r273: cdna mapping mode for testing

Differences from the typical mapping mode:

* banded alignment disabled
* log gap cost during chaining
* zero long-gap extension during alignment
* up to 100kb (by default) reference gap
* bad seeding not filtered (to tune later)
This commit is contained in:
Heng Li
2017-08-08 11:31:49 -04:00
parent 4badf2fcbf
commit 1a7d782131
6 changed files with 33 additions and 15 deletions
+6 -3
View File
@@ -19,6 +19,7 @@ void mm_mapopt_init(mm_mapopt_t *opt)
opt->min_chain_score = 40;
opt->bw = 500;
opt->max_gap = 5000;
opt->max_gap_ref = -1;
opt->max_chain_skip = 25;
opt->mask_level = 0.5f;
@@ -167,7 +168,7 @@ void mm_pair_thin(mm_tbuf_t *b, int radius, mm_match_t *m1, mm_match_t *m2)
#endif
mm_reg1_t *mm_map_frag(const mm_mapopt_t *opt, const mm_idx_t *mi, mm_tbuf_t *b, uint32_t m_st, uint32_t m_en, const char *qname, int qlen, const char *seq, int *n_regs)
{
int i, n = m_en - m_st, j, n_u;
int i, n = m_en - m_st, j, n_u, max_gap_ref;
int64_t n_a;
uint64_t *u;
mm_match_t *m;
@@ -243,7 +244,8 @@ mm_reg1_t *mm_map_frag(const mm_mapopt_t *opt, const mm_idx_t *mi, mm_tbuf_t *b,
fprintf(stderr, "SD\t%s\t%d\t%c\t%d\t%d\t%d\n", mi->seq[a[i].x<<1>>33].name, (int32_t)a[i].x, "+-"[a[i].x>>63], (int32_t)a[i].y, (int32_t)(a[i].y>>32&0xff),
i == 0? 0 : ((int32_t)a[i].y - (int32_t)a[i-1].y) - ((int32_t)a[i].x - (int32_t)a[i-1].x));
n_u = mm_chain_dp(opt->max_gap, opt->bw, opt->max_chain_skip, opt->min_cnt, opt->min_chain_score, n_a, a, &u, b->km);
max_gap_ref = opt->max_gap_ref >= 0? opt->max_gap_ref : opt->max_gap;
n_u = mm_chain_dp(max_gap_ref, opt->max_gap, opt->bw, opt->max_chain_skip, opt->min_cnt, opt->min_chain_score, !!(opt->flag&MM_F_CDNA), n_a, a, &u, b->km);
regs = mm_gen_regs(b->km, qlen, n_u, u, a);
*n_regs = n_u;
@@ -256,7 +258,8 @@ mm_reg1_t *mm_map_frag(const mm_mapopt_t *opt, const mm_idx_t *mi, mm_tbuf_t *b,
if (!(opt->flag & MM_F_AVA)) { // don't choose primary mapping(s) for read overlap
mm_set_parent(b->km, opt->mask_level, *n_regs, regs);
mm_select_sub(b->km, opt->mask_level, opt->pri_ratio, mi->k*2, opt->best_n, n_regs, regs);
mm_join_long(b->km, opt, qlen, n_regs, regs, a); // TODO: this can be applied to all-vs-all in principle
if (!(opt->flag & MM_F_CDNA))
mm_join_long(b->km, opt, qlen, n_regs, regs, a); // TODO: this can be applied to all-vs-all in principle
}
if (opt->flag & MM_F_CIGAR) {
regs = mm_align_skeleton(b->km, opt, mi, qlen, seq, n_regs, regs, a); // this calls mm_filter_regs()