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;
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'},
{"targets",1,0,'t'},
{0,0,0,0}
};
while ((c = getopt_long(argc, argv, "hg:p:s:S:Hr:at: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 't': targets = optarg; 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)<0 ) error("Failed to read the regions: %s\n", regions);
if ( targets && bcf_sr_set_targets(args->files, targets,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;
}
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);
}
int i;
if ( !bcf_sr_add_reader(args->aux.srs, args->bcf_fname) ) error("Failed to open: %s\n", args->bcf_fname);
if ( args->nsamples && args->nsamples != bcf_hdr_nsamples(args->aux.srs->readers[0].header) )
{
args->samples_map = (int *) malloc(sizeof(int)*args->nsamples);
args->aux.hdr = bcf_hdr_subset(args->aux.srs->readers[0].header, args->nsamples, args->samples, args->samples_map);
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(args->aux.srs->readers[0].header);
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]);
}
// Reorder ploidy and family indexes to match mpileup's output and exclude samples which are not available
if ( args->aux.ploidy )
{
for (i=0; i<args->aux.nfams; i++)
{
int j;
for (j=0; j<3; j++)
{
int k = bcf_hdr_id2int(args->aux.hdr, BCF_DT_SAMPLE, args->samples[ args->aux.fams[i].sample[j] ]);
if ( k<0 ) error("No such sample: %s\n", args->samples[ args->aux.fams[i].sample[j] ]);
args->aux.fams[i].sample[j] = k;
}
}
uint8_t *ploidy = (uint8_t*) calloc(bcf_hdr_nsamples(args->aux.hdr), 1);
for (i=0; i<args->nsamples; i++) // i index in -s sample list
{
int j = bcf_hdr_id2int(args->aux.hdr, BCF_DT_SAMPLE, args->samples[i]); // j index in the output VCF / subset VCF
if ( j<0 )
{
fprintf(stderr,"Warning: no such sample: \"%s\"\n", args->samples[i]);
continue;
}
ploidy[j] = args->aux.ploidy[i];
}
args->nsamples = bcf_hdr_nsamples(args->aux.hdr);
for (i=0; i<args->nsamples; i++)
assert( ploidy[i]==0 || ploidy[i]==1 || ploidy[i]==2 );
free(args->aux.ploidy);
args->aux.ploidy = ploidy;
}
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);
}
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_fname = NULL;
args->output_type = FT_VCF;
args->n_threads = 0;
args->record_cmd_line = 1;
int targets_is_file = 0, regions_is_file = 0;
static struct option loptions[] =
{
{"help",no_argument,NULL,'h'},
{"exclude",required_argument,NULL,'e'},
{"include",required_argument,NULL,'i'},
{"collapse",required_argument,NULL,'c'},
{"complement",no_argument,NULL,'C'},
{"apply-filters",required_argument,NULL,'f'},
{"nfiles",required_argument,NULL,'n'},
{"prefix",required_argument,NULL,'p'},
{"write",required_argument,NULL,'w'},
{"targets",required_argument,NULL,'t'},
{"targets-file",required_argument,NULL,'T'},
{"regions",required_argument,NULL,'r'},
{"regions-file",required_argument,NULL,'R'},
{"output",required_argument,NULL,'o'},
{"output-type",required_argument,NULL,'O'},
{"threads",required_argument,NULL,9},
{"no-version",no_argument,NULL,8},
{NULL,0,NULL,0}
};
while ((c = getopt_long(argc, argv, "hc:r:R:p:n:w:t:T:Cf:o:O:i:e:",loptions,NULL)) >= 0) {
switch (c) {
case 'o': args->output_fname = optarg; break;
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_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 'p': args->prefix = optarg; break;
case 'w': args->write_files = optarg; break;
case 'i': add_filter(args, optarg, FLT_INCLUDE); break;
