int main_vcfgtcheck(int argc, char *argv[])
{
int c;
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->files = bcf_sr_init();
args->argc = argc;
args->argv = argv;
set_cwd(args);
char *regions = NULL, *targets = NULL;
int regions_is_file = 0, targets_is_file = 0;
static struct option loptions[] =
{
{"GTs-only",1,0,'G'},
{"all-sites",0,0,'a'},
{"homs-only",0,0,'H'},
{"help",0,0,'h'},
{"genotypes",1,0,'g'},
{"plot",1,0,'p'},
{"target-sample",1,0,'S'},
{"query-sample",1,0,'s'},
{"regions",1,0,'r'},
{"regions-file",1,0,'R'},
{"targets",1,0,'t'},
{"targets-file",1,0,'T'},
{0,0,0,0}
};
while ((c = getopt_long(argc, argv, "hg:p:s:S:Hr:R:at:T:G:",loptions,NULL)) >= 0) {
switch (c) {
case 'G':
args->no_PLs = atoi(optarg);
break;
case 'a':
args->all_sites = 1;
break;
case 'H':
args->hom_only = 1;
break;
case 'g':
args->gt_fname = optarg;
break;
case 'p':
args->plot = optarg;
break;
case 'S':
args->target_sample = optarg;
break;
case 's':
args->query_sample = optarg;
break;
case 'r':
regions = optarg;
break;
case 'R':
regions = optarg;
regions_is_file = 1;
break;
case 't':
targets = optarg;
break;
case 'T':
targets = optarg;
targets_is_file = 1;
break;
case 'h':
case '?':
usage();
default:
error("Unknown argument: %s\n", optarg);
}
}
if ( argc==optind || argc>optind+1 ) usage(); // none or too many files given
if ( !args->gt_fname ) args->cross_check = 1; // no genotype file, run in cross-check mode
else args->files->require_index = 1;
if ( regions && bcf_sr_set_regions(args->files, regions, regions_is_file)<0 ) error("Failed to read the regions: %s\n", regions);
if ( targets && bcf_sr_set_targets(args->files, targets, targets_is_file, 0)<0 ) error("Failed to read the targets: %s\n", targets);
if ( !bcf_sr_add_reader(args->files, argv[optind]) ) error("Failed to open or the file not indexed: %s\n", argv[optind]);
if ( args->gt_fname && !bcf_sr_add_reader(args->files, args->gt_fname) ) error("Failed to open or the file not indexed: %s\n", args->gt_fname);
args->files->collapse = COLLAPSE_SNPS|COLLAPSE_INDELS;
if ( args->plot ) args->plot = init_prefix(args->plot);
init_data(args);
if ( args->cross_check )
cross_check_gts(args);
else
check_gt(args);
destroy_data(args);
bcf_sr_destroy(args->files);
if (args->plot) free(args->plot);
free(args);
return 0;
}
int main_vcfroh(int argc, char *argv[])
{
int c;
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->argc = argc; args->argv = argv;
args->files = bcf_sr_init();
args->t2AZ = 6.7e-8;
args->t2HW = 5e-9;
args->rec_rate = 0;
int regions_is_file = 0, targets_is_file = 0;
static struct option loptions[] =
{
{"AF-tag",1,0,0},
{"AF-file",1,0,1},
{"AF-dflt",1,0,2},
{"estimate-AF",1,0,'e'},
{"GTs-only",1,0,'G'},
{"sample",1,0,'s'},
{"hw-to-az",1,0,'a'},
{"az-to-hw",1,0,'H'},
{"viterbi-training",0,0,'V'},
{"targets",1,0,'t'},
{"targets-file",1,0,'T'},
{"regions",1,0,'r'},
{"regions-file",1,0,'R'},
{"genetic-map",1,0,'m'},
{"rec-rate",1,0,'M'},
{"skip-indels",0,0,'I'},
{0,0,0,0}
};
int naf_opts = 0;
char *tmp;
while ((c = getopt_long(argc, argv, "h?r:R:t:T:H:a:s:m:M:G:Ia:e:V",loptions,NULL)) >= 0) {
switch (c) {
case 0: args->af_tag = optarg; naf_opts++; break;
case 1: args->af_fname = optarg; naf_opts++; break;
case 2:
args->dflt_AF = strtod(optarg,&tmp);
if ( *tmp ) error("Could not parse: --AF-dflt %s\n", optarg);
break;
case 'e': args->estimate_AF = optarg; naf_opts++; break;
case 'I': args->snps_only = 1; break;
case 'G':
args->fake_PLs = 1;
args->unseen_PL = strtod(optarg,&tmp);
if ( *tmp ) error("Could not parse: -G %s\n", optarg);
args->unseen_PL = pow(10,-args->unseen_PL/10.);
