line, BCF_UN_STR);
string sid = string(line->d.id);
if (filter_genotype.check(sid)) {
genotype_id.push_back(sid);
genotype_chr.push_back(string(bcf_hdr_id2name(sr->readers[0].header, line->rid)));
string genotype_ref = string(line->d.allele[0]);
genotype_start.push_back(line->pos + 1);
nsl = bcf_get_info_int32(sr->readers[0].header, line, "END", &sl_arr, &nsl_arr);
if (nsl >= 0 && nsl_arr == 1) genotype_end.push_back(sl_arr[0]);
else genotype_end.push_back(genotype_start.back() + genotype_ref.size() - 1);
genotype_val.push_back(vector < float > (sample_count, 0.0));
for(int i = 0 ; i < n_samples ; i ++) {
if (mappingS[i] >= 0) {
if (nds > 0) genotype_val.back()[mappingS[i]] = ds_arr[i];
else {
if (gt_arr[2*i+0] == bcf_gt_missing || gt_arr[2*i+1] == bcf_gt_missing) genotype_val.back()[mappingS[i]] = bcf_float_missing;
else genotype_val.back()[mappingS[i]] = bcf_gt_allele(gt_arr[2*i+0]) + bcf_gt_allele(gt_arr[2*i+1]);
}
}
}
pair < string, int > temp (sid,n_includedG);
genotype_id_to_idx.insert(temp);
n_includedG++;
} else n_excludedG_user ++;
} else n_excludedG_void ++;
} else n_excludedG_mult ++;
}
//Finalize
free(gt_arr);
free(ds_arr);
bcf_sr_destroy(sr);
genotype_count = n_includedG;
//vrb.bullet(stb.str(n_includedG) + " variants included");
//if (n_excludedG_user > 0) vrb.bullet(stb.str(n_excludedG_user) + " variants excluded by user");
//if (n_excludedG_mult > 0) vrb.bullet(stb.str(n_excludedG_mult) + " multi-allelic variants excluded");
//if (n_excludedG_void > 0) vrb.bullet(stb.str(n_excludedG_void) + " uninformative variants excluded [no GT/DS]");
//if (genotype_count == 0) vrb.leave("Cannot find genotypes in target region!");
}
void union_data::readGenotypesBED(string fbed,string region) {
string buffer;
int n_includedG = 0;
int n_excludedG_user = 0;
int n_includedS = 0;
int n_excludedS = 0;
int n_missingS = 0;
vector < int > mappingS;
genotype_id.clear();
genotype_chr.clear();
genotype_start.clear();
genotype_end.clear();
genotype_val.clear();
genotype_count=0;
genotype_id_to_idx.clear();
//Opening files
htsFile *fp = hts_open(fbed.c_str(),"r");
if (!fp) vrb.error("Cannot open file!");
tbx_t * tbx = tbx_index_load(fbed.c_str());
if (!tbx) vrb.error("Cannot load index file!");
kstring_t str = {0,0,0};
if (hts_getline(fp, KS_SEP_LINE, &str) <= 0 || !str.l || str.s[0] != tbx->conf.meta_char ) vrb.error("Cannot read header line!");
//Process sample names
vector < string > tokens;
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
for (int i0 = 6 ; i0 < tokens.size() ; i0 ++) {
string sid = tokens[i0];
if (filter_sample.check(sid)) {
mappingS.push_back(findSample(sid));
if (mappingS.back() >= 0) n_includedS ++;
else n_missingS ++;
} else {
mappingS.push_back(-1);
n_excludedS ++;
}
}
//vrb.bullet(stb.str(n_includedS) + " samples included");
//if (n_excludedS > 0) vrb.bullet(stb.str(n_excludedS) + " samples excluded by user");
//if (n_missingS > 0) vrb.bullet(stb.str(n_missingS) + " samples without phenotype data");
//if (n_includedS != sample_count) vrb.error("Cannot find genotype for " + stb.str(sample_count - n_includedS) + " samples!");
unsigned int linecount = 0;
//Jump to interesting region
hts_itr_t *itr = tbx_itr_querys(tbx, region.c_str());
//vrb.bullet("target region [" + regionGenotype.get() + "]");
//if (!itr) vrb.error("Cannot jump to region!");
while (tbx_itr_next(fp, tbx, itr, &str) >= 0) {
linecount ++;
if (linecount % 100000 == 0) vrb.bullet("Read " + stb.str(linecount) + " lines");
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
if (filter_genotype.check(tokens[3])) {
genotype_id.push_back(tokens[3]);
genotype_chr.push_back(tokens[0]);
genotype_start.push_back(atoi(tokens[1].c_str()) + 1);
genotype_end.push_back(atoi(tokens[2].c_str()));
genotype_val.push_back(vector < float > (sample_count, 0.0));
for (int t = 6 ; t < tokens.size() ; t ++) {
if (mappingS[t-6] >= 0) {