case 'e': add_filter(args, optarg, FLT_EXCLUDE); 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 ( *p=='~' ) { args->isec_op = OP_EXACT; p++; }
else if ( isdigit(*p) ) args->isec_op = OP_EQUAL;
else error("Could not parse --nfiles %s\n", optarg);
if ( args->isec_op == OP_EXACT ) args->isec_exact = p;
else if ( sscanf(p,"%d",&args->isec_n)!=1 ) error("Could not parse --nfiles %s\n", optarg);
}
break;
case 9 : args->n_threads = strtol(optarg, 0, 0); break;
case 8 : args->record_cmd_line = 0; break;
case 'h':
case '?': usage();
default: error("Unknown argument: %s\n", optarg);
}
}
if ( argc-optind<1 ) usage(); // no file given
if ( args->targets_list && 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->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 ( argc-optind==2 && !args->isec_op )
{
args->isec_op = OP_VENN;
if ( !args->prefix ) error("Expected the -p option\n");
}
if ( !args->targets_list )
{
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: %s\n", argv[optind],bcf_sr_strerror(args->files->errnum));
optind++;
}
init_data(args);
isec_vcf(args);
destroy_data(args);
bcf_sr_destroy(args->files);
free(args);
return 0;
}
static void init_data(args_t *args)
{
bcf_srs_t *files = bcf_sr_init();
if ( args->regions_list )
{
if ( bcf_sr_set_regions(files, args->regions_list, args->regions_is_file)<0 )
error("Failed to read the regions: %s\n", args->regions_list);
}
if ( args->targets_list )
{
if ( bcf_sr_set_targets(files, args->targets_list, args->targets_is_file, 0)<0 )
error("Failed to read the targets: %s\n", args->targets_list);
}
if ( !bcf_sr_add_reader(files, args->fname) ) error("Failed to open %s: %s\n", args->fname,bcf_sr_strerror(files->errnum));
bcf_hdr_t *hdr = files->readers[0].header;
if ( !args->sample )
{
if ( bcf_hdr_nsamples(hdr)>1 ) error("Missing the option -s, --sample\n");
args->sample = hdr->samples[0];
}
else if ( bcf_hdr_id2int(hdr,BCF_DT_SAMPLE,args->sample)<0 ) error("No such sample: %s\n", args->sample);
int ret = bcf_hdr_set_samples(hdr, args->sample, 0);
if ( ret<0 ) error("Error setting the sample: %s\n", args->sample);
if ( !bcf_hdr_idinfo_exists(hdr,BCF_HL_FMT,bcf_hdr_id2int(hdr,BCF_DT_ID,"BAF")) )
error("The tag FORMAT/BAF is not present in the VCF: %s\n", args->fname);
int i;
args->xvals = (double*) calloc(args->nbins,sizeof(double));
for (i=0; i<args->nbins; i++) args->xvals[i] = 1.0*i/(args->nbins-1);
// collect BAF distributions for all chromosomes
int idist = -1, nbaf = 0, nprocessed = 0, ntotal = 0, prev_chr = -1;
float *baf = NULL;
while ( bcf_sr_next_line(files) )
{
ntotal++;
bcf1_t *line = bcf_sr_get_line(files,0);
if ( bcf_get_format_float(hdr,line,"BAF",&baf,&nbaf) != 1 ) continue;
if ( bcf_float_is_missing(baf[0]) ) continue;
nprocessed++;
if ( prev_chr==-1 || prev_chr!=line->rid )
{
// new chromosome
idist = args->ndist++;
args->dist = (dist_t*) realloc(args->dist, sizeof(dist_t)*args->ndist);
memset(&args->dist[idist],0,sizeof(dist_t));
args->dist[idist].chr = strdup(bcf_seqname(hdr,line));
args->dist[idist].yvals = (double*) calloc(args->nbins,sizeof(double));
args->dist[idist].xvals = args->xvals;
args->dist[idist].nvals = args->nbins;
prev_chr = line->rid;
}
int bin = baf[0]*(args->nbins-1);
args->dist[idist].yvals[bin]++; // the distribution
}
free(baf);
bcf_sr_destroy(files);
for (idist=0; idist<args->ndist; idist++)
{
#if 0
int j;
for (j=0; j<args->nbins; j++)
{
double x = args->dist[idist].xvals[j];
args->dist[idist].yvals[j] = exp(-(x-0.5)*(x-0.5)/1e-3);
}
#endif
init_dist(args, &args->dist[idist],args->verbose);