break;
case 'm': args->genmap_fname = optarg; break;
case 'M':
args->rec_rate = strtod(optarg,&tmp);
if ( *tmp ) error("Could not parse: -M %s\n", optarg);
break;
case 's': args->sample = strdup(optarg); break;
case 'a':
args->t2AZ = strtod(optarg,&tmp);
if ( *tmp ) error("Could not parse: -a %s\n", optarg);
break;
case 'H':
args->t2HW = strtod(optarg,&tmp);
if ( *tmp ) error("Could not parse: -H %s\n", optarg);
break;
case 't': args->targets_list = optarg; break;
case 'T': args->targets_list = optarg; targets_is_file = 1; break;
case 'r': args->regions_list = optarg; break;
case 'R': args->regions_list = optarg; regions_is_file = 1; break;
case 'V': args->vi_training = 1; break;
case 'h':
case '?': usage(args); break;
default: error("Unknown argument: %s\n", optarg);
}
}
if ( argc<optind+1 ) usage(args);
if ( args->t2AZ<0 || args->t2AZ>1 ) error("Error: The parameter --hw-to-az is not in [0,1]\n", args->t2AZ);
if ( args->t2HW<0 || args->t2HW>1 ) error("Error: The parameter --az-to-hw is not in [0,1]\n", args->t2HW);
if ( naf_opts>1 ) error("Error: The options --AF-tag, --AF-file and -e are mutually exclusive\n");
if ( args->af_fname && args->targets_list ) error("Error: The options --AF-file and -t are mutually exclusive\n");
if ( args->regions_list )
{
if ( bcf_sr_set_regions(args->files, args->regions_list, regions_is_file)<0 )
error("Failed to read the regions: %s\n", args->regions_list);
}
if ( args->targets_list )
{
if ( bcf_sr_set_targets(args->files, args->targets_list, targets_is_file, 0)<0 )
error("Failed to read the targets: %s\n", args->targets_list);
}
if ( args->af_fname )
{
if ( bcf_sr_set_targets(args->files, args->af_fname, 1, 3)<0 )
error("Failed to read the targets: %s\n", args->af_fname);
}
if ( !bcf_sr_add_reader(args->files, argv[optind]) ) error("Failed to open %s: %s\n", argv[optind],bcf_sr_strerror(args->files->errnum));
init_data(args);
while ( bcf_sr_next_line(args->files) )
{
vcfroh(args, args->files->readers[0].buffer[0]);
}
vcfroh(args, NULL);
fprintf(pysamerr,"Number of lines: total/processed: %d/%d\n", args->ntot,args->nused);
destroy_data(args);
free(args);
return 0;
}
static void init_data(args_t *args)
{
bcf1_t *line = NULL;
// With phased concat, the chunks overlap and come in the right order. To
// avoid opening all files at once, store start positions to recognise need
// for the next one. This way we can keep only two open chunks at once.
if ( args->phased_concat )
{
args->start_pos = (int*) malloc(sizeof(int)*args->nfnames);
line = bcf_init();
}
kstring_t str = {0,0,0};
int i, prev_chrid = -1;
for (i=0; i<args->nfnames; i++)
{
htsFile *fp = hts_open(args->fnames[i], "r"); if ( !fp ) error("Failed to open: %s\n", args->fnames[i]);
bcf_hdr_t *hdr = bcf_hdr_read(fp); if ( !hdr ) error("Failed to parse header: %s\n", args->fnames[i]);
if ( !args->out_hdr )
args->out_hdr = bcf_hdr_dup(hdr);
else
{
bcf_hdr_combine(args->out_hdr, hdr);
if ( bcf_hdr_nsamples(hdr) != bcf_hdr_nsamples(args->out_hdr) )
error("Different number of samples in %s. Perhaps \"bcftools merge\" is what you are looking for?\n", args->fnames[i]);
int j;
for (j=0; j<bcf_hdr_nsamples(hdr); j++)
if ( strcmp(args->out_hdr->samples[j],hdr->samples[j]) )
error("Different sample names in %s. Perhaps \"bcftools merge\" is what you are looking for?\n", args->fnames[i]);
}
if ( args->phased_concat )
{
int ret = bcf_read(fp, hdr, line);
if ( ret!=0 ) args->start_pos[i] = -2; // empty file
else
{
int chrid = bcf_hdr_id2int(args->out_hdr,BCF_DT_CTG,bcf_seqname(hdr,line));
args->start_pos[i] = chrid==prev_chrid ? line->pos : -1;
