/**
* Gets a sorted string representation of the alleles of a variant.
*/
void bcf_alleles2string_sorted(bcf_hdr_t *h, bcf1_t *v, kstring_t *var)
{
bcf_unpack(v, BCF_UN_STR);
var->l = 0;
if (v->n_allele==2)
{
kputs(bcf_get_alt(v, 0), var);
kputc(',', var);
kputs(bcf_get_alt(v, 1), var);
}
else
{
char** allele = bcf_get_allele(v);
char** temp = (char**) malloc((bcf_get_n_allele(v)-1)*sizeof(char*));
for (int32_t i=1; i<v->n_allele; ++i)
{
temp[i-1] = allele[i];
}
std::qsort(temp, bcf_get_n_allele(v)-1, sizeof(char*), cmpstr);
kputs(bcf_get_alt(v, 0), var);
for (int32_t i=0; i<v->n_allele-1; ++i)
{
kputc(',', var);
kputs(temp[i], var);
}
free(temp);
}
}
/**
* Pick candidate motifs in different modes.
* Invokes motif tree and the candidate motifs are stored in a
* heap within the motif tree.
*/
void CandidateMotifPicker::generate_candidate_motifs(bcf_hdr_t* h, bcf1_t* v, Variant& variant)
{
if (debug)
{
std::cerr << "********************************************\n";
std::cerr << "PICK CANDIDATE MOTIFS\n\n";
}
if (variant.ins)
{
char** alleles = bcf_get_allele(v);
if (debug)
{
const char* repeat_tract = variant.vntr.exact_repeat_tract.c_str();
std::cerr << "Longest Allele : " << alleles[0][0] << "[" << &alleles[1][1] << "]" << &repeat_tract[1] << "\n";
}
//spike in inserted allele
std::string spiked_seq(alleles[1]);
std::string insertion = variant.vntr.exact_repeat_tract.substr(strlen(alleles[0]), variant.vntr.exact_repeat_tract.size()-strlen(alleles[0]));
spiked_seq.append(insertion);
mt->detect_candidate_motifs(spiked_seq);
indel_sequence.assign(&alleles[1][1]);
}
else
{
mt->detect_candidate_motifs(variant.vntr.exact_repeat_tract);
char** alleles = bcf_get_allele(v);
indel_sequence.assign(&alleles[0][1]);
}
if (debug)
{
std::cerr << "Indel : " << indel_sequence << "\n";
}
}
/**
* Pick candidate region.
*
* @mode - REFERENCE use refence field
* - ALLELE_EXACT by exact alignment
* - ALLELE_FUZZY by fuzzy alignment
*/
void CandidateRegionExtractor::pick_candidate_region(bcf_hdr_t* h, bcf1_t* v, Variant& variant, uint32_t mode)
{
if (mode==REFERENCE)
{
VNTR& vntr = variant.vntr;
vntr.exact_repeat_tract.assign(bcf_get_ref(v));
vntr.exact_rbeg1 = bcf_get_pos1(v);
char** alleles = bcf_get_allele(v);
vntr.exact_rend1 = strlen(alleles[0]);
vntr.fuzzy_rbeg1 = vntr.exact_rbeg1;
vntr.fuzzy_rend1 = vntr.exact_rend1;
}
else if (mode==EXACT_LEFT_RIGHT_ALIGNMENT)
{
extract_regions_by_exact_alignment(h, v, variant);
}
else if (mode==FUZZY_LEFT_RIGHT_ALIGNMENT)
{
extract_regions_by_fuzzy_alignment(h, v, variant);
}
}
/**
* Classifies variants.