if (tokens[t] == "NA") genotype_val.back()[mappingS[t-6]] = bcf_float_missing;
else genotype_val.back()[mappingS[t-6]] = stof(tokens[t]);
}
}
pair < string, int > temp (tokens[3],n_includedG);
genotype_id_to_idx.insert(temp);
n_includedG++;
} else n_excludedG_user ++;
}
tbx_itr_destroy(itr);
//Finalize & verbose
tbx_destroy(tbx);
if (hts_close(fp)) vrb.error("Cannot properly close file!");
genotype_count = n_includedG;
//vrb.bullet(stb.str(n_includedG) + " variants included");
//if (n_excludedG_user > 0) vrb.bullet(stb.str(n_excludedG_user) + " variants excluded by user");
//if (genotype_count == 0) vrb.leave("Cannot find variants in target region!");
}
void union_data::readPhenotypes(string fbed, string region) {
int n_includedS = 0;
int n_includedP = 0;
int n_excludedP = 0;
vector < int > mappingS;
phenotype_id.clear();
phenotype_chr.clear();
phenotype_start.clear();
phenotype_end.clear();
phenotype_val.clear();
phenotype_count=0;
phenotype_id_to_idx.clear();
//Open BED file
//vrb.title("Reading phenotype data in [" + fbed + "]");
htsFile *fp = hts_open(fbed.c_str(),"r");
if (!fp) vrb.error("Cannot open file");
tbx_t *tbx = tbx_index_load(fbed.c_str());
if (!tbx) vrb.error("Cannot open index file");
kstring_t str = {0,0,0};
if (hts_getline(fp, KS_SEP_LINE, &str) <= 0 || !str.l || str.s[0] != tbx->conf.meta_char ) vrb.error("Cannot read header line!");
//Process sample names
vector < string > tokens;
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
for (int t = 6 ; t < tokens.size() ; t ++) {
mappingS.push_back(findSample(tokens[t]));
if (mappingS.back() >= 0) n_includedS++;
}
unsigned int linecount =0;
//Read phenotypes
hts_itr_t *itr = tbx_itr_querys(tbx, region.c_str());
//vrb.bullet("target region [" + regionPhenotype.get() + "]");
//if (!itr) vrb.error("Cannot jump to region!");
//Read data
while (tbx_itr_next(fp, tbx, itr, &str) >= 0) {
linecount ++;
if (linecount % 100000 == 0) vrb.bullet("Read " + stb.str(linecount) + " lines");
stb.split(string(str.s), tokens);
if (tokens.size() < 7) vrb.error("Incorrect number of columns!");
if (filter_phenotype.check(tokens[3])) {
phenotype_id.push_back(tokens[3]);
phenotype_chr.push_back(tokens[0]);
phenotype_start.push_back(atoi(tokens[1].c_str()) + 1);
phenotype_end.push_back(atoi(tokens[2].c_str()));
phenotype_val.push_back(vector < float > (sample_count, 0.0));
for (int t = 6 ; t < tokens.size() ; t ++) {
if (mappingS[t-6] >= 0) {
if (tokens[t] == "NA") phenotype_val.back()[mappingS[t-6]] = bcf_float_missing;
else phenotype_val.back()[mappingS[t-6]] = stof(tokens[t]);
}
}
pair < string, int > temp (tokens[3],n_includedP);
phenotype_id_to_idx.insert(temp);
n_includedP++;
} else n_excludedP ++;
}
tbx_itr_destroy(itr);
//Finalize & verbose
tbx_destroy(tbx);
if (hts_close(fp)) vrb.error("Cannot properly close file");
phenotype_count = phenotype_id.size();
//vrb.bullet(stb.str(n_includedP) + " phenotypes included");
//if (n_excludedP > 0) vrb.bullet(stb.str(n_excludedP) + " phenotypes excluded by user");
//if (phenotype_count == 0) vrb.leave("Cannot find phenotypes in target region!");
}
void MenuScreen::handleEvent(SDL_Event &e)
{
if (e.type != SDL_KEYDOWN)
return;
if (pwdMode)
{
if (e.key.keysym.sym >= SDLK_a && e.key.keysym.sym <= SDLK_z)
{
if (pwdLen == 6)
{
playSample(findSample(makeCode("slct")));
return;
}
playSample(findSample(makeCode("tick")));
pwd[pwdLen++] = 'a'+(e.key.keysym.sym-SDLK_a);
return;
}
if (e.key.keysym.sym == SDLK_BACKSPACE)
{
if (!pwdLen)
{
playSample(findSample(makeCode("slct")));
return;
}
playSample(findSample(makeCode("tick")));
pwdLen--;
return;
}
if (e.key.keysym.sym == SDLK_RETURN)
{
playSample(findSample(makeCode("tick")));
if (pwdLen == 6)
{
pwd[6] = 0;
game->newGame(pwd);
e.key.keysym.sym = SDLK_a;
game->activate();
}
pwdMode = false;
return;
}
if (e.key.keysym.sym == SDLK_ESCAPE)
pwdMode = false;
return;
}
switch (e.key.keysym.sym)
{