prev_chrid = chrid;
}
}
bcf_hdr_destroy(hdr);
hts_close(fp);
}
free(str.s);
if ( line ) bcf_destroy(line);
args->seen_seq = (int*) calloc(args->out_hdr->n[BCF_DT_CTG],sizeof(int));
if ( args->phased_concat )
{
bcf_hdr_append(args->out_hdr,"##FORMAT=<ID=PQ,Number=1,Type=Integer,Description=\"Phasing Quality (bigger is better)\">");
bcf_hdr_append(args->out_hdr,"##FORMAT=<ID=PS,Number=1,Type=Integer,Description=\"Phase Set\">");
}
bcf_hdr_append_version(args->out_hdr, args->argc, args->argv, "bcftools_concat");
args->out_fh = hts_open(args->output_fname,hts_bcf_wmode(args->output_type));
if ( args->out_fh == NULL ) error("Can't write to \"%s\": %s\n", args->output_fname, strerror(errno));
bcf_hdr_write(args->out_fh, args->out_hdr);
if ( args->allow_overlaps )
{
args->files = bcf_sr_init();
args->files->require_index = 1;
if ( args->regions_list )
{
if ( bcf_sr_set_regions(args->files, args->regions_list, args->regions_is_file)<0 )
error("Failed to read the regions: %s\n", args->regions_list);
}
for (i=0; i<args->nfnames; i++)
if ( !bcf_sr_add_reader(args->files,args->fnames[i]) ) error("Failed to open %s: %s\n", args->fnames[i],bcf_sr_strerror(args->files->errnum));
}
else if ( args->phased_concat )
{
// Remove empty files from the list
int nok = 0;
while (1)
{
while ( nok<args->nfnames && args->start_pos[nok]!=-2 ) nok++;
if ( nok==args->nfnames ) break;
i = nok;
while ( i<args->nfnames && args->start_pos[i]==-2 ) i++;
if ( i==args->nfnames ) break;
int tmp = args->start_pos[nok]; args->start_pos[nok] = args->start_pos[i]; args->start_pos[i] = tmp;
char *str = args->fnames[nok]; args->fnames[nok] = args->fnames[i]; args->fnames[i] = str;
}
for (i=nok; i<args->nfnames; i++) free(args->fnames[i]);
args->nfnames = nok;
for (i=1; i<args->nfnames; i++)
if ( args->start_pos[i-1]!=-1 && args->start_pos[i]!=-1 && args->start_pos[i]<args->start_pos[i-1] )
error("The files not in ascending order: %d in %s, %d in %s\n", args->start_pos[i-1]+1,args->fnames[i-1],args->start_pos[i]+1,args->fnames[i]);
args->prev_chr = -1;
args->swap_phase = (int*) calloc(bcf_hdr_nsamples(args->out_hdr),sizeof(int));
args->nmatch = (int*) calloc(bcf_hdr_nsamples(args->out_hdr),sizeof(int));
args->nmism = (int*) calloc(bcf_hdr_nsamples(args->out_hdr),sizeof(int));
args->phase_qual = (int32_t*) malloc(bcf_hdr_nsamples(args->out_hdr)*sizeof(int32_t));
args->phase_set = (int32_t*) malloc(bcf_hdr_nsamples(args->out_hdr)*sizeof(int32_t));
args->files = bcf_sr_init();
args->files->require_index = 1;
args->ifname = 0;
}
}
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");
}
int main_vcfisec(int argc, char *argv[])
{
int c;
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->files = bcf_sr_init();
args->argc = argc; args->argv = argv;
args->output_type = FT_VCF;
static struct option loptions[] =
{
{"help",0,0,'h'},
{"collapse",1,0,'c'},
{"complement",1,0,'C'},
{"apply-filters",1,0,'f'},
{"nfiles",1,0,'n'},
{"prefix",1,0,'p'},
{"write",1,0,'w'},
{"targets",1,0,'t'},
{"regions",1,0,'r'},
{"output-type",1,0,'O'},
{0,0,0,0}
};
while ((c = getopt_long(argc, argv, "hc:r:p:n:w:t:Cf:O:",loptions,NULL)) >= 0) {
switch (c) {
case 'O':
switch (optarg[0]) {
case 'b': args->output_type = FT_BCF_GZ; break;
case 'u': args->output_type = FT_BCF; break;
case 'z': args->output_type = FT_VCF_GZ; break;
case 'v': args->output_type = FT_VCF; break;
default: error("The output type \"%s\" not recognised\n", optarg);
}
break;
case 'c':
if ( !strcmp(optarg,"snps") ) args->files->collapse |= COLLAPSE_SNPS;