}
args->dat_fp = open_file(&args->dat_fname,"w","%s/dist.dat", args->output_dir);
fprintf(args->dat_fp, "# This file was produced by: bcftools polysomy(%s+htslib-%s), the command line was:\n", bcftools_version(),hts_version());
fprintf(args->dat_fp, "# \t bcftools %s ", args->argv[0]);
for (i=1; i<args->argc; i++)
fprintf(args->dat_fp, " %s",args->argv[i]);
fprintf(args->dat_fp,"\n#\n");
fprintf(args->dat_fp,"# DIST\t[2]Chrom\t[3]BAF\t[4]Normalized Count\n");
fprintf(args->dat_fp,"# FIT\t[2]Goodness of Fit\t[3]iFrom\t[4]iTo\t[5]The Fitted Function\n");
fprintf(args->dat_fp,"# CN\t[2]Chrom\t[3]Estimated Copy Number\t[4]Absolute fit deviation\n");
char *fname = NULL;
FILE *fp = open_file(&fname,"w","%s/dist.py", args->output_dir);
//-------- matplotlib script --------------
fprintf(fp,
"#!/usr/bin/env python\n"
"#\n"
"import matplotlib as mpl\n"
"mpl.use('Agg')\n"
"import matplotlib.pyplot as plt\n"
"import csv,sys,argparse\n"
"from math import exp\n"
"\n"
"outdir = '%s'\n"
"\n"
"def read_dat(dat,fit,cn):\n"
" csv.register_dialect('tab', delimiter='\t', quoting=csv.QUOTE_NONE)\n"
" with open(outdir+'/dist.dat', 'rb') as f:\n"
" reader = csv.reader(f, 'tab')\n"
" for row in reader:\n"
" if row[0][0]=='#': continue\n"
" type = row[0]\n"
" chr = row[1]\n"
" if type=='DIST':\n"
" if chr not in dat: dat[chr] = []\n"
" dat[chr].append(row)\n"
" elif type=='FIT':\n"
" if chr not in fit: fit[chr] = []\n"
" fit[chr].append(row)\n"
" elif type=='CN':\n"
" cn[chr] = row[2]\n"
"\n"
"def plot_dist(dat,fit,chr):\n"
" fig, ax = plt.subplots(1, 1, figsize=(7,5))\n"
" ax.plot([x[2] for x in dat[chr]],[x[3] for x in dat[chr]],'k-',label='Distribution')\n"
" if chr in fit:\n"
" for i in range(len(fit[chr])):\n"
" pfit = fit[chr][i]\n"
" exec('def xfit(x): return '+pfit[5])\n"
" istart = int(pfit[3])\n"
" iend = int(pfit[4])+1\n"
" vals = dat[chr][istart:iend]\n"
" args = {}\n"
" if i==0: args = {'label':'Target to Fit'}\n"
" ax.plot([x[2] for x in vals],[x[3] for x in vals],'r-',**args)\n"
" if i==0: args = {'label':'Best Fit'}\n"
" ax.plot([x[2] for x in vals],[xfit(float(x[2])) for x in vals],'g-',**args)\n"
" ax.set_title('BAF distribution, chr'+chr)\n"
" ax.set_xlabel('BAF')\n"
" ax.set_ylabel('Frequency')\n"
" ax.legend(loc='best',prop={'size':7},frameon=False)\n"
" plt.savefig(outdir+'/dist.chr'+chr+'.png')\n"
" plt.close()\n"
"\n"
"def plot_copy_number(cn):\n"
" fig, ax = plt.subplots(1, 1, figsize=(7,5))\n"
" xlabels = sorted(cn.keys())\n"
" xvals = range(len(xlabels))\n"
" yvals = [float(cn[x]) for x in xlabels]\n"
" ax.plot(xvals,yvals,'o',color='red')\n"
" for i in range(len(xvals)):\n"
" if yvals[i]==-1: ax.annotate('?', xy=(xvals[i],0.5),va='center',ha='center',color='red',fontweight='bold')\n"
" ax.tick_params(axis='both', which='major', labelsize=9)\n"
" ax.set_xticks(xvals)\n"
" ax.set_xticklabels(xlabels,rotation=45)\n"
" ax.set_xlim(-1,len(xlabels))\n"
" ax.set_ylim(0,5.0)\n"
" ax.set_yticks([1.0,2.0,3.0,4.0])\n"
" ax.set_xlabel('Chromosome')\n"
" ax.set_ylabel('Copy Number')\n"
" plt.savefig(outdir+'/copy-number.png')\n"
" plt.close()\n"
"\n"
"class myParser(argparse.ArgumentParser):\n"
" def error(self, message):\n"
" self.print_help()\n"
" sys.stderr.write('error: %%s\\n' %% message)\n"
" sys.exit(2)\n"
"\n"
"def main():\n"
" parser = myParser()\n"
" parser.add_argument('-a', '--all', action='store_true', help='Create all plots')\n"
" parser.add_argument('-c', '--copy-number', action='store_true', help='Create copy-number plot')\n"
