/**
* @param ref_pos is 0-based here
* @param info is extra text to print in the info field of each variant (may be NULL)
* @param genotypes is strings to print in genotypes columns, of length num_samples.
* It may be NULL.
* @return number of variants printed
*/
static void align_biallelic(const char *ref, const char *alt,
const read_t *chrom,
const uint8_t *gts, size_t nsamples,
CallDecomp *dc, const AlignedCall *call,
size_t max_allele_len)
{
int32_t start, len;
size_t ref_nbases, alt_nbases, ref_pos = call->start, ref_end, vcf_pos;
int8_t prev_base, next_base;
bool is_snp;
// printf("--\n ref: %s\n alt: %s\n", ref, alt);
while((start = align_get_start(ref, alt)) > -1)
{
ref_pos += start; // assume ref[i]==alt[i] means ref[i]!='-'
ref += start;
alt += start;
len = align_get_end(ref, alt);
// printf("ref: %.*s\nalt: %.*s\nref_pos: %zu start: %i len %i\n",
// len, ref, len, alt, ref_pos, start, len);
ref_nbases = align_get_nbases(ref, len);
alt_nbases = align_get_nbases(alt, len);
is_snp = (ref_nbases == 1 && alt_nbases == 1);
ref_end = ref_pos+ref_nbases;
vcf_pos = ref_pos; // copy in case we need left padding base
// If one allele is going to be empty, we need a padding base
// If ref_pos == 0, add extra base to end instead
prev_base = next_base = -1;
if(!is_snp) {
if(ref_pos > 0) prev_base = chrom->seq.b[--vcf_pos];
else if(ref_end < chrom->seq.end) next_base = chrom->seq.b[ref_end];
}
if(is_snp || prev_base > 0 || next_base > 0) {
print_vcf_entry(vcf_pos, prev_base, next_base, ref, alt, len,
gts, nsamples, dc, call, max_allele_len);
}
ref_pos += ref_nbases;
ref += len;
alt += len;
}
}
/**
* @param ref_pos is 0-based here
* @param info is extra text to print in the info field of each variant (may be NULL)
* @param genotypes is strings to print in genotypes columns, of length num_samples.
* It may be NULL.
*/
static void align_biallelic(const char *ref, const char *alt,
const read_t *chr, size_t ref_pos,
const char *info, const char **genotypes,
FILE *fout)
{
int start, len;
size_t ref_allele_len, alt_allele_len;
int prev_base, vcf_pos;
bool is_snp;
// printf("--\n ref: %s\n alt: %s\n", ref, alt);
while((start = align_get_start(ref, alt)) > -1)
{
ref_pos += start; // assume ref[i]==alt[i] means ref[i]!='-'
ref += start;
alt += start;
len = align_get_end(ref, alt);
// printf("ref: %.*s\nalt: %.*s\nref_pos: %zu start: %i len %i\n",
// len, ref, len, alt, ref_pos, start, len);
ref_allele_len = align_get_len(ref, len);
alt_allele_len = align_get_len(alt, len);
is_snp = (ref_allele_len == 1 && alt_allele_len == 1);
vcf_pos = ref_pos+1; // Convert to 1-based
if(!is_snp) {
prev_base = ref_pos > 0 ? chr->seq.b[ref_pos-1] : 'N';
vcf_pos--;
} else {
prev_base = -1;
}
print_vcf_entry(chr->name.b, vcf_pos, prev_base,
ref, alt, len,
info, genotypes, fout);
ref_pos += ref_allele_len;
ref += len;
alt += len;
}
}