*/
int32_t VariantManip::classify_variant(bcf_hdr_t *h, bcf1_t *v, Variant& var)
{
bcf_unpack(v, BCF_UN_STR);
const char* chrom = bcf_get_chrom(h, v);
uint32_t pos1 = bcf_get_pos1(v);
char** allele = bcf_get_allele(v);
int32_t n_allele = bcf_get_n_allele(v);
int32_t pos0 = pos1-1;
var.ts = 0;
var.tv = 0;
var.ins = 0;
var.del = 0;
var.clear(); // this sets the type to VT_REF by default.
bool homogeneous_length = true;
char* ref = allele[0];
int32_t rlen = strlen(ref);
//if only ref allele, skip this entire for loop
for (size_t i=1; i<n_allele; ++i)
{
int32_t type = VT_REF;
//check for tags
if (strchr(allele[i],'<'))
{
size_t len = strlen(allele[i]);
if (len>=5)
{
//VN/d+
if (allele[i][0]=='<' && allele[i][1]=='V' && allele[i][2]=='N' && allele[i][len-1]=='>' )
{
for (size_t j=3; j<len-1; ++j)
{
if (allele[i][j]<'0' || allele[i][j]>'9')
{
type = VT_VNTR;
}
}
}
//VNTR
else if (len==6 &&
allele[i][0]=='<' &&
allele[i][1]=='V' && allele[i][2]=='N' && allele[i][3]=='T' && allele[i][4]=='R' &&
allele[i][5]=='>' )
{
type = VT_VNTR;
}
//ST/d+
else if (allele[i][0]=='<' && allele[i][1]=='S' && allele[i][2]=='T' && allele[i][len-1]=='>' )
{
for (size_t j=3; j<len-1; ++j)
{
if (allele[i][j]<'0' || allele[i][j]>'9')
{
type = VT_VNTR;
}
}
}
//STR
else if (len==5 &&
allele[i][0]=='<' &&
allele[i][1]=='S' && allele[i][2]=='T' && allele[i][3]=='R' &&
allele[i][4]=='>' )
{
type = VT_VNTR;
}
}
if (type==VT_VNTR)
{
type = VT_VNTR;
var.type |= type;
var.alleles.push_back(Allele(type));
}
else
{
type = VT_SV;
var.type |= type;
std::string sv_type(allele[i]);
var.alleles.push_back(Allele(type, sv_type));
}
}
else if (allele[i][0]=='.' || strcmp(allele[i],allele[0])==0)
{
type = VT_REF;
}
else
{
kstring_t REF = {0,0,0};
kstring_t ALT = {0,0,0};
ref = allele[0];
char* alt = allele[i];
int32_t alen = strlen(alt);
if (rlen!=alen)
{
homogeneous_length = false;
}
//trimming
//this is required in particular for the
//characterization of multiallelics and
//in general, any unnormalized variant
int32_t rl = rlen;
int32_t al = alen;
//trim right
while (rl!=1 && al!=1)
{
if (ref[rl-1]==alt[al-1])
{
--rl;
--al;
}
else
{
break;
}
}
//trim left
while (rl !=1 && al!=1)
{
if (ref[0]==alt[0])
{
++ref;
++alt;
--rl;
--al;
}
else
{
break;
}
}
kputsn(ref, rl, &REF);
kputsn(alt, al, &ALT);
ref = REF.s;
alt = ALT.s;
int32_t mlen = std::min(rl, al);
int32_t dlen = al-rl;
int32_t diff = 0;
int32_t ts = 0;
int32_t tv = 0;
if (mlen==1 && dlen)
{
char ls, le, ss;
if (rl>al)
{
ls = ref[0];
le = ref[rl-1];
ss = alt[0];
}
else
{
ls = alt[0];
le = alt[al-1];
ss = ref[0];
}
if (ls!=ss && le!=ss)
{
++diff;
if ((ls=='G' && ss=='A') ||
(ls=='A' && ss=='G') ||
(ls=='C' && ss=='T') ||
(ls=='T' && ss=='C'))
{
++ts;
}
else
{
++tv;
}
}
}
else
{
for (int32_t j=0; j<mlen; ++j)
{
if (ref[j]!=alt[j])
{
++diff;
if ((ref[j]=='G' && alt[j]=='A') ||
(ref[j]=='A' && alt[j]=='G') ||
(ref[j]=='C' && alt[j]=='T') ||
(ref[j]=='T' && alt[j]=='C'))
{
++ts;
}
else
{
++tv;
}
}
}
}
//substitution variants
if (mlen==diff)
{
type |= mlen==1 ? VT_SNP : VT_MNP;
}
//indel variants
if (dlen)
{
type |= VT_INDEL;
}
//clumped SNPs and MNPs
if (diff && diff < mlen) //internal gaps
{
type |= VT_CLUMPED;
}
var.type |= type;
var.alleles.push_back(Allele(type, diff, alen, dlen, mlen, ts, tv));
var.ts += ts;
var.tv += tv;
var.ins = dlen>0?1:0;
var.del = dlen<0?1:0;
if (REF.m) free(REF.s);
if (ALT.m) free(ALT.s);
}
}
if (var.type==VT_VNTR)
{
bcf_unpack(v, BCF_UN_INFO);
//populate motif, motif len etc. etc.