" parser.add_argument('-d', '--distrib', metavar='CHR', help='Plot BAF distribution of a single chromosome')\n"
" args = parser.parse_args()\n"
" dat = {}; fit = {}; cn = {}\n"
" read_dat(dat,fit,cn)\n"
" if args.distrib!=None:\n"
" plot_dist(dat,fit,args.distrib)\n"
" if args.all:\n"
" for chr in dat: plot_dist(dat,fit,chr)\n"
" plot_copy_number(cn)\n"
" elif args.copy_number:\n"
" plot_copy_number(cn)\n"
" else:\n"
" for chr in dat: plot_dist(dat,fit,chr)\n"
"\n"
"if __name__ == '__main__':\n"
" main()\n",
args->output_dir);
//---------------------------------------
chmod(fname, S_IWUSR|S_IRUSR|S_IRGRP|S_IROTH|S_IXUSR|S_IXGRP|S_IXOTH);
free(fname);
fclose(fp);
}
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]);
args->out_hdr = bcf_hdr_merge(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\">");
}
if (args->record_cmd_line) 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));
if ( args->n_threads ) hts_set_threads(args->out_fh, args->n_threads);
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);
}
if ( args->remove_dups )
{
if ( !strcmp(args->remove_dups,"snps") ) args->files->collapse |= COLLAPSE_SNPS;
else if ( !strcmp(args->remove_dups,"indels") ) args->files->collapse |= COLLAPSE_INDELS;
else if ( !strcmp(args->remove_dups,"both") ) args->files->collapse |= COLLAPSE_SNPS | COLLAPSE_INDELS;
else if ( !strcmp(args->remove_dups,"any") ) args->files->collapse |= COLLAPSE_ANY;
else if ( !strcmp(args->remove_dups,"all") ) args->files->collapse |= COLLAPSE_ANY;
else if ( !strcmp(args->remove_dups,"none") ) args->files->collapse = COLLAPSE_NONE;
else error("The -D string \"%s\" not recognised.\n", args->remove_dups);
}
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;
}
}
static void init_data(args_t *args)
{
args->sr = bcf_sr_init();
if ( args->region )
{
args->sr->require_index = 1;
if ( bcf_sr_set_regions(args->sr, args->region, args->region_is_file)<0 ) error("Failed to read the regions: %s\n",args->region);
}
if ( args->target && bcf_sr_set_targets(args->sr, args->target, args->target_is_file, 0)<0 ) error("Failed to read the targets: %s\n",args->target);
if ( !bcf_sr_add_reader(args->sr,args->fname) ) error("Error: %s\n", bcf_sr_strerror(args->sr->errnum));
args->hdr_in = bcf_sr_get_header(args->sr,0);
args->hdr_out = bcf_hdr_dup(args->hdr_in);
if ( args->filter_str )
args->filter = filter_init(args->hdr_in, args->filter_str);
mkdir_p("%s/",args->output_dir);
int i, nsmpl = bcf_hdr_nsamples(args->hdr_in);
if ( !nsmpl ) error("No samples to split: %s\n", args->fname);
args->fh = (htsFile**)calloc(nsmpl,sizeof(*args->fh));
args->bnames = set_file_base_names(args);
kstring_t str = {0,0,0};
for (i=0; i<nsmpl; i++)
{
if ( !args->bnames[i] ) continue;
str.l = 0;
kputs(args->output_dir, &str);
if ( str.s[str.l-1] != '/' ) kputc('/', &str);
int k, l = str.l;
kputs(args->bnames[i], &str);
for (k=l; k<str.l; k++) if ( isspace(str.s[k]) ) str.s[k] = '_';
if ( args->output_type & FT_BCF ) kputs(".bcf", &str);
else if ( args->output_type & FT_GZ ) kputs(".vcf.gz", &str);
else kputs(".vcf", &str);
args->fh[i] = hts_open(str.s, hts_bcf_wmode(args->output_type));
if ( args->fh[i] == NULL ) error("Can't write to \"%s\": %s\n", str.s, strerror(errno));
bcf_hdr_nsamples(args->hdr_out) = 1;
args->hdr_out->samples[0] = args->bnames[i];
bcf_hdr_write(args->fh[i], args->hdr_out);
}
free(str.s);
// parse tags
int is_info = 0, is_fmt = 0;
char *beg = args->keep_tags;
while ( beg && *beg )
{
