LVAL xlc_seq_get(void)
{
seq_type arg1 = getseq(xlgaseq());
long arg2 = 0;
long arg3 = 0;
long arg4 = 0;
long arg5 = 0;
long arg6 = 0;
long arg7 = 0;
long arg8 = 0;
LVAL result;
xllastarg();
seq_get(arg1, &arg2, &arg3, &arg4, &arg5, &arg6, &arg7, &arg8);
{ LVAL *next = &getvalue(RSLT_sym);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg2); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg3); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg4); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg5); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg6); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg7); next = &cdr(*next);
*next = cons(NIL, NIL);
car(*next) = cvfixnum(arg8);
}
result = getvalue(RSLT_sym);
return result;
}
LVAL xlc_seq_reset(void)
{
seq_type arg1 = getseq(xlgaseq());
xllastarg();
seq_reset(arg1);
return NIL;
}
LVAL xlc_seq_write_smf(void)
{
seq_type arg1 = getseq(xlgaseq());
FILE * arg2 = getfile(xlgastream());
xllastarg();
seq_write_smf(arg1, arg2);
return NIL;
}
LVAL xlc_seq_copy(void)
{
seq_type arg1 = getseq(xlgaseq());
seq_type result;
xllastarg();
result = seq_copy(arg1);
return cvseq(result);
}
LVAL xlc_seq_next(void)
{
seq_type arg1 = getseq(xlgaseq());
boolean result;
xllastarg();
result = seq_next(arg1);
return cvboolean(result);
}
LVAL xlc_seq_write(void)
{
seq_type arg1 = getseq(xlgaseq());
FILE * arg2 = getfile(xlgastream());
int arg3 = getboolean(xlgetarg());
xllastarg();
seq_write(arg1, arg2, arg3);
return NIL;
}
main(int argc, char **argv) {
char *aa0;
char libstr[MAX_FN];
char qname[MAX_FN];
int sq0off;
int i, n0;
FILE *fp;
struct pstruct pst, *ppst;
/* stuff from initfa.c/h_init() */
memcpy(qascii,aascii,sizeof(qascii));
/* initialize a pam matrix */
ppst = &pst;
strncpy(ppst->pamfile,"BL50",MAX_FN);
standard_pam(ppst->pamfile,ppst,0,0);
/* this is always protein by default */
ppst->nsq = naa;
ppst->nsqx = naax;
for (i=0; i<=ppst->nsqx; i++) {
ppst->sq[i] = aa[i];
ppst->hsq[i] = haa[i];
ppst->sqx[i]=aax[i]; /* sq = aa */
ppst->hsqx[i]=haax[i]; /* hsq = haa */
}
ppst->sq[ppst->nsqx+1] = ppst->sqx[ppst->nsqx+1] = '\0';
if ((aa0 = calloc(MAXTST,sizeof(char)))==NULL) {
fprintf(stderr,"Cannot allocate aa0\n");
exit(1);
}
initenv(argc, argv, &pst, qname);
alloc_pam(pst.nsq+1,pst.nsq+1, &pst);
initpam2(&pst);
n0 = getseq (qname, qascii, aa0, MAXTST, libstr,&sq0off);
if (!pst.pam_pssm) {
fprintf(stderr," ** ERROR ** No -P PSSM provided\n");
}
else {
ppst->pam2p[0] = alloc_pam2p(n0,pst.nsq);
ppst->pam2p[1] = alloc_pam2p(n0,pst.nsq);