// char* str = NULL;
// int32_t n = 0;
// int32_t ret = bcf_get_info_string(h, v, "MOTIF", &str, &n);
// if (ret>0)
// {
// var.motif = std::string(str);
// var.mlen = var.motif.size();
// }
// ret = bcf_get_info_string(h, v, "RU", &str, &n);
// if (ret>0)
// {
// var.ru = std::string(str);
// var.mlen = var.ru.size();
// }
// if (n) free(str);
//
// int32_t* no = NULL;
// n = 0;
// ret = bcf_get_info_int32(h, v, "RL", &no, &n);
// if (ret>0) var.rlen = *no;
// if (n) free(no);
//
// int32_t* fl = NULL;
// n = 0;
// ret = bcf_get_info_int32(h, v, "REF", &fl, &n);
// if (ret>0) var.rcn = *fl;
// if (n) free(fl);
}
//additionally define MNPs by length of all alleles
if (!(var.type&(VT_VNTR|VT_SV)) && var.type!=VT_REF)
{
if (homogeneous_length && rlen>1 && n_allele>1)
{
var.type |= VT_MNP;
}
}
return var.type;
}
/**
* Checks if a variant is normalized.
* Ignores if entry is not a variant.
*/
bool VariantManip::is_normalized(bcf1_t *v)
{
char** alleles = bcf_get_allele(v);
int32_t n_allele = bcf_get_n_allele(v);
if (n_allele==1) return true;
char first_base;
char last_base;
size_t rlen, alen, len;
bool exists_len_one_allele = false;
bool first_base_same = true;
bool last_base_same = true;
if (n_allele==2)
{
rlen = strlen(alleles[0]);
alen = strlen(alleles[1]);
if (rlen==1&&alen==1)
{
return true;
}
else
{
//check if variant is reference.
if (rlen==alen)
{
if (strcmp(alleles[0], alleles[1])==0)
{
return true;
}
}
//ref
if (rlen==1) exists_len_one_allele = true;
first_base = alleles[0][0];
last_base = alleles[0][rlen-1];
//alt
if (alen==1) exists_len_one_allele = true;
if (first_base!=alleles[1][0]) first_base_same = false;
if (last_base!=alleles[1][alen-1]) last_base_same = false;
if (last_base_same || (!exists_len_one_allele && first_base_same))
{
return false;
}
return true;
}
}
else
{
bool same = true;
for (size_t i=0; i<n_allele; ++i)
{
if (i)
{
len = strlen(alleles[i]);
if (len==1) exists_len_one_allele = true;
if (first_base!=alleles[i][0]) first_base_same = false;
if (last_base!=alleles[i][len-1]) last_base_same = false;
same = same && strcmp(alleles[i],alleles[0])==0;
}
else
{
len = strlen(alleles[0]);
if (len==1) exists_len_one_allele = true;
first_base = alleles[0][0];
last_base = alleles[0][len-1];
}
}
//reference entry
if (same)
{
return true;
}
if (last_base_same || (!exists_len_one_allele && first_base_same))
{
return false;
}
return true;
}
}
/**
* Updates VNTR related information from INFO fields.