case SDLK_DOWN:
if (opt == 3)
opt = 0;
else
opt++;
playSample(findSample(makeCode("slct")));
break;
case SDLK_UP:
if (opt)
opt--;
else
opt = 3;
playSample(findSample(makeCode("slct")));
break;
case SDLK_SPACE:
case SDLK_RETURN:
{
playSample(findSample(makeCode("tick")));
switch (opt)
{
case 0:
game->newGame();
game->activate();
break;
case 1:
pwdLen = 0;
pwdMode = true;
break;
case 2:
imageScreen->showImage(this,
"data/objects/credits.ut");
break;
case 3:
running = false;
break;
}
break;
}
case SDLK_ESCAPE:
if (game->playing)
game->activate();
break;
default:;
}
}
void genrich_data::readReferenceGenotypes(string fvcf) {
vector < int > mappingS;
//Opening files
vrb.title("Reading variant list in [" + fvcf + "] MAF=" + stb.str(threshold_maf));
bcf_srs_t * sr = bcf_sr_init();
if(!(bcf_sr_add_reader (sr, fvcf.c_str()))) {
switch (sr->errnum) {
case not_bgzf: vrb.error("File not compressed with bgzip!");
case idx_load_failed: vrb.error("Impossible to load index file!");
case file_type_error: vrb.error("File format not detected by htslib!");
default : vrb.error("Unknown error!");
}
}
//Sample processing
int included_sample = 0;
int n_samples = bcf_hdr_nsamples(sr->readers[0].header);
for (int i = 0 ; i < n_samples ; i ++) {
mappingS.push_back(findSample(string(sr->readers[0].header->samples[i])));
if (mappingS.back() >= 0) included_sample ++;
}
vrb.bullet("#samples = " + stb.str(included_sample));
//Variant processing
unsigned int n_excludedV_mult = 0, n_excludedV_void = 0, n_excludedV_rare = 0, n_excludedV_uchr = 0, n_line = 0, n_excludedV_toofar = 0;
int ngt, ngt_arr = 0, *gt_arr = NULL;
bcf1_t * line;
while(bcf_sr_next_line (sr)) {
line = bcf_sr_get_line(sr, 0);
if (line->n_allele == 2) {
bcf_unpack(line, BCF_UN_STR);
string sid = string(line->d.id);
string chr = string(bcf_hdr_id2name(sr->readers[0].header, line->rid));
int chr_idx = findCHR(chr);
if (chr_idx >= 0) {
unsigned int pos = line->pos + 1;
ngt = bcf_get_genotypes(sr->readers[0].header, line, >_arr, &ngt_arr);
if (ngt == 2*n_samples) {
double freq = 0.0, tot = 0.0;
for(int i = 0 ; i < n_samples ; i ++) {
assert(gt_arr[2*i+0] != bcf_gt_missing && gt_arr[2*i+1] != bcf_gt_missing);
if (mappingS[i] >= 0) {
freq += bcf_gt_allele(gt_arr[2*i+0]) + bcf_gt_allele(gt_arr[2*i+1]);
tot += 2.0;
}
}
double maf = freq / tot;
if (maf > 0.5) maf = 1.0 - maf;
if (maf >= threshold_maf) {
int dist_tss = getDistance(chr_idx, pos);
if (dist_tss < 1e6) {
string tmp_id = chr + "_" + stb.str(pos);
genotype_uuid.insert(pair < string, unsigned int > (tmp_id, genotype_pos.size()));
genotype_chr.push_back(chr_idx);
genotype_pos.push_back(pos);
genotype_maf.push_back(maf);
genotype_dist.push_back(dist_tss);
genotype_haps.push_back(vector < bool > (2 * included_sample, false));
for(int i = 0 ; i < n_samples ; i ++) {
if (mappingS[i] >= 0) {
genotype_haps.back()[2 * mappingS[i] + 0] = bcf_gt_allele(gt_arr[2 * i + 0]);
genotype_haps.back()[2 * mappingS[i] + 1] = bcf_gt_allele(gt_arr[2 * i + 1]);
}
}
} else n_excludedV_toofar++;
} else n_excludedV_rare ++;
} else n_excludedV_void ++;
} else n_excludedV_uchr ++;
} else n_excludedV_mult ++;
if (n_line % 100000 == 0) vrb.bullet("#lines = " + stb.str(n_line));
n_line ++;
}
genotype_qtl = vector < bool > (genotype_pos.size(), false);
genotype_gwas = vector < bool > (genotype_pos.size(), false);
genotype_bin = vector < int > (genotype_pos.size(), -1);
//Finalize
bcf_sr_destroy(sr);
vrb.bullet(stb.str(genotype_pos.size()) + " variants included");
if (n_excludedV_mult > 0) vrb.bullet(stb.str(n_excludedV_mult) + " multi-allelic variants excluded");
if (n_excludedV_uchr > 0) vrb.bullet(stb.str(n_excludedV_uchr) + " variants with unreferenced chromosome in --tss");
if (n_excludedV_rare > 0) vrb.bullet(stb.str(n_excludedV_rare) + " maf filtered variants");
if (n_excludedV_toofar > 0) vrb.bullet(stb.str(n_excludedV_toofar) + " too far variants");
}
