xxay/hifiasm
1
0
Fork 0
mirror of https://github.com/chhylp123/hifiasm.git synced 2026-09-15 12:47:57 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
37b07e4d3344fd8b7f5b696d947f4fd865d1c20e
hifiasm/sketch.cpp
chhylp123 37b07e4d33 better seeding
2021-08-22 09:53:34 -04:00

839 lines
30 KiB
C++
Raw Blame History

#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
#include <string.h>
#include "kvec.h"
#include "htab.h"
#include "ksort.h"
#include "Correct.h"
#define MAX_HIGH_OCC 8 // TODO: don't hard code if we need to tune this parameter
#define MAX_MAX_HIGH_OCC 16
#define GMC(a, x,y,xn) ((a)[(x)*(xn)+(y)])
static inline int mzcmp(const ha_mz1_t *a, const ha_mz1_t *b)
{
return a->rid < b->rid? -1 : a->rid > b->rid? 1 : ((a->x > b->x) - (a->x < b->x));
}
#define mz_lt(a, b) (mzcmp(&(a), &(b)) < 0)
KSORT_INIT(mz, ha_mz1_t, mz_lt)
void debug_refine(ha_mz1_t *ma, uint64_t *mmt, int32_t sn, int32_t n, int32_t m, int32_t end)
{
uint64_t ks = end;
int64_t t = 0, i, k, sp = -1, ep = -1, ovlp, tot = mmt[end]&0xffffffff, nt = 0;;
while (ks != 0xffffffff)
{
i = ks/m; k = ks%m;
ks = mmt[ks]>>32;
if(ks == 0xffffffff || (int32_t)(ks/m) == (i-1))
{
t++;
ovlp = ((MIN(ep, (int64_t)ma[k].pos) >= MAX(sp, (int64_t)(ma[k].pos+1-ma[k].span)))?
MIN(ep, (int64_t)ma[k].pos) - MAX(sp, (int64_t)(ma[k].pos+1-ma[k].span)) + 1:0);
if(ovlp != 0) fprintf(stderr, "ERROR-OVLP\n");
if(sp == -1 || sp > (ma[k].pos+1-ma[k].span)) sp = ma[k].pos+1-ma[k].span;
if(ep == -1 || ep < ma[k].pos) ep = ma[k].pos;
nt += (ma[k].rid);
}
}
if(t != sn) fprintf(stderr, "ERROR-TN, t: %ld, sn: %d\n", t, sn);
if(nt != tot) fprintf(stderr, "ERROR-TOT, nt: %ld, tot: %ld\n", nt, tot);
}
void refine_select(ha_mz1_v *mz, int32_t sidx, int32_t eidx, int32_t sn, int32_t min_freq, st_mt_t *mm,
int32_t *rsi, int32_t *rei)
{
int32_t n = sn, m = eidx + 1 - sidx, i, k, t, mk=-1;
uint64_t ix, kx, ks;
kv_resize(uint64_t, *mm, mm->n+n*m);
ha_mz1_t *ma = mz->a + sidx;
uint64_t *mmt = mm->a + mm->n;
// fprintf(stderr, "[M::%s::] ==> +n: %d, m: %d, sn: %d, sidx: %d, eidx: %d\n", __func__, n, m, sn, sidx, eidx);
for (i = 0; i < n; i++) ///how many selected minimizers
{
for (k = 0, mk = -1; k < m; k++) ///how many minimizers in total
{
if((int32_t)(ma[k].rid)<min_freq) continue;
ks = ma[k].pos + 1 - ma[k].span; t = -1;
if(i > 0)
{
for (t = k-1; t >= 0 && (ma[t].pos >= ks||(int32_t)(ma[t].rid)<min_freq); t--);
}
ix = (i <= 0?0:(t<0?0xffffffff:(GMC(mmt, i-1,t,m)&0xffffffff)));
if(ix < 0xffffffff) ix += (ma[k].rid);
kx = (mk < 0?0xffffffff:(GMC(mmt, i, mk,m)&0xffffffff));
ks = MIN(ix, kx);
// fprintf(stderr, "ks: %lu, i: %d (n-%d), k: %d (m-%d), ix: %lu, kx: %lu, t: %d, mk: %d\n",
// ks, i, n, k, m, ix, kx, t, mk);
if((ks&0xffffffff) == 0xffffffff) ks |= ((uint64_t)0xffffffff)<<32;
else if(ks == ix) ks |= (uint64_t)(i>0?(i-1)*m+t:0xffffffff)<<32;
else if(ks == kx) ks |= (uint64_t)(mk>=0?i*m+mk:0xffffffff)<<32;
GMC(mmt, i,k,m) = ks;
mk = k;
}
}
// fprintf(stderr, "[M::%s::] ==> ++n: %d, m: %d, sn: %d, sidx: %d, eidx: %d\n", __func__, n, m, sn, sidx, eidx);
ks = (n-1)*m + mk; ix = (uint64_t)-1; kx = 0;
while (ks != 0xffffffff)
{
i = ks/m; k = ks%m;
ks = mmt[ks]>>32;
// fprintf(stderr, "i: %d, k: %d, ks: %lu\n", i, k, ks);
if(ks == 0xffffffff || (int32_t)(ks/m) == (i-1))
{
mm->a[sidx+k] = 1;
ix = MIN((uint64_t)k, ix); kx = MAX((uint64_t)k, kx);
}
}
///debug
// debug_refine(ma, mmt, sn, n, m, (n-1)*m + mk);