*/
void Variant::update_vntr_from_info_fields()
{
vntr.motif = bcf_get_rid(v);
char** allele = bcf_get_allele(v);
// vntr.exact_repeat_tract.assign(allele[0]);
// std::string tags[16] = {"MOTIF", "RU", "BASIS", "MLEN", "BLEN", "REPEAT_TRACT", "COMP", "ENTROPY", "ENTROPY2", "KL_DIVERGENCE", "KL_DIVERGENCE2", "RL", "LL", "RU_COUNTS", "SCORE", "TRF_SCORE"};
vntr.motif = bcf_get_info_str(h, v, "MOTIF");
vntr.ru = bcf_get_info_str(h, v, "RU");
vntr.basis = bcf_get_info_str(h, v, "BASIS");
if (vntr.basis=="") vntr.basis = VNTR::get_basis(vntr.motif);
vntr.mlen = vntr.motif.size();
vntr.blen = (int32_t) vntr.basis.size();
std::vector<int32_t> i_vec = bcf_get_info_int_vec(h, v, "REPEAT_TRACT", 2, 0);
vntr.beg1 = i_vec[0];
vntr.end1 = i_vec[1];
i_vec = bcf_get_info_int_vec(h, v, "COMP", 4, 0);
vntr.comp[0] = i_vec[0];
vntr.comp[1] = i_vec[1];
vntr.comp[2] = i_vec[2];
vntr.comp[3] = i_vec[3];
vntr.entropy = bcf_get_info_flt(h, v, "ENTROPY");
vntr.entropy2 = bcf_get_info_flt(h, v, "ENTROPY2");
vntr.kl_divergence = bcf_get_info_flt(h, v, "KL_DIVERGENCE");
vntr.kl_divergence2 = bcf_get_info_flt(h, v, "KL_DIVERGENCE2");
vntr.rl = bcf_get_info_int(h, v, "RL");
vntr.ll = bcf_get_info_int(h, v, "LL");
i_vec = bcf_get_info_int_vec(h, v, "RU_COUNTS", 2, 0);
vntr.no_perfect_ru = i_vec[0];
vntr.no_ru = i_vec[1];
vntr.score = bcf_get_info_flt(h, v, "SCORE");
vntr.trf_score = bcf_get_info_int(h, v, "TRF_SCORE");
vntr.exact_motif = bcf_get_info_str(h, v, "EX_MOTIF");
vntr.exact_ru = bcf_get_info_str(h, v, "EX_RU");
vntr.exact_basis = bcf_get_info_str(h, v, "EX_BASIS");
vntr.exact_mlen = (int32_t) vntr.exact_motif.size();
vntr.exact_blen = (int32_t) vntr.exact_basis.size();
i_vec = bcf_get_info_int_vec(h, v, "EX_REPEAT_TRACT", 2, 0);
vntr.exact_beg1 = i_vec[0];
vntr.exact_end1 = i_vec[1];
i_vec = bcf_get_info_int_vec(h, v, "EX_COMP", 4, 0);
vntr.exact_comp[0] = i_vec[0];