if ((fp = fopen(pst.pgpfile,"rb"))!=NULL) {
read_pssm(aa0, n0, pst.nsq, pst.pamscale,fp,ppst);
}
}
}
LVAL xlc_seq_insert_macctrl(void)
{
seq_type arg1 = getseq(xlgaseq());
long arg2 = getfixnum(xlgafixnum());
long arg3 = getfixnum(xlgafixnum());
long arg4 = getfixnum(xlgafixnum());
long arg5 = getfixnum(xlgafixnum());
long arg6 = getfixnum(xlgafixnum());
xllastarg();
insert_macctrl(arg1, arg2, arg3, arg4, arg5, arg6);
return NIL;
}
int
main(int argc, char * argv[])
{
if (argc != 3)
usage(argv[0]);
int n = atol(argv[2]);
srand(701888);
env = sp_env();
sp_ctl(env, SPDIR, SPO_CREAT|SPO_RDWR, "./sophia_bench_data");
db = sp_open(env);
if (db == NULL) {
printf("open: %s\n", sp_error(env));
return 1;
}
int rc = 0;
unsigned long long start = now();
switch (benchof(argv[1]))
{
case SETSEQ: rc = setseq(n);
break;
case SETRAND: rc = setrand(n);
break;
case GETSEQ:rc = getseq(n);
break;
case GETRAND: rc = getrand(n);
break;
case RANGEFWD: rc = rangefwd(n);
break;
case UNKNOWN:
printf("unknown operation\n");
}
if (rc == 0) {
unsigned long long diff = now() - start;
float rps = n / (diff / 1000.0);
printf("%d rps\n", (int)rps);
}
if (keytraversed > 0 && keytraversed != n)
printf("(%d keys traversed)\n", keytraversed);
sp_destroy(db);
sp_destroy(env);
return rc;
}
LVAL xlc_seq_insert_ramp(void)
{
seq_type arg1 = getseq(xlgaseq());
long arg2 = getfixnum(xlgafixnum());
long arg3 = getfixnum(xlgafixnum());
long arg4 = getfixnum(xlgafixnum());
long arg5 = getfixnum(xlgafixnum());
long arg6 = getfixnum(xlgafixnum());
long arg7 = getfixnum(xlgafixnum());
long arg8 = getfixnum(xlgafixnum());
long arg9 = getfixnum(xlgafixnum());
xllastarg();
insert_ctrlramp(arg1, arg2, arg3, arg4, arg5, arg6, arg7, arg8, arg9);
return NIL;
}
static void elcount( int *npnts, int *npols )
{
int i, apts, apls, bpts, bpls, ppts, ppls;
getseq();
for ( i = 0, bpls = 0; i < seqlen; i++ )
bpls += bond_count( seq[ i ] );
if ( bond_type == BTYPE_LINE ) {
for ( i = 0, bpts = 0; i < seqlen; i++ )
bpts += point_count( seq[ i ] ) - 4;
bpts += 2;
}
else {
bpts = bpls * bond_nsides * ( 1 + bond_nsegments );
bpls *= bond_nsides * bond_nsegments + 2;
}
for ( i = 0, apls = 0; i < seqlen; i++ )
apls += atom_count( seq[ i ] );
if ( atom_type == ATYPE_POINT )
apts = ( bond_type == BTYPE_LINE ) ? 0 : apls;
else if ( atom_type == ATYPE_SPHERE ) {
apts = apls * ( atom_nsides * ( atom_nsegments - 1 ) + 2 );
apls *= atom_nsides * atom_nsegments;
}
else {
apts = 20 * apls;
apls *= 12;
}
if ( do_plates ) {
ppts = ( atom_type == ATYPE_POINT || bond_type == BTYPE_LINE ) ?