if ( !strncasecmp("INFO/",beg,5) ) { is_info = 1; is_fmt = 0; beg += 5; }
else if ( !strcasecmp("INFO",beg) ) { args->keep_info = 1; break; }
else if ( !strncasecmp("INFO,",beg,5) ) { args->keep_info = 1; beg += 5; continue; }
else if ( !strncasecmp("FMT/",beg,4) ) { is_info = 0; is_fmt = 1; beg += 4; }
else if ( !strncasecmp("FORMAT/",beg,7) ) { is_info = 0; is_fmt = 1; beg += 7; }
else if ( !strcasecmp("FMT",beg) ) { args->keep_fmt = 1; break; }
else if ( !strcasecmp("FORMAT",beg) ) { args->keep_fmt = 1; break; }
else if ( !strncasecmp("FMT,",beg,4) ) { args->keep_fmt = 1; beg += 4; continue; }
else if ( !strncasecmp("FORMAT,",beg,7) ) { args->keep_fmt = 1; beg += 7; continue; }
char *end = beg;
while ( *end && *end!=',' ) end++;
char tmp = *end; *end = 0;
int id = bcf_hdr_id2int(args->hdr_in, BCF_DT_ID, beg);
beg = tmp ? end + 1 : end;
if ( is_info && bcf_hdr_idinfo_exists(args->hdr_in,BCF_HL_INFO,id) )
{
if ( id >= args->ninfo_tags ) args->ninfo_tags = id + 1;
hts_expand0(uint8_t, args->ninfo_tags, args->minfo_tags, args->info_tags);
args->info_tags[id] = 1;
}
if ( is_fmt && bcf_hdr_idinfo_exists(args->hdr_in,BCF_HL_FMT,id) )
{
if ( id >= args->nfmt_tags ) args->nfmt_tags = id + 1;
hts_expand0(uint8_t, args->nfmt_tags, args->mfmt_tags, args->fmt_tags);
args->fmt_tags[id] = 1;
}
}
if ( !args->keep_info && !args->keep_fmt && !args->ninfo_tags && !args->nfmt_tags )
{
args->keep_info = args->keep_fmt = 1;
}
}
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!");
}
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}
};
char *tmp;
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 = 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 main_plugin(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->output_fname = "-";
args->output_type = FT_VCF;
args->nplugin_paths = -1;
int regions_is_file = 0, targets_is_file = 0, plist_only = 0;
if ( argc==1 ) usage(args);
char *plugin_name = NULL;
if ( argv[1][0]!='-' ) { plugin_name = argv[1]; argc--; argv++; }
static struct option loptions[] =
{
{"verbose",0,0,'v'},
{"help",0,0,'h'},
{"list-plugins",0,0,'l'},
{"output",1,0,'o'},
{"output-type",1,0,'O'},
{"include",1,0,'i'},
{"exclude",1,0,'e'},
{"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, "h?o:O:r:R:li:e:v",loptions,NULL)) >= 0)
{
switch (c) {
case 'v': args->verbose = 1; break;
case 'o': args->output_fname = optarg; break;
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 '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': plist_only = 1; break;
case '?':
case 'h': load_plugin(args, plugin_name, 1, &args->plugin); fprintf(stderr,"%s",args->plugin.usage()); return 0; break;
default: error("Unknown argument: %s\n", optarg);
}
}
if ( plist_only ) return list_plugins(args);
char *fname = NULL;
if ( optind>=argc || argv[optind][0]=='-' )
{
if ( !isatty(fileno((FILE *)stdin)) ) fname = "-"; // reading from stdin
else usage(args);
args->plugin.argc = argc - optind + 1;
args->plugin.argv = argv + optind - 1;
}
else
{
fname = argv[optind];
args->plugin.argc = argc - optind;
args->plugin.argv = argv + optind;
}
optind = 0;
args->plugin.argv[0] = plugin_name;
load_plugin(args, plugin_name, 1, &args->plugin);
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);
args->files->collapse |= COLLAPSE_SOME;
}
if ( !bcf_sr_add_reader(args->files, fname) ) error("Failed to open or the file not indexed: %s\n", fname);
init_data(args);
while ( bcf_sr_next_line(args->files) )
{
bcf1_t *line = bcf_sr_get_line(args->files,0);
if ( args->filter )
{