else if ( !strcmp(optarg,"indels") ) args->files->collapse |= COLLAPSE_INDELS;
else if ( !strcmp(optarg,"both") ) args->files->collapse |= COLLAPSE_SNPS | COLLAPSE_INDELS;
else if ( !strcmp(optarg,"any") ) args->files->collapse |= COLLAPSE_ANY;
else if ( !strcmp(optarg,"all") ) args->files->collapse |= COLLAPSE_ANY;
else if ( !strcmp(optarg,"some") ) args->files->collapse |= COLLAPSE_SOME;
else if ( !strcmp(optarg,"none") ) args->files->collapse = COLLAPSE_NONE;
else error("The --collapse string \"%s\" not recognised.\n", optarg);
break;
case 'f': args->files->apply_filters = optarg; break;
case 'C': args->isec_op = OP_COMPLEMENT; break;
case 'r': args->regions_fname = optarg; break;
case 't': args->targets_fname = optarg; break;
case 'p': args->prefix = optarg; break;
case 'w': args->write_files = optarg; break;
case 'n':
{
char *p = optarg;
if ( *p=='-' ) { args->isec_op = OP_MINUS; p++; }
else if ( *p=='+' ) { args->isec_op = OP_PLUS; p++; }
else if ( *p=='=' ) { args->isec_op = OP_EQUAL; p++; }
else if ( isdigit(*p) ) args->isec_op = OP_EQUAL;
else error("Could not parse --nfiles %s\n", optarg);
if ( sscanf(p,"%d",&args->isec_n)!=1 ) error("Could not parse --nfiles %s\n", optarg);
}
break;
case 'h':
case '?': usage();
default: error("Unknown argument: %s\n", optarg);
}
}
if ( argc-optind<1 ) usage(); // no file given
if ( args->targets_fname && bcf_sr_set_targets(args->files, args->targets_fname,0)<0 )
error("Failed to read the targets: %s\n", args->targets_fname);
if ( args->regions_fname && bcf_sr_set_regions(args->files, args->regions_fname)<0 )
error("Failed to read the regions: %s\n", args->regions_fname);
if ( argc-optind==2 && !args->isec_op )
{
args->isec_op = OP_VENN;
if ( !args->prefix ) error("Expected the -p option\n");
}
if ( !args->targets_fname )
{
if ( argc-optind<2 ) error("Expected multiple files or the --targets option\n");
if ( !args->isec_op ) error("Expected two file names or one of the options --complement, --nfiles or --targets\n");
}
args->files->require_index = 1;
while (optind<argc)
{
if ( !bcf_sr_add_reader(args->files, argv[optind]) ) error("Failed to open: %s\n", argv[optind]);
optind++;
}
init_data(args);
isec_vcf(args);
destroy_data(args);
bcf_sr_destroy(args->files);
free(args);
return 0;
}
int main_vcfquery(int argc, char *argv[])
{
int c, collapse = 0;
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->argc = argc; args->argv = argv;
int regions_is_file = 0, targets_is_file = 0;
static struct option loptions[] =
{
{"help",0,0,'h'},
{"list-samples",0,0,'l'},
{"include",1,0,'i'},
{"exclude",1,0,'e'},
{"format",1,0,'f'},
{"output-file",1,0,'o'},
{"regions",1,0,'r'},
{"regions-file",1,0,'R'},
{"targets",1,0,'t'},
{"targets-file",1,0,'T'},
{"annots",1,0,'a'},
{"samples",1,0,'s'},
{"samples-file",1,0,'S'},
{"print-header",0,0,'H'},
{"collapse",1,0,'c'},
{"vcf-list",1,0,'v'},
{"allow-undef-tags",0,0,'u'},
{0,0,0,0}
};
while ((c = getopt_long(argc, argv, "hlr:R:f:a:s:S:Ht:T:c:v:i:e:o:u",loptions,NULL)) >= 0) {
switch (c) {
case 'o': args->fn_out = optarg; break;
case 'f': args->format_str = strdup(optarg); break;
case 'H': args->print_header = 1; break;
case 'v': args->vcf_list = optarg; break;
case 'c':
if ( !strcmp(optarg,"snps") ) collapse |= COLLAPSE_SNPS;
else if ( !strcmp(optarg,"indels") ) collapse |= COLLAPSE_INDELS;
else if ( !strcmp(optarg,"both") ) collapse |= COLLAPSE_SNPS | COLLAPSE_INDELS;
else if ( !strcmp(optarg,"any") ) collapse |= COLLAPSE_ANY;
else if ( !strcmp(optarg,"all") ) collapse |= COLLAPSE_ANY;
else if ( !strcmp(optarg,"some") ) collapse |= COLLAPSE_SOME;
