fork/minimap2
1
0
Fork 0
mirror of https://github.com/lh3/minimap2.git synced 2026-09-30 20:18:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into utec

Browse Source
This commit is contained in:
Heng Li
2021-07-16 13:32:47 -04:00
parent 28a37a017a 581f2d7123
commit b046052d82
39 changed files with 2143 additions and 531 deletions
Download Patch File Download Diff File Expand all files Collapse all files
map.c
+31 -58
View File
@@ -80,49 +80,7 @@ static void collect_minimizers(void *km, const mm_mapopt_t *opt, const mm_idx_t
#define heap_lt(a, b) ((a).x > (b).x)
KSORT_INIT(heap, mm128_t, heap_lt)
typedef struct {
uint32_t n;
uint32_t q_pos, q_span;
uint32_t seg_id:31, is_tandem:1;
const uint64_t *cr;
} mm_match_t;
static mm_match_t *collect_matches(void *km, int *_n_m, int max_occ, const mm_idx_t *mi, const mm128_v *mv, int64_t *n_a, int *rep_len, int *n_mini_pos, uint64_t **mini_pos)
{
int rep_st = 0, rep_en = 0, n_m;
size_t i;
mm_match_t *m;
*n_mini_pos = 0;
*mini_pos = (uint64_t*)kmalloc(km, mv->n * sizeof(uint64_t));
m = (mm_match_t*)kmalloc(km, mv->n * sizeof(mm_match_t));
for (i = 0, n_m = 0, *rep_len = 0, *n_a = 0; i < mv->n; ++i) {
const uint64_t *cr;
mm128_t *p = &mv->a[i];
uint32_t q_pos = (uint32_t)p->y, q_span = p->x & 0xff;
int t;
cr = mm_idx_get(mi, p->x>>8, &t);
if (t >= max_occ) {
int en = (q_pos >> 1) + 1, st = en - q_span;
if (st > rep_en) {
*rep_len += rep_en - rep_st;
rep_st = st, rep_en = en;
} else rep_en = en;
} else {
mm_match_t *q = &m[n_m++];
q->q_pos = q_pos, q->q_span = q_span, q->cr = cr, q->n = t, q->seg_id = p->y >> 32;
q->is_tandem = 0;
if (i > 0 && p->x>>8 == mv->a[i - 1].x>>8) q->is_tandem = 1;
if (i < mv->n - 1 && p->x>>8 == mv->a[i + 1].x>>8) q->is_tandem = 1;
*n_a += q->n;
(*mini_pos)[(*n_mini_pos)++] = (uint64_t)q_span<<32 | q_pos>>1;
}
}
*rep_len += rep_en - rep_st;
*_n_m = n_m;
return m;
}
static inline int skip_seed(int flag, uint64_t r, const mm_match_t *q, const char *qname, int qlen, const mm_idx_t *mi, int *is_self)
static inline int skip_seed(int flag, uint64_t r, const mm_seed_t *q, const char *qname, int qlen, const mm_idx_t *mi, int *is_self)
{
*is_self = 0;
if (qname && (flag & (MM_F_NO_DIAG|MM_F_NO_DUAL))) {
@@ -151,10 +109,10 @@ static mm128_t *collect_seed_hits_heap(void *km, const mm_mapopt_t *opt, int max
{
int i, n_m, heap_size = 0;
int64_t j, n_for = 0, n_rev = 0;
mm_match_t *m;
mm_seed_t *m;
mm128_t *a, *heap;
m = collect_matches(km, &n_m, max_occ, mi, mv, n_a, rep_len, n_mini_pos, mini_pos);
m = mm_collect_matches(km, &n_m, qlen, max_occ, opt->max_max_occ, opt->occ_dist, mi, mv, n_a, rep_len, n_mini_pos, mini_pos);
heap = (mm128_t*)kmalloc(km, n_m * sizeof(mm128_t));
a = (mm128_t*)kmalloc(km, *n_a * sizeof(mm128_t));
@@ -168,7 +126,7 @@ static mm128_t *collect_seed_hits_heap(void *km, const mm_mapopt_t *opt, int max
}
ks_heapmake_heap(heap_size, heap);
while (heap_size > 0) {
mm_match_t *q = &m[heap->y>>32];
mm_seed_t *q = &m[heap->y>>32];
mm128_t *p;
uint64_t r = heap->x;
int32_t is_self, rpos = (uint32_t)r >> 1;
@@ -216,12 +174,12 @@ static mm128_t *collect_seed_hits(void *km, const mm_mapopt_t *opt, int max_occ,