if(rsi) (*rsi) = ix + sidx;
if(rei) (*rei) = kx + sidx;
}
void refine_sketch(ha_mz1_v *p, ha_pt_t *pt, int32_t rlen, int32_t dp_min_len, float er, int32_t min_freq, st_mt_t *mt)
{
// fprintf(stderr, "[M::%s::] ==> #########10#########, rlen: %d\n", __func__, rlen);
int32_t i, n = p->n, bd, len = MIN(rlen, dp_min_len), sublen, cnt, ei, li, ri;
int32_t sn = len*er + 1;
kv_resize(uint64_t, *mt, (int64_t)p->n);
mt->n = p->n; memset(mt->a, 0, sizeof(uint64_t)*p->n);
for (i = 0; i < n; i++) p->a[i].rid = ha_pt_cnt(pt, p->a[i].x);
for (i = cnt = 0, bd = -1, ei = -1; i < n; i++)
{
if((int32_t)(p->a[i].rid)<min_freq) continue;
sublen = p->a[i].pos + 1;
if(sublen > len) break;
else ei = i;
if((int32_t)(p->a[i].pos + 1 - p->a[i].span) > bd)
{
bd = p->a[i].pos;
cnt++;
}
}
// fprintf(stderr, "[M::%s::] ==> +cnt: %d, sn: %d, ei: %d, n: %d\n", __func__, cnt, sn, ei, n);
if(cnt >= sn) refine_select(p, 0, ei, sn, min_freq, mt, NULL, &li);
else
{
li = i-1;
for (i = 0; i <= li; i++) mt->a[i] = 1;
}
if(len < rlen)
{
for (i = n-1, cnt = 0, bd = rlen+1, ei = -1; i >= 0; i--)
{
if((int32_t)(p->a[i].rid)<min_freq) continue;
sublen = rlen - (p->a[i].pos + 1 - p->a[i].span);
if(sublen > len) break;
else ei = i;
if((int32_t)(p->a[i].pos) < bd)
{
bd = p->a[i].pos + 1 - p->a[i].span;
cnt++;
}
}
// fprintf(stderr, "[M::%s::] ==> -cnt: %d, sn: %d, ei: %d, n: %d\n", __func__, cnt, sn, ei, n);
if(cnt >= sn) refine_select(p, ei, n-1, sn, min_freq, mt, &ri, NULL);
else
{
ri = i+1;
for (i = ri; i <= n-1; i++) mt->a[i] = 1;
}
// fprintf(stderr, "[M::%s::] ==> --cnt: %d, sn: %d, ei: %d, n: %d\n", __func__, cnt, sn, ei, n);
if(ri - li >= 2)
{
li++; ri--;
sn = (p->a[ri].pos - p->a[li].pos + p->a[li].span)*er + 1;
for (i = li, cnt = 0, bd = -1; i <= ri; i++)
{
if((int32_t)(p->a[i].rid)<min_freq) continue;
if((int32_t)(p->a[i].pos + 1 - p->a[i].span) > bd)
{
bd = p->a[i].pos;
cnt++;
if(cnt >= sn) break;
}
}
if(cnt >= sn) refine_select(p, li, ri, sn, min_freq, mt, NULL, NULL);
else for (i = li; i <= ri; i++) mt->a[i] = 1;
}
}
// fprintf(stderr, "[M::%s::] ==> #########20#########, p->n: %u, n: %d\n", __func__, p->n, n);
for (i = sn = 0; i < n; i++)
{
if(mt->a[i])
{
p->a[sn] = p->a[i];
sn++;
}
}
// if(p->n != sn) fprintf(stderr, "[M::%s::] ==> #########21#########, p->n: %u, sn: %d\n", __func__, p->n, sn);
p->n = sn;
}
inline int hf_dp(ha_mz1_v *mz, int32_t sidx, int32_t eidx, int32_t sn, int32_t min_freq, st_mt_t *mm,
int32_t *rsi, int32_t *rei)
{
return 0;
}
inline void hf_select(ha_mz1_v *p, int32_t si, int32_t ei, int32_t n, int32_t len, int32_t sample_dist, ha_mz1_t *b, int32_t force)
{
if(ei - si <= 1) return;
int32_t ps = si < 0? 0 : p->a[si].pos;
int32_t pe = ei == n? len : p->a[ei].pos;
int32_t j, k, st = si + 1, en = ei;
int32_t max_high_occ = (int32_t)((double)(pe - ps) / sample_dist + .499);
if (max_high_occ > MAX_MAX_HIGH_OCC)
max_high_occ = MAX_MAX_HIGH_OCC;
for (j = st, k = 0; j < en && k < max_high_occ; ++j, ++k)
b[k] = p->a[j], b[k].pos = j; // b[].pos keeps the index in p->a[]
ks_heapmake_mz(k, b); // initialize the binomial heap
for (; j < en; ++j) { // if there are more, choose top max_high_occ
if (mz_lt(p->a[j], b[0])) { // then update the heap
b[0] = p->a[j], b[0].pos = j;
ks_heapdown_mz(0, k, b);
}
}
//ks_heapsort_mz(k, b); // sorting is not needed for now
for (j = 0; j < k; ++j)
if (b[j].rid < pe - ps || force)
p->a[b[j].pos].rid = 0;
}
void select_mz(ha_mz1_v *p, int len, int sample_dist, int32_t dp_min_len)