else error("The --collapse string \"%s\" not recognised.\n", optarg);
break;
case 'a':
{
kstring_t str = {0,0,0};
kputs("%CHROM\t%POS\t%MASK\t%REF\t%ALT\t%", &str);
char *p = optarg;
while ( *p )
{
if ( *p==',' )
kputs("\t%", &str);
else
kputc(*p, &str);
p++;
}
kputc('\n', &str);
args->format_str = str.s;
break;
}
case 'e': args->filter_str = optarg; args->filter_logic |= FLT_EXCLUDE; break;
case 'i': args->filter_str = optarg; args->filter_logic |= FLT_INCLUDE; break;
case 'r': args->regions_list = optarg; break;
case 'R': args->regions_list = optarg; regions_is_file = 1; break;
case 't': args->targets_list = optarg; break;
case 'T': args->targets_list = optarg; targets_is_file = 1; break;
case 'l': args->list_columns = 1; break;
case 'u': args->allow_undef_tags = 1; break;
case 's': args->sample_list = optarg; break;
case 'S': args->sample_list = optarg; args->sample_is_file = 1; break;
case 'h':
case '?': usage();
default: error("Unknown argument: %s\n", optarg);
}
}
char *fname = NULL;
if ( optind>=argc )
{
if ( !isatty(fileno((FILE *)stdin)) ) fname = "-";
}
else fname = argv[optind];
if ( args->list_columns )
{
if ( !fname ) error("Missing the VCF file name\n");
args->files = bcf_sr_init();
if ( !bcf_sr_add_reader(args->files, fname) ) error("Failed to open %s: %s\n", fname,bcf_sr_strerror(args->files->errnum));
list_columns(args);
bcf_sr_destroy(args->files);
free(args);
return 0;
}
if ( !args->format_str ) usage();
args->out = args->fn_out ? fopen(args->fn_out, "w") : stdout;
if ( !args->out ) error("%s: %s\n", args->fn_out,strerror(errno));
if ( !args->vcf_list )
{
if ( !fname ) usage();
args->files = bcf_sr_init();
args->files->collapse = collapse;
if ( optind+1 < argc ) args->files->require_index = 1;
if ( args->regions_list && bcf_sr_set_regions(args->files, args->regions_list, regions_is_file)<0 )
error("Failed to read the regions: %s\n", args->regions_list);
if ( args->targets_list )
{
if ( bcf_sr_set_targets(args->files, args->targets_list, targets_is_file, 0)<0 )
error("Failed to read the targets: %s\n", args->targets_list);
}
while ( fname )
{
if ( !bcf_sr_add_reader(args->files, fname) ) error("Failed to open %s: %s\n", fname,bcf_sr_strerror(args->files->errnum));
fname = ++optind < argc ? argv[optind] : NULL;
}
init_data(args);
query_vcf(args);
free(args->format_str);
destroy_data(args);
bcf_sr_destroy(args->files);
fclose(args->out);
free(args);
return 0;
}
// multiple VCFs
int i, k, nfiles, prev_nsamples = 0;
char **fnames, **prev_samples = NULL;
fnames = hts_readlist(args->vcf_list, 1, &nfiles);
if ( !nfiles ) error("No files in %s?\n", args->vcf_list);
for (i=0; i<nfiles; i++)
{
args->files = bcf_sr_init();
args->files->collapse = collapse;
if ( args->regions_list && bcf_sr_set_regions(args->files, args->regions_list, regions_is_file)<0 )
error("Failed to read the regions: %s\n", args->regions_list);
if ( optind < argc ) args->files->require_index = 1;
if ( args->targets_list )
{
if ( bcf_sr_set_targets(args->files, args->targets_list,targets_is_file, 0)<0 )
error("Failed to read the targets: %s\n", args->targets_list);
}
if ( !bcf_sr_add_reader(args->files, fnames[i]) ) error("Failed to open %s: %s\n", fnames[i],bcf_sr_strerror(args->files->errnum));
for (k=optind; k<argc; k++)
if ( !bcf_sr_add_reader(args->files, argv[k]) ) error("Failed to open %s: %s\n", argv[k],bcf_sr_strerror(args->files->errnum));
init_data(args);
if ( i==0 )
prev_samples = copy_header(args->header, args->files->readers[0].header->samples, bcf_hdr_nsamples(args->files->readers[0].header));
else
{
args->print_header = 0;
