mirrors
/
klib
mirror of https://github.com/attractivechaos/klib
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

keep the ending position

This commit is contained in:
Heng Li 2011-05-07 16:56:00 -04:00
parent 0d5864823f
commit 93f537223b
2 changed files with 65 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

90
ksw.c
View File

@ -37,9 +37,9 @@
#endif
struct _ksw_query_t {
int slen;
int qlen, slen;
uint8_t shift, mdiff;
__m128i *qp, *H0, *H1, *E;
__m128i *qp, *H0, *H1, *E, *Hmax;
};
ksw_query_t *ksw_qinit(int p, int qlen, const uint8_t *query, int m, const int8_t *mat)
@ -50,12 +50,13 @@ ksw_query_t *ksw_qinit(int p, int qlen, const uint8_t *query, int m, const int8_
slen = (qlen + p - 1) / p;
qlen16 = (qlen + 15) >> 4 << 4;
q = malloc(sizeof(ksw_query_t) + 256 + qlen16 * (m + 3)); // a single block of memory
q = malloc(sizeof(ksw_query_t) + 256 + qlen16 * (m + 4)); // a single block of memory
q->qp = (__m128i*)(((size_t)q + sizeof(ksw_query_t) + 15) >> 4 << 4); // align memory
q->H0 = q->qp + (qlen16 * m) / 16;
q->H1 = q->H0 + qlen16 / 16;
q->E = q->H1 + qlen16 / 16;
q->slen = slen;
q->Hmax = q->E + qlen16 / 16;
q->slen = slen; q->qlen = qlen;
// compute shift
tmp = m * m;
for (a = 0, q->shift = 127, q->mdiff = 0; a < tmp; ++a) { // find the minimum and maximum score
@ -77,10 +78,10 @@ ksw_query_t *ksw_qinit(int p, int qlen, const uint8_t *query, int m, const int8_
return q;
}
int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, unsigned _o, unsigned _e) // the first gap costs -(_o+_e)
int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, ksw_aux_t *a) // the first gap costs -(_o+_e)
{
int slen, i, sum, score;
__m128i zero, gapo, gape, shift, gmax, reduce, thres, *H0, *H1, *E;
__m128i zero, gapoe, gape, shift, gmax, reduce, thres, *H0, *H1, *E, *Hmax;
#define __max_16(ret, xx) do { \
(xx) = _mm_max_epu8((xx), _mm_srli_si128((xx), 8)); \
@ -91,13 +92,14 @@ int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, unsigned _o, un
} while (0)
// initialization
a->te = -1;
gmax = zero = _mm_set1_epi32(0);
gapo = _mm_set1_epi8(_o + _e);
gape = _mm_set1_epi8(_e);
gapoe = _mm_set1_epi8(a->gapo + a->gape);
gape = _mm_set1_epi8(a->gape);
shift = _mm_set1_epi8(q->shift);
thres = _mm_set1_epi8(254 - q->mdiff); // when max score exceeds this, shift all scores by "reduce" below
reduce = _mm_set1_epi8(127);
H0 = q->H0; H1 = q->H1; E = q->E;
H0 = q->H0; H1 = q->H1; E = q->E; Hmax = q->Hmax;
slen = q->slen;
for (i = 0; i < slen; ++i) {
_mm_store_si128(E + i, zero);
@ -124,7 +126,7 @@ int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, unsigned _o, un
max = _mm_max_epu8(max, h); // set max
_mm_store_si128(H1 + j, h); // save to H'(i,j)
// now compute E'(i+1,j)
h = _mm_subs_epu8(h, gapo); // h=H'(i,j)-gapo
h = _mm_subs_epu8(h, gapoe); // h=H'(i,j)-gapo
e = _mm_subs_epu8(e, gape); // e=E'(i,j)-gape
e = _mm_max_epu8(e, h); // e=E'(i+1,j)
_mm_store_si128(E + j, e); // save to E'(i+1,j)
@ -134,38 +136,55 @@ int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, unsigned _o, un
// get H'(i-1,j) and prepare for the next j
h = _mm_load_si128(H0 + j); // h=H'(i-1,j)
}
gmax = _mm_max_epu8(gmax, max); // NB: H(i,j) updated in the lazy-F loop cannot exceed max
// NB: we do not need to set E(i,j) as we disallow adjecent insertion and then deletion
for (k = 0; LIKELY(k < 16); ++k) { // this block mimics SWPS3
for (k = 0; LIKELY(k < 16); ++k) { // this block mimics SWPS3; NB: H(i,j) updated in the lazy-F loop cannot exceed max
f = _mm_slli_si128(f, 1);
for (j = 0; LIKELY(j < slen); ++j) {
h = _mm_load_si128(H1 + j);
h = _mm_max_epu8(h, f); // h=H'(i,j)
_mm_store_si128(H1 + j, h);
h = _mm_subs_epu8(h, gapo);
h = _mm_subs_epu8(h, gapoe);
f = _mm_subs_epu8(f, gape);
cmp = _mm_movemask_epi8(_mm_cmpeq_epi8(_mm_subs_epu8(f, h), zero));
if (UNLIKELY(cmp == 0xffff)) goto end_loop;
}
}
end_loop:
if (_mm_movemask_epi8(_mm_cmpeq_epi8(_mm_subs_epu8(gmax, thres), zero)) != 0xffff) { // reduce
/* When overflow is going to happen, subtract 127 from all scores. This heuristic
* may miss the best alignment, but in practice this should happen very rarely. */