vntr.exact_comp[1] = i_vec[1];
vntr.exact_comp[2] = i_vec[2];
vntr.exact_comp[3] = i_vec[3];
vntr.exact_entropy = bcf_get_info_flt(h, v, "EX_ENTROPY");
vntr.exact_entropy2 = bcf_get_info_flt(h, v, "EX_ENTROPY2");
vntr.exact_kl_divergence = bcf_get_info_flt(h, v, "EX_KL_DIVERGENCE");
vntr.exact_kl_divergence2 = bcf_get_info_flt(h, v, "EX_KL_DIVERGENCE2");
vntr.exact_rl = bcf_get_info_int(h, v, "EX_RL");
vntr.exact_ll = bcf_get_info_int(h, v, "EX_LL");
i_vec = bcf_get_info_int_vec(h, v, "EX_RU_COUNTS", 2, 0);
vntr.exact_no_perfect_ru = i_vec[0];
vntr.exact_no_ru = i_vec[1];
vntr.exact_score = bcf_get_info_flt(h, v, "EX_SCORE");
vntr.exact_trf_score = bcf_get_info_int(h, v, "EX_TRF_SCORE");
vntr.fuzzy_motif = bcf_get_info_str(h, v, "FZ_MOTIF");
vntr.fuzzy_ru = bcf_get_info_str(h, v, "FZ_RU");
vntr.fuzzy_basis = bcf_get_info_str(h, v, "FZ_BASIS");
vntr.fuzzy_mlen = (int32_t) vntr.fuzzy_motif.size();
vntr.fuzzy_blen = (int32_t) vntr.fuzzy_basis.size();
i_vec = bcf_get_info_int_vec(h, v, "FZ_REPEAT_TRACT", 2, 0);
vntr.fuzzy_beg1 = i_vec[0];
vntr.fuzzy_end1 = i_vec[1];
i_vec = bcf_get_info_int_vec(h, v, "FZ_COMP", 4, 0);
vntr.fuzzy_comp[0] = i_vec[0];
vntr.fuzzy_comp[1] = i_vec[1];
vntr.fuzzy_comp[2] = i_vec[2];
vntr.fuzzy_comp[3] = i_vec[3];
vntr.fuzzy_entropy = bcf_get_info_flt(h, v, "FZ_ENTROPY");
vntr.fuzzy_entropy2 = bcf_get_info_flt(h, v, "FZ_ENTROPY2");
vntr.fuzzy_kl_divergence = bcf_get_info_flt(h, v, "FZ_KL_DIVERGENCE");
vntr.fuzzy_kl_divergence2 = bcf_get_info_flt(h, v, "FZ_KL_DIVERGENCE2");
vntr.fuzzy_rl = bcf_get_info_int(h, v, "FZ_RL");
vntr.fuzzy_ll = bcf_get_info_int(h, v, "FZ_LL");
i_vec = bcf_get_info_int_vec(h, v, "FZ_RU_COUNTS", 2, 0);
vntr.fuzzy_no_perfect_ru = i_vec[0];
vntr.fuzzy_no_ru = i_vec[1];
vntr.fuzzy_score = bcf_get_info_flt(h, v, "FZ_SCORE");
vntr.fuzzy_trf_score = bcf_get_info_int(h, v, "FZ_TRF_SCORE");
}
/**
* Classifies variants.