int *n_mini_pos, uint64_t **mini_pos)
{
int i, n_m;
mm_match_t *m;
mm_seed_t *m;
mm128_t *a;
m = collect_matches(km, &n_m, max_occ, mi, mv, n_a, rep_len, n_mini_pos, mini_pos);
m = mm_collect_matches(km, &n_m, qlen, max_occ, opt->max_max_occ, opt->occ_dist, mi, mv, n_a, rep_len, n_mini_pos, mini_pos);
a = (mm128_t*)kmalloc(km, *n_a * sizeof(mm128_t));
for (i = 0, *n_a = 0; i < n_m; ++i) {
mm_match_t *q = &m[i];
mm_seed_t *q = &m[i];
const uint64_t *r = q->cr;
uint32_t k;
for (k = 0; k < q->n; ++k) {
@@ -253,11 +211,9 @@ static mm128_t *collect_seed_hits(void *km, const mm_mapopt_t *opt, int max_occ,
static void chain_post(const mm_mapopt_t *opt, int max_chain_gap_ref, const mm_idx_t *mi, void *km, int qlen, int n_segs, const int *qlens, int *n_regs, mm_reg1_t *regs, mm128_t *a)
{
if (!(opt->flag & MM_F_ALL_CHAINS)) { // don't choose primary mapping(s)
mm_set_parent(km, opt->mask_level, *n_regs, regs, opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
mm_set_parent(km, opt->mask_level, opt->mask_len, *n_regs, regs, opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
if (n_segs <= 1) mm_select_sub(km, opt->pri_ratio, mi->k*2, opt->best_n, n_regs, regs);
else mm_select_sub_multi(km, opt->pri_ratio, 0.2f, 0.7f, max_chain_gap_ref, mi->k*2, opt->best_n, n_segs, qlens, n_regs, regs);
if (!(opt->flag & (MM_F_SPLICE|MM_F_SR|MM_F_NO_LJOIN))) // long join not working well without primary chains
mm_join_long(km, opt, qlen, n_regs, regs, a);
}
}
@@ -266,7 +222,7 @@ static mm_reg1_t *align_regs(const mm_mapopt_t *opt, const mm_idx_t *mi, void *k
if (!(opt->flag & MM_F_CIGAR)) return regs;
regs = mm_align_skeleton(km, opt, mi, qlen, seq, n_regs, regs, a); // this calls mm_filter_regs()
if (!(opt->flag & MM_F_ALL_CHAINS)) { // don't choose primary mapping(s)
mm_set_parent(km, opt->mask_level, *n_regs, regs, opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
mm_set_parent(km, opt->mask_level, opt->mask_len, *n_regs, regs, opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
mm_select_sub(km, opt->pri_ratio, mi->k*2, opt->best_n, n_regs, regs);
mm_set_sam_pri(*n_regs, regs);
}
@@ -317,9 +273,25 @@ void mm_map_frag(const mm_idx_t *mi, int n_segs, const int *qlens, const char **
if (max_chain_gap_ref < opt->max_gap) max_chain_gap_ref = opt->max_gap;
} else max_chain_gap_ref = opt->max_gap;
a = mm_chain_dp(max_chain_gap_ref, max_chain_gap_qry, opt->bw, opt->max_chain_skip, opt->max_chain_iter, opt->min_cnt, opt->min_chain_score, opt->chain_gap_scale, is_splice, n_segs, n_a, a, &n_regs0, &u, b->km);
if (opt->flag & MM_F_RMQ) {
a = mg_lchain_rmq(opt->max_gap, opt->rmq_inner_dist, opt->bw, opt->max_chain_skip, opt->rmq_size_cap, opt->min_cnt, opt->min_chain_score,
opt->chain_gap_scale * 0.01 * mi->k, 0.0f, n_a, a, &n_regs0, &u, b->km);
} else {
a = mg_lchain_dp(max_chain_gap_ref, max_chain_gap_qry, opt->bw, opt->max_chain_skip, opt->max_chain_iter, opt->min_cnt, opt->min_chain_score,
opt->chain_gap_scale * 0.01 * mi->k, 0.0f, is_splice, n_segs, n_a, a, &n_regs0, &u, b->km);
}
if (opt->max_occ > opt->mid_occ && rep_len > 0) {