{ // for high-occ minimizers, choose up to max_high_occ in each high-occ streak
int32_t i, last0 = -1, n = (int32_t)p->n, m = 0, nw[2], min_len;
ha_mz1_t b[MAX_MAX_HIGH_OCC]; // this is to avoid a heap allocation
if (n == 0 || n == 1) return;
assert(n < 1<<27); // 27 is the number of bits for ha_mz1_t::pos; this should be safe as there are more bases than minimizers
for (i = 0; i < n; ++i)
if (p->a[i].rid != 0) ++m;
if (m == 0) return; // no high-frequency k-mers; do nothing
for (i = 0; i <= n; ++i) {
if (i == n || p->a[i].rid == 0) {
if (i - last0 > 1) {
hf_select(p, last0, i, n, len, sample_dist, b, 0);
// int32_t ps = last0 < 0? 0 : p->a[last0].pos;
// int32_t pe = i == n? len : p->a[i].pos;
// int32_t j, k, st = last0 + 1, en = i;
// int32_t max_high_occ = (int32_t)((double)(pe - ps) / sample_dist + .499);
// if (max_high_occ > MAX_MAX_HIGH_OCC)
// max_high_occ = MAX_MAX_HIGH_OCC;
// for (j = st, k = 0; j < en && k < max_high_occ; ++j, ++k)
// b[k] = p->a[j], b[k].pos = j; // b[].pos keeps the index in p->a[]
// ks_heapmake_mz(k, b); // initialize the binomial heap
// for (; j < en; ++j) { // if there are more, choose top max_high_occ
// if (mz_lt(p->a[j], b[0])) { // then update the heap
// b[0] = p->a[j], b[0].pos = j;
// ks_heapdown_mz(0, k, b);
// }
// }
// //ks_heapsort_mz(k, b); // sorting is not needed for now
// for (j = 0; j < k; ++j)
// if (b[j].rid < pe - ps)
// p->a[b[j].pos].rid = 0;
}
last0 = i;
}
}
min_len = MAX(dp_min_len, (p->a[0].pos+1)+sample_dist);
for (i = 0, nw[0] = nw[1] = 0; i < n; i++)
{
nw[(p->a[i].rid!=0)]++;
if((p->a[i].pos + 1) > min_len) break;
}
if(nw[0]==0 && nw[1]>0) hf_select(p, -1, i, n, len, sample_dist, b, 1);
min_len = MAX(dp_min_len, (len - (p->a[n-1].pos + 1 - p->a[n-1].span))+sample_dist);
for (i = n-1, nw[0] = nw[1] = 0; i >= 0; i--)
{
nw[(p->a[i].rid!=0)]++;
if((len - (p->a[i].pos + 1 - p->a[i].span)) > min_len) break;
}
if(nw[0]==0 && nw[1]>0) hf_select(p, i, n, n, len, sample_dist, b, 1);
for (i = n = 0; i < (int32_t)p->n; ++i) // squeeze out filtered minimizers
if (p->a[i].rid == 0)
p->a[n++] = p->a[i];
p->n = n;
}
static inline int mzcmp_l(const ha_mz1_v *p, int32_t ai, int32_t bi)
{
if(ai >= 0 && bi >= 0){
ha_mz1_t *a = &(p->a[ai]), *b = &(p->a[bi]);
if(a->rid > 0 && b->rid > 0) return mzcmp(a, b);
return (a->rid == 0) - (b->rid == 0);
}
return (ai < 0) - (bi < 0);
}
#define GL(x, i) ((int64_t)((uint32_t)((x).a[(i)])))
#define A_M(p, i) ((i) >= 0 && (p).a[(i)].rid > 0)
int32_t qfw(ha_mz1_v *p, st_mt_t *mt, int32_t n, int32_t tot_l, int32_t ws, int32_t i, int32_t *mi)
{
int32_t m, si;
for (si = i, (*mi) = -1; i < n; i++){
if(GL(*mt, i) >= ws || (i+1 < n && GL(*mt, i) < ws && GL(*mt, i+1) > ws) ||
(i+1 == n && tot_l >= ws && GL(*mt, i) < ws)){
for (m = si; m <= i; m++){
if(!A_M(*p, m)) continue;
if(mzcmp_l(p, *mi, m) >= 0) (*mi) = m;
}
if((*mi) >= 0 && A_M(*p, *mi)){
for (m = si; m <= i; m++){
if(!A_M(*p, m)) continue;
if(mzcmp_l(p, *mi, m) == 0) mt->a[m] |= 0x100000000;
}
}
break;
}
}
return i;
}
void dbg_boundary(ha_mz1_v *p, st_mt_t *mt, int32_t w, int32_t k, int32_t tot_l)
{
if(tot_l < w + k -1) return;
int32_t i, m, n = p->n, s, a;
for (i = 0; i < n; i++){
if(GL(*mt, i) >= w+k-1){
for (m = s = a = 0; m <= i; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) <= w+k-1){
a++;
if(mt->a[m]&0x100000000) s++;
}
}
if(a > 0 && s == 0){
