added --mask-len for #659

hit.c

@@ -122,7 +122,7 @@ void mm_split_reg(mm_reg1_t *r, mm_reg1_t *r2, int n, int qlen, mm128_t *a)
 	r->split |= 1, r2->split |= 2;
 }
 
-void mm_set_parent(void *km, float mask_level, int n, mm_reg1_t *r, int sub_diff, int hard_mask_level, float alt_diff_frac) // and compute mm_reg1_t::subsc
+void mm_set_parent(void *km, float mask_level, int mask_len, int n, mm_reg1_t *r, int sub_diff, int hard_mask_level, float alt_diff_frac) // and compute mm_reg1_t::subsc
 {
 	int i, j, k, *w;
 	uint64_t *cov;
@@ -162,7 +162,7 @@ skip_uncov:
 			min = ej - sj < ei - si? ej - sj : ei - si;
 			max = ej - sj > ei - si? ej - sj : ei - si;
 			ol = si < sj? (ei < sj? 0 : ei < ej? ei - sj : ej - sj) : (ej < si? 0 : ej < ei? ej - si : ei - si); // overlap length; TODO: this can be simplified
-			if ((float)ol / min - (float)uncov_len / max > mask_level) {
+			if ((float)ol / min - (float)uncov_len / max > mask_level && uncov_len <= mask_len) { // then this is a secondary hit
 				int cnt_sub = 0, sci = ri->score;
 				ri->parent = rp->parent;
 				if (!rp->is_alt && ri->is_alt) sci = mm_alt_score(sci, alt_diff_frac);

main.c

@@ -70,6 +70,7 @@ static ko_longopt_t long_options[] = {
 	{ "chain-gap-scale",ko_required_argument, 343 },
 	{ "alt",            ko_required_argument, 344 },
 	{ "alt-drop",       ko_required_argument, 345 },
+	{ "mask-len",       ko_required_argument, 346 },
 	{ "help",           ko_no_argument,       'h' },
 	{ "max-intron-len", ko_required_argument, 'G' },
 	{ "version",        ko_no_argument,       'V' },
@@ -217,6 +218,7 @@ int main(int argc, char *argv[])
 		else if (c == 343) opt.chain_gap_scale = atof(o.arg); // --chain-gap-scale
 		else if (c == 344) alt_list = o.arg; // --alt
 		else if (c == 345) opt.alt_drop = atof(o.arg); // --alt-drop
+		else if (c == 346) opt.mask_len = mm_parse_num(o.arg); // --mask-len
 		else if (c == 314) { // --frag
 			yes_or_no(&opt, MM_F_FRAG_MODE, o.longidx, o.arg, 1);
 		} else if (c == 315) { // --secondary

map.c

@@ -249,7 +249,7 @@ 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
@@ -262,7 +262,7 @@ 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);
 	}
@@ -365,7 +365,7 @@ 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);
 		}
@@ -509,7 +509,7 @@ 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]);

minimap.h

@@ -122,6 +122,7 @@ typedef struct {
 	float chain_gap_scale;
 
 	float mask_level;
+	int mask_len;
 	float pri_ratio;
 	int best_n;      // top best_n chains are subjected to DP alignment
 

minimap2.1

@@ -237,7 +237,14 @@ or more of the shorter chain [0.5]
 .B --hard-mask-level
 Honor option
 .B -M
-and disable a heurstic to save unmapped subsequences.
+and disable a heurstic to save unmapped subsequences and disables
+.BR --mask-len .
+.TP
+.BI --mask-len \ NUM
+Keep an alignment if dropping it leaves an unaligned region on query longer than
+.IR INT
+[inf]. Effective without
+.BR --hard-mask-level .
 .TP
 .BI --max-chain-skip \ INT
 A heuristics that stops chaining early [25]. Minimap2 uses dynamic programming

mmpriv.h

@@ -71,7 +71,7 @@ void mm_split_reg(mm_reg1_t *r, mm_reg1_t *r2, int n, int qlen, mm128_t *a);
 void mm_sync_regs(void *km, int n_regs, mm_reg1_t *regs);
 int mm_squeeze_a(void *km, int n_regs, mm_reg1_t *regs, mm128_t *a);
 int mm_set_sam_pri(int n, mm_reg1_t *r);
-void mm_set_parent(void *km, float mask_level, int n, mm_reg1_t *r, int sub_diff, int hard_mask_level, float alt_diff_frac);
+void mm_set_parent(void *km, float mask_level, int mask_len, int n, mm_reg1_t *r, int sub_diff, int hard_mask_level, float alt_diff_frac);
 void mm_select_sub(void *km, float pri_ratio, int min_diff, int best_n, int *n_, mm_reg1_t *r);
 void mm_select_sub_multi(void *km, float pri_ratio, float pri1, float pri2, int max_gap_ref, int min_diff, int best_n, int n_segs, const int *qlens, int *n_, mm_reg1_t *r);
 void mm_filter_regs(const mm_mapopt_t *opt, int qlen, int *n_regs, mm_reg1_t *regs);

options.c

@@ -1,4 +1,5 @@
 #include <stdio.h>
+#include <limits.h>
 #include "mmpriv.h"
 
 void mm_idxopt_init(mm_idxopt_t *opt)
@@ -27,6 +28,7 @@ void mm_mapopt_init(mm_mapopt_t *opt)
 	opt->chain_gap_scale = 1.0f;
 
 	opt->mask_level = 0.5f;
+	opt->mask_len = INT_MAX;
 	opt->pri_ratio = 0.8f;
 	opt->best_n = 5;
 

python/cmappy.pxd

@@ -22,6 +22,7 @@ cdef extern from "minimap.h":
 		int min_chain_score
 		float chain_gap_scale
 		float mask_level
+		int mask_len
 		float pri_ratio
 		int best_n
 		int max_join_long, max_join_short