if (opt->bw_long > opt->bw && (opt->flag & (MM_F_SPLICE|MM_F_SR|MM_F_NO_LJOIN)) == 0 && n_segs == 1 && n_regs0 > 1) { // re-chain/long-join for long sequences
int32_t st = (int32_t)a[0].y, en = (int32_t)a[(int32_t)u[0] - 1].y;
if (qlen_sum - (en - st) > opt->rmq_rescue_size || en - st > qlen_sum * opt->rmq_rescue_ratio) {
int32_t i;
for (i = 0, n_a = 0; i < n_regs0; ++i) n_a += (int32_t)u[i];
kfree(b->km, u);
radix_sort_128x(a, a + n_a);
a = mg_lchain_rmq(opt->max_gap, opt->rmq_inner_dist, opt->bw_long, opt->max_chain_skip, opt->rmq_size_cap, opt->min_cnt, opt->min_chain_score,
opt->chain_gap_scale * 0.01 * mi->k, 0.0f, n_a, a, &n_regs0, &u, b->km);
}
} else if (opt->max_occ > opt->mid_occ && rep_len > 0 && !(opt->flag & MM_F_RMQ)) { // re-chain, mostly for short reads
int rechain = 0;
if (n_regs0 > 0) { // test if the best chain has all the segments
int n_chained_segs = 1, max = 0, max_i = -1, max_off = -1, off = 0;
@@ -339,7 +311,8 @@ void mm_map_frag(const mm_idx_t *mi, int n_segs, const int *qlens, const char **
kfree(b->km, mini_pos);
if (opt->flag & MM_F_HEAP_SORT) a = collect_seed_hits_heap(b->km, opt, opt->max_occ, mi, qname, &mv, qlen_sum, &n_a, &rep_len, &n_mini_pos, &mini_pos);
else a = collect_seed_hits(b->km, opt, opt->max_occ, mi, qname, &mv, qlen_sum, &n_a, &rep_len, &n_mini_pos, &mini_pos);
a = mm_chain_dp(max_chain_gap_ref, max_chain_gap_qry, opt->bw, opt->max_chain_skip, opt->max_chain_iter, opt->min_cnt, opt->min_chain_score, opt->chain_gap_scale, is_splice, n_segs, n_a, a, &n_regs0, &u, b->km);
a = mg_lchain_dp(max_chain_gap_ref, max_chain_gap_qry, opt->bw, opt->max_chain_skip, opt->max_chain_iter, opt->min_cnt, opt->min_chain_score,
opt->chain_gap_scale * 0.01 * mi->k, 0.0f, is_splice, n_segs, n_a, a, &n_regs0, &u, b->km);
}
}
b->frag_gap = max_chain_gap_ref;
@@ -370,7 +343,7 @@ void mm_map_frag(const mm_idx_t *mi, int n_segs, const int *qlens, const char **
seg = mm_seg_gen(b->km, hash, n_segs, qlens, n_regs0, regs0, n_regs, regs, a); // split fragment chain to separate segment chains
free(regs0);
for (i = 0; i < n_segs; ++i) {
mm_set_parent(b->km, opt->mask_level, n_regs[i], regs[i], opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop); // update mm_reg1_t::parent
mm_set_parent(b->km, opt->mask_level, opt->mask_len, n_regs[i], regs[i], opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop); // update mm_reg1_t::parent
regs[i] = align_regs(opt, mi, b->km, qlens[i], seqs[i], &n_regs[i], regs[i], seg[i].a);
mm_set_mapq(b->km, n_regs[i], regs[i], opt->min_chain_score, opt->a, rep_len, is_sr);
}
@@ -514,7 +487,7 @@ static void merge_hits(step_t *s)
}
}
mm_hit_sort(km, &s->n_reg[k], s->reg[k], opt->alt_drop);
mm_set_parent(km, opt->mask_level, s->n_reg[k], s->reg[k], opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
mm_set_parent(km, opt->mask_level, opt->mask_len, s->n_reg[k], s->reg[k], opt->a * 2 + opt->b, opt->flag&MM_F_HARD_MLEVEL, opt->alt_drop);
if (!(opt->flag & MM_F_ALL_CHAINS)) {
mm_select_sub(km, opt->pri_ratio, s->p->mi->k*2, opt->best_n, &s->n_reg[k], s->reg[k]);
mm_set_sam_pri(s->n_reg[k], s->reg[k]);