0 : 15 * seqlen;
ppls = 2 * seqlen;
}
else ppts = ppls = 0;
if ( npnts ) *npnts = apts + bpts + ppts;
if ( npols ) *npols = apls + bpls + ppls;
}
static void
join (
FILE *fp1,
FILE *fp2 )
{
struct seq seq1, seq2;
struct line line;
int diff, i, j, eof1, eof2;
/* Read the first line of each file. */
initseq (&seq1);
getseq (fp1, &seq1);
initseq (&seq2);
getseq (fp2, &seq2);
while (seq1.count && seq2.count)
{
diff = keycmp (&seq1.lines[0], &seq2.lines[0]);
if (diff < 0)
{
if (print_unpairables_1)
prline (&seq1.lines[0]);
freeline (&seq1.lines[0]);
seq1.count = 0;
getseq (fp1, &seq1);
continue;
}
if (diff > 0)
{
if (print_unpairables_2)
prline (&seq2.lines[0]);
freeline (&seq2.lines[0]);
seq2.count = 0;
getseq (fp2, &seq2);
continue;
}
/* Keep reading lines from file1 as long as they continue to
match the current line from file2. */
eof1 = 0;
do
if (!getseq (fp1, &seq1))
{
eof1 = 1;
++seq1.count;
break;
}
while (!keycmp (&seq1.lines[seq1.count - 1], &seq2.lines[0]));
/* Keep reading lines from file2 as long as they continue to
match the current line from file1. */
eof2 = 0;
do
if (!getseq (fp2, &seq2))
{
eof2 = 1;
++seq2.count;
break;
}
while (!keycmp (&seq1.lines[0], &seq2.lines[seq2.count - 1]));
if (print_pairables)
{
for (i = 0; i < seq1.count - 1; ++i)
for (j = 0; j < seq2.count - 1; ++j)
prjoin (&seq1.lines[i], &seq2.lines[j]);
}
for (i = 0; i < seq1.count - 1; ++i)
freeline (&seq1.lines[i]);
if (!eof1)
{
seq1.lines[0] = seq1.lines[seq1.count - 1];
seq1.count = 1;
}
else
seq1.count = 0;
for (i = 0; i < seq2.count - 1; ++i)
freeline (&seq2.lines[i]);
if (!eof2)
{
seq2.lines[0] = seq2.lines[seq2.count - 1];
seq2.count = 1;
}
else
seq2.count = 0;
}
if (print_unpairables_1 && seq1.count)
{
prline (&seq1.lines[0]);
freeline (&seq1.lines[0]);
while (getline (fp1, &line))
{
prline (&line);
freeline (&line);
}
}
if (print_unpairables_2 && seq2.count)
{
prline (&seq2.lines[0]);
freeline (&seq2.lines[0]);
while (getline (fp2, &line))
{
prline (&line);
freeline (&line);
}
}
delseq (&seq1);
delseq (&seq2);
}
unsigned long long * get_kmer_counts_from_file(const char *fn, const unsigned int kmer) {
char *line = NULL;
size_t len = 0;
ssize_t read;
long long i = 0;
long long position = 0;
FILE * const fh = fopen(fn, "r");
if(fh == NULL) {
fprintf(stderr, "Error opening %s - %s\n", fn, strerror(errno));
exit(EXIT_FAILURE);
}
// width is 4^kmer
// there's a sneaky bitshift to avoid pow dependency
const unsigned long width = pow_four(kmer);
// malloc our return array
unsigned long long * counts = malloc((width+ 1) * sizeof(unsigned long long));
memset(counts, 0, width * sizeof(unsigned long long));
if(counts == NULL) {
fprintf(stderr, strerror(errno));
exit(EXIT_FAILURE);
}
char *str = malloc(4096);
if(str == NULL) {
fprintf(stderr, strerror(errno));
exit(EXIT_FAILURE);
}
unsigned long long str_size = 4096;
while ((read = getseq(&line, &len, fh)) != -1) {
// find our first \n, this should be the end of the header
char *start = strchr(line, '\n');
if(start == NULL)
continue;
start = start + 1;
size_t start_len = strlen(start);
// if our current str buffer isn't big enough, realloc
if(start_len + 1 > str_size + 1) {
str = realloc(str, start_len + 1);
if(str == NULL) {
exit(EXIT_FAILURE);
fprintf(stderr, strerror(errno));
}
}
// strip out all other newlines to handle multiline sequences
str = strnstrip(start, str, '\n',start_len);
ssize_t seq_length = strlen(str);
if(seq_length < kmer)
continue;
// relace A, C, G and T with 0, 1, 2, 3 respectively
// everything else is 5
for(i = 0; i < seq_length; i++) {
str[i] = alpha[(int)str[i]];
}
// loop through our string to process each k-mer
for(position = 0; position < (seq_length - kmer + 1); position++) {
unsigned long mer = 0;
unsigned long multiply = 1;
// for each char in the k-mer check if it is an error char
for(i = position + kmer - 1; i >= position; i--){
if(str[i] == 5) {
mer = width;
position = i;
goto next;
}
// if it's a newline, we should skip it
// multiply this char in the mer by the multiply
// and bitshift the multiply for the next round
mer += str[i] * multiply;
multiply = multiply << 2;
}
// use this point to get mer of our loop
next:
// bump up the mer value in the counts array
counts[mer]++;
}
}
free(line);
free(str);
fclose(fh);
return counts;
}