if ( compare_header(args->header, args->files->readers[0].header->samples, bcf_hdr_nsamples(args->files->readers[0].header), prev_samples, prev_nsamples) )
error("Different samples in %s and %s\n", fnames[i-1],fnames[i]);
}
query_vcf(args);
destroy_data(args);
bcf_sr_destroy(args->files);
}
fclose(args->out);
destroy_list(fnames, nfiles);
destroy_list(prev_samples, prev_nsamples);
free(args->format_str);
free(args);
return 0;
}
int main_vcfgtcheck(int argc, char *argv[])
{
int c;
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->files = bcf_sr_init();
args->argc = argc; args->argv = argv; set_cwd(args);
char *regions = NULL, *targets = NULL;
int regions_is_file = 0, targets_is_file = 0;
// In simulated sample swaps the minimum error was 0.3 and maximum intra-sample error was 0.23
// - min_inter: pairs with smaller err value will be considered identical
// - max_intra: pairs with err value bigger than abs(max_intra_err) will be considered
// different. If negative, the cutoff may be heuristically lowered
args->min_inter_err = 0.23;
args->max_intra_err = -0.3;
static struct option loptions[] =
{
{"cluster",1,0,'c'},
{"GTs-only",1,0,'G'},
{"all-sites",0,0,'a'},
{"homs-only",0,0,'H'},
{"help",0,0,'h'},
{"genotypes",1,0,'g'},
{"plot",1,0,'p'},
{"target-sample",1,0,'S'},
{"query-sample",1,0,'s'},
{"regions",1,0,'r'},
{"regions-file",1,0,'R'},
{"targets",1,0,'t'},
{"targets-file",1,0,'T'},
{0,0,0,0}
};
char *tmp;
while ((c = getopt_long(argc, argv, "hg:p:s:S:Hr:R:at:T:G:c:",loptions,NULL)) >= 0) {
switch (c) {
case 'c':
args->min_inter_err = strtod(optarg,&tmp);
if ( *tmp )
{
if ( *tmp!=',') error("Could not parse: -c %s\n", optarg);
args->max_intra_err = strtod(tmp+1,&tmp);
if ( *tmp ) error("Could not parse: -c %s\n", optarg);
}
break;
case 'G':
args->no_PLs = strtol(optarg,&tmp,10);
if ( *tmp ) error("Could not parse argument: --GTs-only %s\n", optarg);
break;
case 'a': args->all_sites = 1; break;
case 'H': args->hom_only = 1; break;
case 'g': args->gt_fname = optarg; break;
case 'p': args->plot = optarg; break;
case 'S': args->target_sample = optarg; break;
case 's': args->query_sample = optarg; break;
case 'r': regions = optarg; break;
case 'R': regions = optarg; regions_is_file = 1; break;
case 't': targets = optarg; break;
case 'T': targets = optarg; targets_is_file = 1; break;
case 'h':
case '?': usage();
default: error("Unknown argument: %s\n", optarg);
}
}
char *fname = NULL;
if ( optind==argc )
{
if ( !isatty(fileno((FILE *)stdin)) ) fname = "-"; // reading from stdin
else usage(); // no files given
}
else fname = argv[optind];
if ( argc>optind+1 ) usage(); // too many files given
if ( !args->gt_fname ) args->cross_check = 1; // no genotype file, run in cross-check mode
else args->files->require_index = 1;
if ( regions && bcf_sr_set_regions(args->files, regions, regions_is_file)<0 ) error("Failed to read the regions: %s\n", regions);
if ( targets && bcf_sr_set_targets(args->files, targets, targets_is_file, 0)<0 ) error("Failed to read the targets: %s\n", targets);
if ( !bcf_sr_add_reader(args->files, fname) ) error("Failed to open %s: %s\n", fname,bcf_sr_strerror(args->files->errnum));
if ( args->gt_fname && !bcf_sr_add_reader(args->files, args->gt_fname) ) error("Failed to open %s: %s\n", args->gt_fname,bcf_sr_strerror(args->files->errnum));
args->files->collapse = COLLAPSE_SNPS|COLLAPSE_INDELS;
if ( args->plot ) args->plot = init_prefix(args->plot);
init_data(args);
if ( args->cross_check )
cross_check_gts(args);
else
check_gt(args);
destroy_data(args);
bcf_sr_destroy(args->files);
if (args->plot) free(args->plot);