*/
int32_t Variant::classify(bcf_hdr_t *h, bcf1_t *v)
{
clear();
this->h = h;
this->v = v;
bcf_unpack(v, BCF_UN_STR);
chrom.assign(bcf_get_chrom(h, v));
rid = bcf_get_rid(v);
pos1 = bcf_get_pos1(v);
end1 = bcf_get_end1(v);
char** allele = bcf_get_allele(v);
int32_t n_allele = bcf_get_n_allele(v);
int32_t pos0 = pos1-1;
bool homogeneous_length = true;
char* ref = allele[0];
int32_t rlen = strlen(ref);
if (strchr(ref, 'N'))
{
contains_N = true;
}
//if only ref allele, skip this entire for loop
for (size_t i=1; i<n_allele; ++i)
{
int32_t allele_type = VT_REF;
//check for symbolic alternative alleles
if (strchr(allele[i],'<'))
{
size_t len = strlen(allele[i]);
if (len>=5)
{
//VN/d+
if (allele[i][0]=='<' && allele[i][1]=='V' && allele[i][2]=='N' && allele[i][len-1]=='>' )
{
for (size_t j=3; j<len-1; ++j)
{
if (allele[i][j]<'0' || allele[i][j]>'9')
{
allele_type = VT_VNTR;
}
}
}
//VNTR
else if (len==6 &&
allele[i][0]=='<' &&
allele[i][1]=='V' && allele[i][2]=='N' && allele[i][3]=='T' && allele[i][4]=='R' &&
allele[i][5]=='>' )
{
allele_type = VT_VNTR;
}
//STR
else if (len==5 &&
allele[i][0]=='<' &&
allele[i][1]=='S' && allele[i][2]=='T' && allele[i][3]=='R' &&
allele[i][4]=='>' )
{
allele_type = VT_VNTR;
}
//ST/d+
else if (allele[i][0]=='<' && allele[i][1]=='S' && allele[i][2]=='T' && allele[i][len-1]=='>' )
{
type = VT_VNTR;
for (size_t j=3; j<len-1; ++j)
{
if ((allele[i][j]<'0' || allele[i][j]>'9') && allele[i][j]!='.')
{
type = VT_SV;
}
}
}
}
if (allele_type==VT_VNTR)
{
allele_type = VT_VNTR;
type |= allele_type;
alleles.push_back(Allele(allele_type));
}
else
{
allele_type = VT_SV;
type |= allele_type;
std::string sv_type(allele[i]);
alleles.push_back(Allele(allele_type, sv_type));
}
}
//checks for chromosomal breakpoints
else if (strchr(allele[i],'[')||strchr(allele[i],']'))
{
allele_type = VT_SV;
type |= allele_type;
std::string sv_type("<BND>");
alleles.push_back(Allele(allele_type, sv_type));
}
//non variant record
else if (allele[i][0]=='.' || strcmp(allele[i],allele[0])==0)
{
type = VT_REF;
}
//explicit sequence of bases
else
{
kstring_t REF = {0,0,0};
kstring_t ALT = {0,0,0};
ref = allele[0];
char* alt = allele[i];
int32_t alen = strlen(alt);
if (strchr(alt, 'N'))
{
contains_N = true;
}
if (rlen!=alen)
{
homogeneous_length = false;
}
//trimming
//this is required in particular for the
//characterization of multiallelics and
//in general, any unnormalized variant
int32_t rl = rlen;
int32_t al = alen;
//trim right
while (rl!=1 && al!=1)
{
if (ref[rl-1]==alt[al-1])
{
--rl;
--al;
}
else
{
break;
}
}
//trim left
while (rl !=1 && al!=1)
{
if (ref[0]==alt[0])
{
++ref;
++alt;
--rl;
--al;
}
else
{
break;
}
}
kputsn(ref, rl, &REF);
kputsn(alt, al, &ALT);
ref = REF.s;
alt = ALT.s;
int32_t mlen = std::min(rl, al);
int32_t dlen = al-rl;
int32_t diff = 0;
int32_t ts = 0;
int32_t tv = 0;
if (mlen==1 && dlen)
{
char ls, le, ss;
if (rl>al)
{
ls = ref[0];
le = ref[rl-1];
ss = alt[0];
}
else
{
ls = alt[0];
le = alt[al-1];
ss = ref[0];
}
if (ls!=ss && le!=ss)
{
++diff;
if ((ls=='G' && ss=='A') ||
(ls=='A' && ss=='G') ||
(ls=='C' && ss=='T') ||
(ls=='T' && ss=='C'))
{
++ts;
}
else
{
++tv;
}
}
}
else
{
for (int32_t j=0; j<mlen; ++j)
{
if (ref[j]!=alt[j])
{
++diff;
if ((ref[j]=='G' && alt[j]=='A') ||
(ref[j]=='A' && alt[j]=='G') ||
(ref[j]=='C' && alt[j]=='T') ||
(ref[j]=='T' && alt[j]=='C'))
{
++ts;
}
else
{
++tv;
}
}
}
}
//substitution variants
if (mlen==diff)
{
allele_type |= mlen==1 ? VT_SNP : VT_MNP;
}
//indel variants
if (dlen)
{
allele_type |= VT_INDEL;
}
//clumped SNPs and MNPs
if (diff && diff < mlen) //internal gaps
{
allele_type |= VT_CLUMPED;
}
type |= allele_type;
alleles.push_back(Allele(type, diff, alen, dlen, mlen, ts, tv));
ts += ts;
tv += tv;
ins = dlen>0?1:0;
del = dlen<0?1:0;
if (REF.m) free(REF.s);
if (ALT.m) free(ALT.s);
}
}
if (type==VT_VNTR)
{
update_vntr_from_info_fields(h, v);
}
//additionally define MNPs by length of all alleles
if (!(type&(VT_VNTR|VT_SV)) && type!=VT_REF)
{
if (homogeneous_length && rlen>1 && n_allele>1)
{
type |= VT_MNP;
}
}
return type;
}
/**
* Constructor.