sum += 127;
gmax = _mm_subs_epu8(gmax, reduce);
//int k;for (k=0;k<16;++k)printf("%d ", ((uint8_t*)&thres)[k]);printf("\n");
for (j = 0; LIKELY(j < slen); ++j) {
h = _mm_subs_epu8(_mm_load_si128(H1 + j), reduce);
_mm_store_si128(H1 + j, h);
e = _mm_subs_epu8(_mm_load_si128(E + j), reduce);
_mm_store_si128(E + j, e);
if (_mm_movemask_epi8(_mm_cmpeq_epi8(_mm_subs_epu8(max, gmax), zero)) != 0xffff) { // if max > gmax at any position
int imax, igmax;
__max_16(imax, max); // maximum in max
__max_16(igmax, gmax); // maximum in gmax
if (imax > igmax) { // if the max has been updated
a->te = i; // the end position on the target
for (j = 0; LIKELY(j < slen); ++j) // keep the H1 vector
_mm_store_si128(Hmax + j, _mm_load_si128(H1 + j));
}
gmax = _mm_max_epu8(gmax, max); // update gmax
if (_mm_movemask_epi8(_mm_cmpeq_epi8(_mm_subs_epu8(gmax, thres), zero)) != 0xffff) { // reduce
/* When overflow is going to happen, subtract 127 from all scores. This heuristic
* may miss the best alignment, but in practice this should happen very rarely. */
sum += 127;
gmax = _mm_subs_epu8(gmax, reduce);
//int k;for (k=0;k<16;++k)printf("%d ", ((uint8_t*)&thres)[k]);printf("\n");
for (j = 0; LIKELY(j < slen); ++j) {
h = _mm_subs_epu8(_mm_load_si128(H1 + j), reduce);
_mm_store_si128(H1 + j, h);
e = _mm_subs_epu8(_mm_load_si128(E + j), reduce);
_mm_store_si128(E + j, e);
}
}
}
S = H1; H1 = H0; H0 = S; // swap H0 and H1
}
__max_16(score, gmax);
return score + sum;
a->score = score + sum;
{ // get a->qe, the end of query match
int max = -1;
uint8_t *t = (uint8_t*)Hmax;
for (i = 0, score = 0, a->qe = -1; i < q->qlen; ++i, ++t)
if ((int)*t > max) max = *t, a->qe = i / 16 + i % 16 * slen;
}
return a->score;
}
/*******************************************
@ -201,22 +220,25 @@ unsigned char seq_nt4_table[256] = {
int main(int argc, char *argv[])
{
int c, sa = 1, sb = 3, sq = 5, sr = 2, i, j, k, forward_only = 0;
int c, sa = 1, sb = 3, i, j, k, forward_only = 0;
int8_t mat[25];
ksw_aux_t a;
gzFile fpt, fpq;
kseq_t *kst, *ksq;
// parse command line
while ((c = getopt(argc, argv, "a:b:q:r:f")) >= 0) {
a.gapo = 5; a.gape = 2; a.T = 20;
while ((c = getopt(argc, argv, "a:b:q:r:ft:")) >= 0) {
switch (c) {
case 'a': sa = atoi(optarg); break;
case 'b': sb = atoi(optarg); break;
case 'q': sq = atoi(optarg); break;
case 'r': sr = atoi(optarg); break;
case 'q': a.gapo = atoi(optarg); break;
case 'r': a.gape = atoi(optarg); break;
case 't': a.T = atoi(optarg); break;
case 'f': forward_only = 1; break;
}
}
if (optind + 2 > argc) {
fprintf(stderr, "Usage: ksw [-a%d] [-b%d] [-q%d] [-r%d] <target.fa> <query.fa>\n", sa, sb, sq, sr);
fprintf(stderr, "Usage: ksw [-a%d] [-b%d] [-q%d] [-r%d] <target.fa> <query.fa>\n", sa, sb, a.gapo, a.gape);
return 1;
}
// initialize scoring matrix
@ -248,11 +270,11 @@ int main(int argc, char *argv[])
while (kseq_read(kst) > 0) {
int s;
for (i = 0; i < kst->seq.l; ++i) kst->seq.s[i] = seq_nt4_table[(int)kst->seq.s[i]];
s = ksw_sse2_16(q[0], kst->seq.l, (uint8_t*)kst->seq.s, sq, sr);
printf("%s\t%s\t+\t%d\n", ksq->name.s, kst->name.s, s);
s = ksw_sse2_16(q[0], kst->seq.l, (uint8_t*)kst->seq.s, &a);
printf("%s\t%s\t+\t%d\t%d\t%d\n", ksq->name.s, kst->name.s, s, a.te+1, a.qe+1);
if (q[1]) {
s = ksw_sse2_16(q[1], kst->seq.l, (uint8_t*)kst->seq.s, sq, sr);
printf("%s\t%s\t-\t%d\n", ksq->name.s, kst->name.s, s);
s = ksw_sse2_16(q[1], kst->seq.l, (uint8_t*)kst->seq.s, &a);
printf("%s\t%s\t-\t%d\t%d\t%d\n", ksq->name.s, kst->name.s, s, a.te+1, a.qe+1);
}
}
free(q[0]); free(q[1]);

10
ksw.h
View File

@ -4,12 +4,20 @@
struct _ksw_query_t;
typedef struct _ksw_query_t ksw_query_t;
typedef struct {
// input
unsigned gapo, gape; // the first gap costs gapo+gape
unsigned T; // threshold
// output
int score, te, qe;
} ksw_aux_t;
#ifdef __cplusplus
extern "C" {
#endif
ksw_query_t *ksw_qinit(int p, int qlen, const uint8_t *query, int m, const int8_t *mat); // to free, simply call free()
int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, unsigned o, unsigned e); // first gap costs -(o+e)
int ksw_sse2_16(ksw_query_t *q, int tlen, const uint8_t *target, ksw_aux_t *a);
#ifdef __cplusplus
}