int pass = filter_test(args->filter, line, NULL);
if ( args->filter_logic & FLT_EXCLUDE ) pass = pass ? 0 : 1;
if ( !pass ) continue;
}
line = args->plugin.process(line);
if ( line ) bcf_write1(args->out_fh, args->hdr_out, line);
}
destroy_data(args);
bcf_sr_destroy(args->files);
free(args);
return 0;
}
int main_vcfview(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->clevel = -1;
args->print_header = 1;
args->update_info = 1;
args->output_type = FT_VCF;
int targets_is_file = 0, regions_is_file = 0;
static struct option loptions[] =
{
{"genotype",1,0,'g'},
{"compression-level",1,0,'l'},
{"header-only",0,0,'h'},
{"no-header",0,0,'H'},
{"exclude",1,0,'e'},
{"include",1,0,'i'},
{"trim-alt-alleles",0,0,'a'},
{"no-update",0,0,'I'},
{"drop-genotypes",0,0,'G'},
{"private",0,0,'x'},
{"exclude-private",0,0,'X'},
{"uncalled",0,0,'u'},
{"exclude-uncalled",0,0,'U'},
{"apply-filters",1,0,'f'},
{"known",0,0,'k'},
{"novel",0,0,'n'},
{"min-alleles",1,0,'m'},
{"max-alleles",1,0,'M'},
{"samples",1,0,'s'},
{"samples-file",1,0,'S'},
{"force-samples",0,0,1},
{"output-type",1,0,'O'},
{"output-file",1,0,'o'},
{"types",1,0,'v'},
{"exclude-types",1,0,'V'},
{"targets",1,0,'t'},
{"targets-file",1,0,'T'},
{"regions",1,0,'r'},
{"regions-file",1,0,'R'},
{"min-ac",1,0,'c'},
{"max-ac",1,0,'C'},
{"min-af",1,0,'q'},
{"max-af",1,0,'Q'},
{"phased",0,0,'p'},
{"exclude-phased",0,0,'P'},
{0,0,0,0}
};
char *tmp;
while ((c = getopt_long(argc, argv, "l:t:T:r:R:o:O:s:S:Gf:knv:V:m:M:auUhHc:C:Ii:e:xXpPq:Q:g:",loptions,NULL)) >= 0)
{
char allele_type[8] = "nref";
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 'l':
args->clevel = strtol(optarg,&tmp,10);
if ( *tmp ) error("Could not parse argument: --compression-level %s\n", optarg);
args->output_type |= FT_GZ;
break;
case 'o': args->fn_out = optarg; break;
case 'H': args->print_header = 0; break;
case 'h': args->header_only = 1; 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 's': args->sample_names = optarg; break;
case 'S': args->sample_names = optarg; args->sample_is_file = 1; break;
case 1 : args->force_samples = 1; break;
case 'a': args->trim_alts = 1; args->calc_ac = 1; break;
case 'I': args->update_info = 0; break;
case 'G': args->sites_only = 1; break;
case 'f': args->files->apply_filters = optarg; break;
case 'k': args->known = 1; break;
case 'n': args->novel = 1; break;
case 'm':
args->min_alleles = strtol(optarg,&tmp,10);
if ( *tmp ) error("Could not parse argument: --min-alleles %s\n", optarg);
break;
case 'M':
args->max_alleles = strtol(optarg,&tmp,10);
if ( *tmp ) error("Could not parse argument: --max-alleles %s\n", optarg);
break;
case 'v': args->include_types = optarg; break;
case 'V': args->exclude_types = optarg; 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 'c':
{
args->min_ac_type = ALLELE_NONREF;
if ( sscanf(optarg,"%d:%s",&args->min_ac, allele_type)!=2 && sscanf(optarg,"%d",&args->min_ac)!=1 )
error("Error: Could not parse --min-ac %s\n", optarg);
set_allele_type(&args->min_ac_type, allele_type);
args->calc_ac = 1;
break;
}
case 'C':
{
args->max_ac_type = ALLELE_NONREF;
if ( sscanf(optarg,"%d:%s",&args->max_ac, allele_type)!=2 && sscanf(optarg,"%d",&args->max_ac)!=1 )
error("Error: Could not parse --max-ac %s\n", optarg);
set_allele_type(&args->max_ac_type, allele_type);
args->calc_ac = 1;
break;
}
case 'q':
{
args->min_af_type = ALLELE_NONREF;
if ( sscanf(optarg,"%f:%s",&args->min_af, allele_type)!=2 && sscanf(optarg,"%f",&args->min_af)!=1 )
error("Error: Could not parse --min_af %s\n", optarg);
