void pe_clean_core(char *fa_fn, clean_opt *opt) {
bwa_seq_t *seqs, *s = NULL;
int n_seqs = 0, i = 0;
char *item = (char*) malloc(BUFSIZE), *solid = malloc(BUFSIZE);
FILE *solid_file;
clock_t t = clock();
GPtrArray *solid_reads = NULL;
show_debug_msg(__func__, "Loading library %s...\n", fa_fn);
seqs = load_reads(fa_fn, &n_seqs);
show_debug_msg(__func__, "Saving k-mer frequencies: %.2f sec...\n",
(float) (clock() - t) / CLOCKS_PER_SEC);
sprintf(solid, "%s.solid", opt->lib_name);
solid_file = xopen(solid, "w");
solid_reads = calc_solid_reads(seqs, n_seqs, opt, n_seqs * opt->stop_thre,
0, 1);
for (i = 0; i < solid_reads->len; i++) {
s = g_ptr_array_index(solid_reads, i);
sprintf(item, "%s\n", s->name);
fputs(item, solid_file);
}
free(item);
free(solid);
g_ptr_array_free(solid_reads, TRUE);
bwa_free_read_seq(n_seqs, seqs);
fclose(solid_file);
}
void Depot::load_reads(ReadSet& dst) {
load_reads(dst, 0, -1); // read all
}
Read* Depot::load_read(uint32_t index) {
ReadSet temp;
load_reads(temp, index, 1);
return temp.front();
}
int main(int argc, char **argv)
{
// compiler complains about unused function without these linese
(void)kh_clear_ghash;
(void)kh_del_ghash;
if(argc < 2) print_usage(usage, NULL);
char swap_alleles = 0;
int c;
while((c = getopt(argc, argv, "s")) >= 0) {
switch (c) {
case 's': swap_alleles = 1; break;
default: die("Unknown option: %c", c);
}
}
if(optind == argc) print_usage(usage, "Not enough arguments");
char *inputpath = argv[optind];
char **refpaths = argv + optind + 1;
size_t num_refs = argc - optind - 1;
gzFile gzin = gzopen(inputpath, "r");
if(gzin == NULL) die("Cannot read file: %s", inputpath);
size_t i, nchroms = 0, capacity = 1024;
khash_t(ghash) *genome = kh_init(ghash);
read_t *reads = malloc(capacity * sizeof(read_t)), *r;
int hret;
khiter_t k;
for(i = 0; i < num_refs; i++) {
fprintf(stderr, "Loading %s\n", refpaths[i]);
load_reads(refpaths[i], &reads, &capacity, &nchroms);
}
if(num_refs == 0) {
fprintf(stderr, "Loading from stdin\n");
load_reads("-", &reads, &capacity, &nchroms);
}
if(nchroms == 0) die("No chromosomes loaded");
for(i = 0; i < nchroms; i++) {
r = reads + i;
fprintf(stderr, "Loaded: '%s'\n", r->name.b);
k = kh_put(ghash, genome, r->name.b, &hret);
if(hret == 0) warn("Duplicate read name (taking first): %s", r->name.b);
else kh_value(genome, k) = r;
}
// Now read VCF
StrBuf line;
strbuf_alloc(&line, 1024);
char *fields[9];
char *chr;
int pos, reflen, altlen;
while(strbuf_reset_gzreadline(&line, gzin) > 0)
{
if(line.b[0] == '#') fputs(line.b, stdout);
else
{
strbuf_chomp(&line);
vcf_columns(line.b, fields);
fields[1][-1] = fields[2][-1] = '\0';
chr = line.b;
pos = atoi(fields[1])-1;
k = kh_get(ghash, genome, chr);
r = kh_value(genome, k);
fields[1][-1] = fields[2][-1] = '\t';
reflen = fields[4] - fields[3] - 1;
altlen = fields[5] - fields[4] - 1;
if(k == kh_end(genome)) warn("Cannot find chrom: %s", chr);
else if(pos < 0) warn("Bad line: %s\n", line.b);
else if((reflen == 1 && altlen == 1) || fields[3][0] == fields[4][0])
{
if((unsigned)pos + reflen <= r->seq.end &&
strncasecmp(r->seq.b+pos,fields[3],reflen) == 0)
{
fputs(line.b, stdout);
fputc('\n', stdout);
}
else if(swap_alleles && (unsigned)pos + altlen <= r->seq.end &&
strncasecmp(r->seq.b+pos,fields[4],altlen) == 0)
{
// swap alleles
char tmp[altlen], *ref = fields[3], *alt = fields[4];
memcpy(tmp, alt, altlen);
memmove(ref+altlen+1, ref, reflen);
memcpy(ref, tmp, altlen);
ref[altlen] = '\t';
fputs(line.b, stdout);
fputc('\n', stdout);
}
// else printf("FAIL0\n");
}
// else printf("FAIL1\n");
}
}
kh_destroy(ghash, genome);
strbuf_dealloc(&line);
gzclose(gzin);
for(i = 0; i < nchroms; i++) seq_read_dealloc(reads+i);
free(reads);
fprintf(stderr, " Done.\n");
return 0;
}