fprintf(stderr, "\nERROR1, s: %d, n: %d, tot_l: %d, end_l: %ld\n", s, n, tot_l, GL(*mt, i));
for (m = s = a = 0; m <= i; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) <= w+k-1){
fprintf(stderr, "lp: %ld\n", GL(*mt, m));
a++;
if(mt->a[m]&0x100000000) s++;
}
}
}
break;
}
}
if(i == n){
for (m = s = a = 0; m < n; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) <= w+k-1){
a++;
if(mt->a[m]&0x100000000) s++;
}
}
if(a > 0 && s == 0) fprintf(stderr, "ERROR2\n");
}
for (i = n-1; i >= 0; i--)
{
if (GL(*mt, i) + w <= tot_l + 1) {
for (m = i, s = a = 0; m < n; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) + w >= tot_l + 1){
a++;
if(mt->a[m]&0x100000000) s++;
}
}
if(a > 0 && s == 0) {
fprintf(stderr, "\nERROR3, s: %d, n: %d, tot_l: %d, end_l: %ld\n", s, n, tot_l, GL(*mt, i));
for (m = i, s = a = 0; m < n; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) + w >= tot_l + 1){
fprintf(stderr, "lp: %ld\n", GL(*mt, m));
a++;
if(mt->a[m]&0x100000000) s++;
}
}
}
break;
}
}
if(i < 0){
for (m = s = a = 0; m < n; m++){
if(!A_M(*p, m)) continue;
if(GL(*mt, m) + w >= tot_l + 1){
a++;
if(mt->a[m]&0x100000000) s++;
}
}
if(a > 0 && s == 0) fprintf(stderr, "ERROR4\n");
}
}
static void select_mz_h(ha_mz1_v *p, st_mt_t *mt, int len, int sample_dist, int32_t w, int32_t k, int32_t tot_l)
{ // for high-occ minimizers, choose up to max_high_occ in each high-occ streak
int32_t i, mi = -1, si, last0 = -1, n = (int32_t)p->n, m = 0, ws = w + k - 1;
if (n == 0) return;
assert(n < 1<<27); // 27 is the number of bits for ha_mz1_t::pos; this should be safe as there are more bases than minimizers
for (i = m = 0, last0 = -1; i <= n; ++i) {
if (i == n || p->a[i].rid == 0) {
if (i - last0 > 1) {
int32_t ps = last0 < 0? 0 : p->a[last0].pos;
int32_t pe = i == n? len : p->a[i].pos;
if(((int32_t)((double)(pe - ps) / sample_dist + .499)) > 0){
last0 = -2;
m++;
break;
}
}
last0 = i;
}
}
if (m == 0) return; // no high-frequency k-mers; do nothing
if(last0 >= -1) goto ff;
i = 0;
i = qfw(p, mt, n, tot_l, ws, i, &mi);
if(i == n) goto ff;
for (si = 0, i++; i < n; i++){
for (; si < i; si++){
if(GL(*mt, si) + w > GL(*mt, i)) break;
}
// a new minimum; then write the old min
if(mzcmp_l(p, i, mi) <= 0) {
if(A_M(*p, mi)) mt->a[mi] |= 0x100000000;
mi = i;
}// old min has moved outside the window
else if(si > mi){
if(A_M(*p, mi)) mt->a[mi] |= 0x100000000;
for (m = si, mi = -1; m <= i; m++){
if(mzcmp_l(p, mi, m) >= 0) mi = m;
}
if(A_M(*p, mi)){
for (m = si; m <= i; m++){
if(!A_M(*p, m)) continue;
if(mzcmp_l(p, mi, m) == 0) mt->a[m] |= 0x100000000;
}
}
}
}
if(A_M(*p, mi)) mt->a[mi] |= 0x100000000;
for (i = n - 1; si < n && GL(*mt, si) + w <= tot_l + 1; si++){
if(si > mi){
if(A_M(*p, mi)) mt->a[mi] |= 0x100000000;
for (m = si, mi = -1; m <= i; m++){
if(mzcmp_l(p, mi, m) >= 0) mi = m;
}
if(A_M(*p, mi)){
for (m = si; m <= i; m++){
if(!A_M(*p, m)) continue;
if(mzcmp_l(p, mi, m) == 0) mt->a[m] |= 0x100000000;
}
}
}
}
/**
dbg_boundary(p, mt, w, k, tot_l);
fprintf(stderr, "\n");
for (i = 0; i < (int32_t)p->n; ++i){
if(p->a[i].rid == 0) continue;
fprintf(stderr, "%cl: %u, pos: %lu, cnt: %lu, key: %lu, i: %d\n", "+-"[!!(mt->a[i]&0x100000000)],
(uint32_t)mt->a[i], p->a[i].pos, p->a[i].rid, p->a[i].x, i);
// if (mt->a[i]&0x100000000){
// fprintf(stderr, "+l: %u, pos: %lu, cnt: %lu\n", (uint32_t)mt->a[i], p->a[i].pos, p->a[i].rid);
// }
}
**/
ha_mz1_t b[MAX_MAX_HIGH_OCC];
for (i = 0, last0 = -1; i <= n; ++i) {
if (i == n || p->a[i].rid == 0) {