set_allele_type(&args->min_af_type, allele_type);
args->calc_ac = 1;
break;
}
case 'Q':
{
args->max_af_type = ALLELE_NONREF;
if ( sscanf(optarg,"%f:%s",&args->max_af, allele_type)!=2 && sscanf(optarg,"%f",&args->max_af)!=1 )
error("Error: Could not parse --min_af %s\n", optarg);
set_allele_type(&args->max_af_type, allele_type);
args->calc_ac = 1;
break;
}
case 'x': args->private_vars |= FLT_INCLUDE; args->calc_ac = 1; break;
case 'X': args->private_vars |= FLT_EXCLUDE; args->calc_ac = 1; break;
case 'u': args->uncalled |= FLT_INCLUDE; args->calc_ac = 1; break;
case 'U': args->uncalled |= FLT_EXCLUDE; args->calc_ac = 1; break;
case 'p': args->phased |= FLT_INCLUDE; break; // phased
case 'P': args->phased |= FLT_EXCLUDE; break; // exclude-phased
case 'g':
{
if ( !strcasecmp(optarg,"hom") ) args->gt_type = GT_NEED_HOM;
else if ( !strcasecmp(optarg,"het") ) args->gt_type = GT_NEED_HET;
else if ( !strcasecmp(optarg,"miss") ) args->gt_type = GT_NEED_MISSING;
else if ( !strcasecmp(optarg,"^hom") ) args->gt_type = GT_NO_HOM;
else if ( !strcasecmp(optarg,"^het") ) args->gt_type = GT_NO_HET;
else if ( !strcasecmp(optarg,"^miss") ) args->gt_type = GT_NO_MISSING;
else error("The argument to -g not recognised. Expected one of hom/het/miss/^hom/^het/^miss, got \"%s\".\n", optarg);
break;
}
case '?': usage(args);
default: error("Unknown argument: %s\n", optarg);
}
}
if ( args->filter_logic == (FLT_EXCLUDE|FLT_INCLUDE) ) error("Only one of -i or -e can be given.\n");
if ( args->private_vars > FLT_EXCLUDE ) error("Only one of -x or -X can be given.\n");
if ( args->uncalled > FLT_EXCLUDE ) error("Only one of -u or -U can be given.\n");
if ( args->phased > FLT_EXCLUDE ) error("Only one of -p or -P can be given.\n");
if ( args->sample_names && args->update_info) args->calc_ac = 1;
char *fname = NULL;
if ( optind>=argc )
{
if ( !isatty(fileno((FILE *)stdin)) ) fname = "-"; // reading from stdin
else usage(args);
}
else fname = argv[optind];
// read in the regions from the command line
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);
}
else if ( optind+1 < argc )
{
int i;
kstring_t tmp = {0,0,0};
kputs(argv[optind+1],&tmp);
for (i=optind+2; i<argc; i++) { kputc(',',&tmp); kputs(argv[i],&tmp); }
if ( bcf_sr_set_regions(args->files, tmp.s, 0)<0 )
error("Failed to read the regions: %s\n", tmp.s);
free(tmp.s);
}
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, fname) ) error("Failed to open %s: %s\n", fname,bcf_sr_strerror(args->files->errnum));
init_data(args);
bcf_hdr_t *out_hdr = args->hnull ? args->hnull : (args->hsub ? args->hsub : args->hdr);
if (args->print_header)
bcf_hdr_write(args->out, out_hdr);
else if ( args->output_type & FT_BCF )
error("BCF output requires header, cannot proceed with -H\n");
if (!args->header_only)
{
while ( bcf_sr_next_line(args->files) )
{
bcf1_t *line = args->files->readers[0].buffer[0];
if ( line->errcode && out_hdr!=args->hdr ) error("Undefined tags in the header, cannot proceed in the sample subset mode.\n");
if ( subset_vcf(args, line) )
bcf_write1(args->out, out_hdr, line);
}
}
hts_close(args->out);
destroy_data(args);
bcf_sr_destroy(args->files);
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;
}
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;
}
void set_region(const std::string ®ion) {
if (bcf_sr_set_regions(m_sr, region.c_str(), false) != 0)
throw std::runtime_error("Could not set region: [" + region + "]");
assert(m_sr->require_index == 1);
}
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);
}