* @v - VCF record.
*/
GenotypingRecord::GenotypingRecord(bcf_hdr_t *h, bcf1_t *v, int32_t vtype)
{
clear();
this->h = h;
this->v = v;
rid = bcf_get_rid(v);
pos1 = bcf_get_pos1(v);
this->vtype = vtype;
int32_t n_allele = bcf_get_n_allele(v);
if (vtype==VT_SNP && n_allele==2)
{
rid = bcf_get_rid(v);
beg1 = bcf_get_pos1(v);
end1 = beg1;
}
else if (vtype==VT_INDEL && bcf_get_n_allele(v)==2)
{
rid = bcf_get_rid(v);
char** alleles = bcf_get_allele(v);
dlen = strlen(alleles[1])-strlen(alleles[0]);
len = abs(dlen);
int32_t *flanks = NULL;
int32_t n = 0;
if (bcf_get_info_int32(h, v, "FLANKS", &flanks, &n)>0)
{
lend1 = flanks[0];
rbeg1 = flanks[1];
free(flanks);
}
else
{
lend1 = bcf_get_pos1(v) - 1;
rbeg1 = bcf_get_end_pos1(v) + 1;
}
int32_t *fuzzy_flanks = NULL;
n = 0;
if (bcf_get_info_int32(h, v, "FZ_FLANKS", &fuzzy_flanks, &n)>0)
{
fuzzy_lend1 = fuzzy_flanks[0];
fuzzy_rbeg1 = fuzzy_flanks[1];
free(fuzzy_flanks);
}
else
{
fuzzy_lend1 = bcf_get_pos1(v) - 1;
fuzzy_rbeg1 = bcf_get_end_pos1(v) + 1;
}
beg1 = std::min(lend1-2, fuzzy_lend1-2);
end1 = std::max(rbeg1+2, fuzzy_rbeg1+2);
//construct alleles
//get reference sequence
// char* ref_seq = NULL;
// int32_t ref_len = 0;
//// ref_seq = faidx_fetch_seq(fai, bcf_get_chrom(h,v), lend1+1-1, rbeg1-1-1, &ref_len);
//
// for (uint32_t i=0; i<n_allele; ++i)
// {
//
// }
// for ()
// {
// }
//
if (dlen>0)
{
indel.append(&alleles[1][1]);
}
else
{
indel.append(&alleles[0][1]);
}
}
else if (vtype==VT_VNTR)
{
rid = bcf_get_rid(v);
beg1 = bcf_get_pos1(v) - 1;
end1 = bcf_get_end_pos1(v) + 1;
char *motif = NULL;
int32_t n = 0;
if (bcf_get_info_string(h, v, "MOTIF", &motif, &n)>0)
{
this->motif.assign(motif);
free(motif);
}
}
}