if (i - last0 > 1) {
int32_t ps = last0 < 0? 0 : p->a[last0].pos;
int32_t pe = i == n? len : p->a[i].pos;
if(((int32_t)((double)(pe - ps) / sample_dist + .499)) > 0){
for (m = last0 + 1, mi = 0; m < i; ++m){
if(mt->a[m]&0x100000000) p->a[m].rid = 0, mi++;
}
if(mi == 0) hf_select(p, last0, i, n, len, sample_dist, b, 0);
}
}
last0 = i;
}
}
ff:
for (i = n = 0; i < (int32_t)p->n; ++i) // squeeze out filtered minimizers
if (p->a[i].rid == 0)
p->a[n++] = p->a[i];
p->n = n;
}
void debug_pl(const char *str, int len, int w, int k, int is_hpc, ha_mz1_v *p, const void *hf, st_mt_t *mt)
{
int i, l, dbi, dbcnt = 0, kmer_span = 0;
tiny_queue_t tq;
memset(&tq, 0, sizeof(tiny_queue_t));
uint64_t shift1 = k - 1, mask = (1ULL<<k) - 1, kmer[4] = {0,0,0,0};
for (i = l = dbi = 0; i < len; ++i) {
int c = seq_nt4_table[(uint8_t)str[i]];
if (c < 4) { // not an ambiguous base
int z;
if (is_hpc) {
int skip_len = 1;
if (i + 1 < len && seq_nt4_table[(uint8_t)str[i + 1]] == c) {
for (skip_len = 2; i + skip_len < len; ++skip_len)
if (seq_nt4_table[(uint8_t)str[i + skip_len]] != c)
break;
i += skip_len - 1; // put $i at the end of the current homopolymer run
}
tq_push(&tq, skip_len);
kmer_span += skip_len;
///how many bases that are covered by this HPC k-mer
///kmer_span includes at most k HPC elements
if (tq.count > k) kmer_span -= tq_shift(&tq);
} else kmer_span = l + 1 < k? l + 1 : k;
///kmer_span should be used for HPC k-mer
///non-HPC k-mer, kmer_span should be k
///kmer_span is used to calculate anchor pos on reverse complementary strand
kmer[0] = (kmer[0] << 1 | (c&1)) & mask; // forward k-mer
kmer[1] = (kmer[1] << 1 | (c>>1)) & mask;
kmer[2] = kmer[2] >> 1 | (uint64_t)(1 - (c&1)) << shift1; // reverse k-mer
kmer[3] = kmer[3] >> 1 | (uint64_t)(1 - (c>>1)) << shift1;
if (kmer[1] == kmer[3]) continue; // skip "symmetric k-mers" as we don't know it strand
z = kmer[1] < kmer[3]? 0 : 1; // strand
++l;
if (l >= k && kmer_span < 256) {
uint64_t y;
int32_t cnt;
y = yak_hash64_64(kmer[z<<1|0]) + yak_hash64_64(kmer[z<<1|1]);
cnt = hf? ha_ft_cnt(hf, y) : 0;
for (dbi = 0; dbi < mt->n; dbi++)
{
if(p->a[dbi].x == y && p->a[dbi].rid == cnt && p->a[dbi].pos == i && p->a[dbi].rev == z && p->a[dbi].span == kmer_span)
{
if(l != (int)mt->a[dbi]) fprintf(stderr, "ERROR\n");
dbcnt++;
}
}
}
} else l = 0, tq.count = tq.front = 0, kmer_span = 0;
}
if(dbcnt != mt->n) fprintf(stderr, "ERROR\n");
if(mt->n != (int)p->n) fprintf(stderr, "ERROR\n");
for (dbi = 1; dbi < mt->n; dbi++)
{
if(p->a[dbi].pos <= p->a[dbi-1].pos || (int)mt->a[dbi] <= (int)mt->a[dbi-1])
{
fprintf(stderr, "ERROR\n");
}
}
}
/**
* Find symmetric (w,k)-minimizers on a DNA sequence
*
* @param str DNA sequence
* @param len length of $str
* @param w find a minimizer for every $w consecutive k-mers
* @param k k-mer size
* @param rid reference ID; will be copied to the output $p array
* @param is_hpc homopolymer-compressed or not
* @param p minimizers
*/
void ha_sketch(const char *str, int len, int w, int k, uint32_t rid, int is_hpc, ha_mz1_v *p, const void *hf, int sample_dist, kvec_t_u8_warp* k_flag, kvec_t_u64_warp* dbg_ct, ha_pt_t *pt, int min_freq, int32_t dp_min_len, float dp_e, st_mt_t *mt, int32_t ws)
{ ///in default, w = 51, k = 51, is_hpc = 1
/**
uint64_t x;
uint64_t rid:28, pos:27, rev:1, span:8;
**/
extern void *ha_ct_table;
static const ha_mz1_t dummy = { UINT64_MAX, (1<<28) - 1, 0, 0, 0};