free(args);
return 0;
}
int run(int argc, char **argv)
{
args_t *args = (args_t*) calloc(1,sizeof(args_t));
args->nsites = 10;
args->min_hets = 0.3;
args->background = "X:60001-2699520";
static struct option loptions[] =
{
{"verbose",1,0,'v'},
{"ploidy",1,0,'p'},
{"nsites",1,0,'n'},
{"guess",1,0,'g'},
{"min-hets",1,0,'m'},
{"background",1,0,'b'},
{0,0,0,0}
};
char c, *tmp, *ploidy_fname = NULL;
while ((c = getopt_long(argc, argv, "p:n:g:m:vb:",loptions,NULL)) >= 0)
{
switch (c) {
case 'b':
if ( !strcmp("-",optarg) ) args->background = NULL;
else args->background = optarg;
break;
case 'v': args->verbose = 1; break;
case 'g':
if ( !strcasecmp(optarg,"GT") ) args->guess = GUESS_GT;
else if ( !strcasecmp(optarg,"PL") ) args->guess = GUESS_PL;
else if ( !strcasecmp(optarg,"GL") ) args->guess = GUESS_GL;
else error("The argument not recognised, expected --guess GT, --guess PL or --guess GL: %s\n", optarg);
break;
case 'm':
args->min_hets = strtod(optarg,&tmp);
if ( *tmp ) error("Unexpected argument to --min-hets: %s\n", optarg);
break;
case 'p': ploidy_fname = optarg; break;
case 'n':
args->nsites = strtol(optarg,&tmp,10);
if (*tmp) error("Unexpected argument to --nsites: %s\n", optarg); break;
case 'h':
case '?':
default: error("%s", usage()); break;
}
}
args->sr = bcf_sr_init();
args->sr->require_index = 1;
if ( !argv[0] ) error("%s", usage());
if ( !bcf_sr_add_reader(args->sr,argv[0]) ) error("Error: %s\n", bcf_sr_strerror(args->sr->errnum));
args->hdr = args->sr->readers[0].header;
args->nsample = bcf_hdr_nsamples(args->hdr);
args->dflt_ploidy = 2;
if ( ploidy_fname )
{
args->ploidy = ploidy_init(ploidy_fname, args->dflt_ploidy);
if ( !args->ploidy ) error("Could not read %s\n", ploidy_fname);
}
else
{
args->ploidy = ploidy_init_string(
"X 1 60000 M 1\n"
"X 2699521 154931043 M 1\n"
"Y 1 59373566 M 1\n"
"Y 1 59373566 F 0\n", args->dflt_ploidy);
}
args->nsex = ploidy_nsex(args->ploidy);
args->sex2ploidy = (int*) malloc(sizeof(int)*args->nsex);
args->max_ploidy = ploidy_max(args->ploidy);
if ( args->guess && args->max_ploidy > 2 ) error("Sorry, ploidy %d not supported with -g\n", args->max_ploidy);
args->ncounts = args->nsample * ((args->max_ploidy>2 ? args->max_ploidy : 2)+1);
args->counts = (int*) malloc(sizeof(int)*args->ncounts);
args->bg_counts = (count_t*) calloc(args->nsample,sizeof(count_t));
args->sex2prob = (float*) calloc(args->nsample*args->nsex,sizeof(float));
int i, nseq;
for (i=0; i<args->nsample*args->nsex; i++) args->sex2prob[i] = 1;
if ( args->verbose && args->guess )
printf("# [1]REG\t[2]Region\t[3]Sample\t[4]Het fraction\t[5]nHet\t[6]nHom\t[7]nMissing\n");
// First get the counts from expected haploid regions
regidx_t *idx = ploidy_regions(args->ploidy);
char **seqs = regidx_seq_names(idx, &nseq);
for (i=0; i<nseq; i++)
{
regitr_t itr;
regidx_overlap(idx, seqs[i], 0, UINT32_MAX, &itr);
while ( itr.i < itr.n )
{
if ( args->guess )
itr.i += process_region_guess(args, seqs[i], &itr);
else
itr.i += process_region_precise(args, seqs[i], &itr);
}
}
// Get the counts from a PAR (the background diploid region) and see if the fraction
// of hets is different
if ( args->guess ) sex2prob_guess(args);
for (i=0; i<args->nsample; i++)
{
int j, jmax = 0;
float max = 0, sum = 0;
for (j=0; j<args->nsex; j++)
{
sum += args->sex2prob[i*args->nsex+j];
if ( max < args->sex2prob[i*args->nsex+j] )
{
jmax = j;
max = args->sex2prob[i*args->nsex+j];
}
}
if ( args->verbose )
printf("%s\t%s\t%f\n", args->hdr->samples[i],ploidy_id2sex(args->ploidy,jmax),args->sex2prob[i*args->nsex+jmax]/sum);
else