uint64_t shift1 = k - 1, mask = (1ULL<<k) - 1, kmer[4] = {0,0,0,0};
int i, j, l, tl = 0, buf_pos, min_pos, kmer_span = 0;
ha_mz1_t buf[256], min = dummy;
uint32_t buf_p[256], min_s = (uint32_t)-1;
tiny_queue_t tq;
assert(len > 0 && len < 1<<27 && rid < 1<<28 && (w > 0 && w < 256) && (k > 0 && k <= 63));
if (dbg_ct != NULL) dbg_ct->a.n = 0;
if (k_flag != NULL) {
kv_resize(uint8_t, k_flag->a, (uint64_t)len);
k_flag->a.n = len;
memset(k_flag->a.a, 0, k_flag->a.n);
}
memset(buf, 0xff, w * sizeof(ha_mz1_t));
memset(&tq, 0, sizeof(tiny_queue_t));
///len/w is the evaluated minimizer numbers
kv_resize(ha_mz1_t, *p, p->n + len/w);
kv_resize(uint64_t, *mt, (int64_t)p->m); mt->n = p->n;
for (i = l = tl = buf_pos = min_pos = 0; i < len; ++i) {
int c = seq_nt4_table[(uint8_t)str[i]];
ha_mz1_t info = dummy;
if (c < 4) { // not an ambiguous base
int z;
if (is_hpc) {
int skip_len = 1;
if (i + 1 < len && seq_nt4_table[(uint8_t)str[i + 1]] == c) {
for (skip_len = 2; i + skip_len < len; ++skip_len)
if (seq_nt4_table[(uint8_t)str[i + skip_len]] != c)
break;
i += skip_len - 1; // put $i at the end of the current homopolymer run
}
tq_push(&tq, skip_len);
kmer_span += skip_len;
///how many bases that are covered by this HPC k-mer
///kmer_span includes at most k HPC elements
if (tq.count > k) kmer_span -= tq_shift(&tq);
} else kmer_span = l + 1 < k? l + 1 : k;
///kmer_span should be used for HPC k-mer
///non-HPC k-mer, kmer_span should be k
///kmer_span is used to calculate anchor pos on reverse complementary strand
if (k_flag != NULL) k_flag->a.a[i] = 1;///lable all useful base, which are not ignored by HPC
kmer[0] = (kmer[0] << 1 | (c&1)) & mask; // forward k-mer
kmer[1] = (kmer[1] << 1 | (c>>1)) & mask;
kmer[2] = kmer[2] >> 1 | (uint64_t)(1 - (c&1)) << shift1; // reverse k-mer
kmer[3] = kmer[3] >> 1 | (uint64_t)(1 - (c>>1)) << shift1;
if (kmer[1] == kmer[3]) continue; // skip "symmetric k-mers" as we don't know it strand
z = kmer[1] < kmer[3]? 0 : 1; // strand
++l; tl++;
if (l >= k && kmer_span < 256) {
uint64_t y;
int32_t cnt, filtered;
y = yak_hash64_64(kmer[z<<1|0]) + yak_hash64_64(kmer[z<<1|1]);
cnt = hf? ha_ft_cnt(hf, y) : 0;
filtered = (cnt >= 1<<28);
if (dbg_ct != NULL) kv_push(uint64_t, dbg_ct->a, ((((uint64_t)(query_ct_index(ha_ct_table, y))<<1)|filtered)<<32)|(uint64_t)(i));
if (!filtered) info.x = y, info.rid = cnt, info.pos = i, info.rev = z, info.span = kmer_span; // initially ha_mz1_t::rid keeps the k-mer count
if (k_flag != NULL) k_flag->a.a[i]++;
if (k_flag != NULL && filtered > 0) k_flag->a.a[i]++;
}
} else l = 0, tq.count = tq.front = 0, kmer_span = 0;
buf[buf_pos] = info; // need to do this here as appropriate buf_pos and buf[buf_pos] are needed below
buf_p[buf_pos] = l;
if (l == w + k - 1 && min.x != UINT64_MAX) { // special case for the first window - because identical k-mers are not stored yet
for (j = buf_pos + 1; j < w; ++j){
if (mzcmp(&min, &buf[j]) == 0 && buf[j].pos != min.pos){
kv_push(ha_mz1_t, *p, buf[j]); kv_push(uint64_t, *mt, buf_p[j]);
}
}
for (j = 0; j < buf_pos; ++j){
if (mzcmp(&min, &buf[j]) == 0 && buf[j].pos != min.pos){
kv_push(ha_mz1_t, *p, buf[j]); kv_push(uint64_t, *mt, buf_p[j]);
}
}
}
/**
* There are three cases:
* 1. info.x <= min.x, means info is a new minimizer
* 2. info.x > min.x, info is not a new minimizer
* (1) buf_pos != min_pos, do nothing
* (2) buf_pos == min_pos, means current minimizer has moved outside the window
* **/
///three cases: 1.