printf("%s\t%s\n", args->hdr->samples[i],ploidy_id2sex(args->ploidy,jmax));
}
bcf_sr_destroy(args->sr);
ploidy_destroy(args->ploidy);
destroy_regs(args);
free(args->sex2ploidy);
free(args->counts);
free(args->bg_counts);
free(args->gts);
free(args->pls);
free(args->sex2prob);
free(args);
return 0;
}
bcf_srs() { m_sr = bcf_sr_init(); }
static void init_data(args_t *args)
{
args->aux.srs = bcf_sr_init();
// Open files for input and output, initialize structures
if ( args->targets )
{
if ( bcf_sr_set_targets(args->aux.srs, args->targets, args->targets_is_file, args->aux.flag&CALL_CONSTR_ALLELES ? 3 : 0)<0 )
error("Failed to read the targets: %s\n", args->targets);
if ( args->aux.flag&CALL_CONSTR_ALLELES && args->flag&CF_INS_MISSED )
{
args->aux.srs->targets->missed_reg_handler = print_missed_line;
args->aux.srs->targets->missed_reg_data = args;
}
}
if ( args->regions )
{
if ( bcf_sr_set_regions(args->aux.srs, args->regions, args->regions_is_file)<0 )
error("Failed to read the targets: %s\n", args->regions);
}
if ( !bcf_sr_add_reader(args->aux.srs, args->bcf_fname) ) error("Failed to open %s: %s\n", args->bcf_fname,bcf_sr_strerror(args->aux.srs->errnum));
args->aux.hdr = bcf_sr_get_header(args->aux.srs,0);
int i;
if ( args->samples_fname )
{
set_samples(args, args->samples_fname, args->samples_is_file);
if ( args->aux.flag&CALL_CONSTR_TRIO )
{
if ( 3*args->aux.nfams!=args->nsamples ) error("Expected only trios in %s, sorry!\n", args->samples_fname);
fprintf(stderr,"Detected %d samples in %d trio families\n", args->nsamples,args->aux.nfams);
}
args->nsex = ploidy_nsex(args->ploidy);
args->sex2ploidy = (int*) calloc(args->nsex,sizeof(int));
args->sex2ploidy_prev = (int*) calloc(args->nsex,sizeof(int));
args->aux.ploidy = (uint8_t*) malloc(args->nsamples);
for (i=0; i<args->nsamples; i++) args->aux.ploidy[i] = 2;
for (i=0; i<args->nsex; i++) args->sex2ploidy_prev[i] = 2;
}
if ( args->samples_map )
{
args->aux.hdr = bcf_hdr_subset(bcf_sr_get_header(args->aux.srs,0), args->nsamples, args->samples, args->samples_map);
if ( !args->aux.hdr ) error("Error occurred while subsetting samples\n");
for (i=0; i<args->nsamples; i++)
if ( args->samples_map[i]<0 ) error("No such sample: %s\n", args->samples[i]);
if ( !bcf_hdr_nsamples(args->aux.hdr) ) error("No matching sample found\n");
}
else
{
args->aux.hdr = bcf_hdr_dup(bcf_sr_get_header(args->aux.srs,0));
for (i=0; i<args->nsamples; i++)
if ( bcf_hdr_id2int(args->aux.hdr,BCF_DT_SAMPLE,args->samples[i])<0 )
error("No such sample: %s\n", args->samples[i]);
}
args->out_fh = hts_open(args->output_fname, hts_bcf_wmode(args->output_type));
if ( args->out_fh == NULL ) error("Can't write to \"%s\": %s\n", args->output_fname, strerror(errno));
if ( args->flag & CF_QCALL )
return;
if ( args->flag & CF_MCALL )
mcall_init(&args->aux);
if ( args->flag & CF_CCALL )
ccall_init(&args->aux);
if ( args->flag&CF_GVCF )
{
bcf_hdr_append(args->aux.hdr,"##INFO=<ID=END,Number=1,Type=Integer,Description=\"End position of the variant described in this record\">");
args->gvcf.rid = -1;
args->gvcf.line = bcf_init1();
args->gvcf.gt = (int32_t*) malloc(2*sizeof(int32_t)*bcf_hdr_nsamples(args->aux.hdr));
for (i=0; i<bcf_hdr_nsamples(args->aux.hdr); i++)
{
args->gvcf.gt[2*i+0] = bcf_gt_unphased(0);
args->gvcf.gt[2*i+1] = bcf_gt_unphased(0);
}
}
bcf_hdr_remove(args->aux.hdr, BCF_HL_INFO, "QS");
bcf_hdr_remove(args->aux.hdr, BCF_HL_INFO, "I16");
bcf_hdr_append_version(args->aux.hdr, args->argc, args->argv, "bcftools_call");
bcf_hdr_write(args->out_fh, args->aux.hdr);
if ( args->flag&CF_INS_MISSED ) init_missed_line(args);
}