if (mzcmp(&min, &info) >= 0) { // a new minimum; then write the old min
if (l >= w + k && min.x != UINT64_MAX){
kv_push(ha_mz1_t, *p, min); kv_push(uint64_t, *mt, min_s);
}
min = info, min_pos = buf_pos, min_s = buf_p[buf_pos];
} else if (buf_pos == min_pos) { // old min has moved outside the window
if (l >= w + k - 1 && min.x != UINT64_MAX){
kv_push(ha_mz1_t, *p, min); kv_push(uint64_t, *mt, min_s);
}
///buf_pos == min_pos, means current minimizer has moved outside the window
///so for now we need to find a new minimizer at the current window (w k-mers)
for (j = buf_pos + 1, min = dummy; j < w; ++j) // the two loops are necessary when there are identical k-mers
if (mzcmp(&min, &buf[j]) >= 0) min = buf[j], min_pos = j, min_s = buf_p[j]; // >= is important s.t. min is always the closest k-mer
for (j = 0; j <= buf_pos; ++j)
if (mzcmp(&min, &buf[j]) >= 0) min = buf[j], min_pos = j, min_s = buf_p[j];
if (l >= w + k - 1 && min.x != UINT64_MAX) { // write identical k-mers
for (j = buf_pos + 1; j < w; ++j) // these two loops make sure the output is sorted
if (mzcmp(&min, &buf[j]) == 0 && min.pos != buf[j].pos){
kv_push(ha_mz1_t, *p, buf[j]); kv_push(uint64_t, *mt, buf_p[j]);
}
for (j = 0; j <= buf_pos; ++j)
if (mzcmp(&min, &buf[j]) == 0 && min.pos != buf[j].pos){
kv_push(ha_mz1_t, *p, buf[j]); kv_push(uint64_t, *mt, buf_p[j]);
}
}
}
if (++buf_pos == w) buf_pos = 0;
}
if (min.x != UINT64_MAX){
kv_push(ha_mz1_t, *p, min); kv_push(uint64_t, *mt, min_s);
}
// debug_pl(str, len, w, k, is_hpc, p, hf, mt);
// if (sample_dist > w) select_mz(p, len, MAX_HIGH_OCC, dp_min_len);
select_mz_h(p, mt, len, sample_dist, ws, k, tl);
if (dp_min_len > 0 && pt && mt) refine_sketch(p, pt, len, dp_min_len, dp_e, min_freq, mt);
for (i = 0; i < (int)p->n; ++i) // populate .rid as this was keeping counts
p->a[i].rid = rid;
}
void ha_sketch_worse(const char *str, int len, int w, int k, uint32_t rid, int is_hpc, ha_mz1_v *p, const void *hf, int sample_dist, kvec_t_u8_warp* k_flag, kvec_t_u64_warp* dbg_ct, ha_pt_t *pt, int min_freq, int32_t dp_min_len, float dp_e, st_mt_t *mt)
{ ///in default, w = 51, k = 51, is_hpc = 1
/**
uint64_t x;
uint64_t rid:28, pos:27, rev:1, span:8;
**/
extern void *ha_ct_table;
static const ha_mz1_t dummy = { UINT64_MAX, (1<<28) - 1, 0, 0 };
uint64_t shift1 = k - 1, mask = (1ULL<<k) - 1, kmer[4] = {0,0,0,0};
int i, j, l, buf_pos, min_pos, kmer_span = 0;
ha_mz1_t buf[256], min = dummy;
tiny_queue_t tq;
assert(len > 0 && len < 1<<27 && rid < 1<<28 && (w > 0 && w < 256) && (k > 0 && k <= 63));
if (dbg_ct != NULL) dbg_ct->a.n = 0;
if (k_flag != NULL) {
kv_resize(uint8_t, k_flag->a, (uint64_t)len);
k_flag->a.n = len;
memset(k_flag->a.a, 0, k_flag->a.n);
}
memset(buf, 0xff, w * sizeof(ha_mz1_t));
memset(&tq, 0, sizeof(tiny_queue_t));
///len/w is the evaluated minimizer numbers
kv_resize(ha_mz1_t, *p, p->n + len/w);
for (i = l = buf_pos = min_pos = 0; i < len; ++i) {
int c = seq_nt4_table[(uint8_t)str[i]];
ha_mz1_t info = dummy;
if (c < 4) { // not an ambiguous base
int z;
if (is_hpc) {
int skip_len = 1;
if (i + 1 < len && seq_nt4_table[(uint8_t)str[i + 1]] == c) {
for (skip_len = 2; i + skip_len < len; ++skip_len)
if (seq_nt4_table[(uint8_t)str[i + skip_len]] != c)
break;
i += skip_len - 1; // put $i at the end of the current homopolymer run
}
tq_push(&tq, skip_len);
kmer_span += skip_len;
///how many bases that are covered by this HPC k-mer
///kmer_span includes at most k HPC elements
if (tq.count > k) kmer_span -= tq_shift(&tq);
} else kmer_span = l + 1 < k? l + 1 : k;
///kmer_span should be used for HPC k-mer
///non-HPC k-mer, kmer_span should be k
///kmer_span is used to calculate anchor pos on reverse complementary strand
if (k_flag != NULL) k_flag->a.a[i] = 1;///lable all useful base, which are not ignored by HPC
kmer[0] = (kmer[0] << 1 | (c&1)) & mask; // forward k-mer
kmer[1] = (kmer[1] << 1 | (c>>1)) & mask;
kmer[2] = kmer[2] >> 1 | (uint64_t)(1 - (c&1)) << shift1; // reverse k-mer
kmer[3] = kmer[3] >> 1 | (uint64_t)(1 - (c>>1)) << shift1;
if (kmer[1] == kmer[3]) continue; // skip "symmetric k-mers" as we don't know it strand
z = kmer[1] < kmer[3]? 0 : 1; // strand
++l;
if (l >= k && kmer_span < 256) {
uint64_t y;
int32_t cnt, filtered;
y = yak_hash64_64(kmer[z<<1|0]) + yak_hash64_64(kmer[z<<1|1]);
cnt = hf? ha_ft_cnt(hf, y) : 0;
filtered = (cnt >= 1<<28);
if (dbg_ct != NULL) kv_push(uint64_t, dbg_ct->a, ((((uint64_t)(query_ct_index(ha_ct_table, y))<<1)|filtered)<<32)|(uint64_t)(i));
if (!filtered) info.x = y, info.rid = pt? ha_pt_cnt(pt, y):cnt, info.pos = i, info.rev = z, info.span = kmer_span; // initially ha_mz1_t::rid keeps the k-mer count
if (k_flag != NULL) k_flag->a.a[i]++;
if (k_flag != NULL && filtered > 0) k_flag->a.a[i]++;
}
} else l = 0, tq.count = tq.front = 0, kmer_span = 0;
//for non-HPC k-mer, l = i; but for HPC k-mer, l is always less than i
//i is the real base iterator, while l is the HPC base iterator
//only if l >= k, info is a useful minimizer (ha_mz1_t.x != UINT64_MAX)
//but even if l < k, infor is still stored into buf
buf[buf_pos] = info; // need to do this here as appropriate buf_pos and buf[buf_pos] are needed below
if (l == w + k - 1 && min.x != UINT64_MAX) { // special case for the first window - because identical k-mers are not stored yet
for (j = buf_pos + 1; j < w; ++j)
if (mzcmp(&min, &buf[j]) == 0 && buf[j].pos != min.pos) kv_push(ha_mz1_t, *p, buf[j]);
for (j = 0; j < buf_pos; ++j)
if (mzcmp(&min, &buf[j]) == 0 && buf[j].pos != min.pos) kv_push(ha_mz1_t, *p, buf[j]);
}
/**
* There are three cases:
* 1. info.x <= min.x, means info is a new minimizer
* 2. info.x > min.x, info is not a new minimizer
* (1) buf_pos != min_pos, do nothing
* (2) buf_pos == min_pos, means current minimizer has moved outside the window
* **/
///three cases: 1.
if (info.x <= min.x) { // a new minimum; then write the old min
if (l >= w + k && min.x != UINT64_MAX) kv_push(ha_mz1_t, *p, min);
min = info, min_pos = buf_pos;
} else if (buf_pos == min_pos) { // old min has moved outside the window
if (l >= w + k - 1 && min.x != UINT64_MAX) kv_push(ha_mz1_t, *p, min);
///buf_pos == min_pos, means current minimizer has moved outside the window
///so for now we need to find a new minimizer at the current window (w k-mers)
for (j = buf_pos + 1, min.x = UINT64_MAX; j < w; ++j) // the two loops are necessary when there are identical k-mers
if (mzcmp(&min, &buf[j]) >= 0) min = buf[j], min_pos = j; // >= is important s.t. min is always the closest k-mer
for (j = 0; j <= buf_pos; ++j)
if (mzcmp(&min, &buf[j]) >= 0) min = buf[j], min_pos = j;
if (l >= w + k - 1 && min.x != UINT64_MAX) { // write identical k-mers
for (j = buf_pos + 1; j < w; ++j) // these two loops make sure the output is sorted
if (mzcmp(&min, &buf[j]) == 0 && min.pos != buf[j].pos) kv_push(ha_mz1_t, *p, buf[j]);
for (j = 0; j <= buf_pos; ++j)
if (mzcmp(&min, &buf[j]) == 0 && min.pos != buf[j].pos) kv_push(ha_mz1_t, *p, buf[j]);
}
}
if (++buf_pos == w) buf_pos = 0;
}
if (min.x != UINT64_MAX)
kv_push(ha_mz1_t, *p, min);
if (dp_min_len > 0 && pt && mt) refine_sketch(p, pt, len, dp_min_len, dp_e, min_freq, mt);
// if (sample_dist > w) select_mz(p, len, MAX_HIGH_OCC);
for (i = 0; i < (int)p->n; ++i) // populate .rid as this was keeping counts
p->a[i].rid = rid;
}
Reference in New Issue View Git Blame Copy Permalink