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;
}
long get_min_seq(seq_t seqs, long min) {
int i, n = seq_length(seqs);
for (i = 0; i < n; ++i)
min = MIN(min, seq_get(&seqs[i]));
return min;
}
int driver (int t_num)
{
int i, j;
/* Actually, WaitTimes are needed... */
while( activate_transaction ){
switch(seq_get()){
case 0:
do_neword(t_num);
break;
case 1:
do_payment(t_num);
break;
case 2:
do_ordstat(t_num);
break;
case 3:
do_delivery(t_num);
break;
case 4:
do_slev(t_num);
break;
default:
printf("Error - Unknown sequence.\n");
}
}
return(0);
}
void c_seq::seek(uint32 keyId, int index, wisdom_IOStream& io)
{
SEQ_INIT();
if (is_clean())
{
wisdom_IOStream os = new c_ostream_array;
remove(os);
}
c_rlock lock(&m_lock);
io->push(ZRESULT_OK);
if (keyId == 0)
{
keyId = m_seq_head.g_index();
}
bool down_up = index > 0 ? true : false;
index = abs(index);
for (int i = 0; i < index; i++)
{
string value;
if (!seq_get(keyId, value))
{
break;
}
char* ptr = (char*)value.c_str() + sizeof(_zmsg_head);
int len = value.length() - sizeof(_zmsg_head);
_zmsg_head* h = (_zmsg_head*)value.c_str();
c_time t(h->g_effective());
io->push(__tos(keyId << "," << t.time_stamp() << ","));
io->push(ptr, len);
if (down_up)
keyId++;
else
keyId--;
}
}
int main(int argc, char *argv[])
{
SEQ *sf;
uchar *s;
FILE *f;
char buf[10000];
char head[MAX_LEN];
char cur[LEN_NAME], chr_name[LEN_NAME], annot[LEN_NAME], gname[LEN_NAME], filter[LEN_NAME];
int gid = -1;
int rid = -1;
int i = 0;
int b = 0, e = 1, num_cds = 0;
char dir[3];
struct exons_list *exons;
char annot_name[LEN_NAME];
float qual = (float)0;
char ref[LEN_NAME], alt[LEN_NAME];
int rest = 0;
char codon[4], alt_codon[4];
char aa1 = '\0', aa2 = '\0';
int num_rmsk = 0;
struct exons_list *rmsk;
int num_snps = 0, num_pass = 0, num_filter = 0, num_coding1 = 0, num_syn1 = 0, num_non1 = 0, num_repeats1 = 0, num_coding_repeats1 = 0;
int num_coding = 0, num_syn = 0, num_non = 0, num_repeats = 0, num_coding_repeats = 0;
bool is_num_print = false;
strcpy(buf, "");
strcpy(head, "");
strcpy(cur, "");
strcpy(chr_name, "");
strcpy(annot, "");
strcpy(gname, "");
strcpy(annot_name, "");
strcpy(ref, "");
strcpy(alt, "");
strcpy(codon, "");
strcpy(alt_codon, "");
strcpy(dir, "");
codon[3] = '\0';
alt_codon[3] = '\0';
if( argc != 7 ) {
printf("link_to_annot vcf_file gff_file seq_file annot_type(exon, gene, ...) rmsk_file print_mode(NUM or SITES)\n");
return EXIT_FAILURE;
}
else {
if(!(f = ckopen(argv[2], "r"))) {
printf("no file %s exists\n", argv[2]);
return EXIT_FAILURE;
}
strcpy(annot_name, argv[4]);
if( strcmp(annot_name, "exon") != 0 ) {
fatalf("seq file is required only when the annot type is exon, but %s here\n", annot_name);
}
sf = seq_get(argv[3]);
s = SEQ_CHARS(sf) - 1;
if( strcmp(argv[6], "NUM") == 0 ) {
is_num_print = true;
}
else if( strcmp(argv[6], "SITES") == 0 ) {
is_num_print = false;
}
else {
fatalf("unsupported print option: %s\n", argv[6]);
}
}
compl['a'] = compl['A'] = 'T';
compl['c'] = compl['C'] = 'G';
compl['g'] = compl['G'] = 'C';
compl['t'] = compl['T'] = 'A';
while(fgets(buf, 10000, f))
{
if( (buf[0] == '#') || (buf[0] == '>') ) {}
else if( sscanf(buf, "%*s %*s %s %d %d %*s", annot, &b, &e) != 3 ) {
fatalf("line in wrong gff format: %s\n", buf);
}
else {
if( strcmp(annot, annot_name) == 0 ) {
num_cds++;
}
}
}
if( num_cds > 0 ) exons = (struct exons_list *) ckalloc(num_cds * sizeof(struct exons_list));
initialize_exons_list(exons, 0, num_cds);
fseek(f, 0, SEEK_SET);
i = 0;
while(fgets(buf, 10000, f))
{
if( (buf[0] == '#') || (buf[0] == '>') ) {}
else if( sscanf(buf, "%s %*s %s %d %d %*s %s %*s %s", chr_name, annot, &b, &e, dir, cur) != 6 ) {
fatalf("line in wrong gff format: %s\n", buf);
}
else {
if( strcmp(annot, annot_name) == 0 ) {
get_gene_name(cur, gname);
strcpy(exons[i].name, gname);
exons[i].reg = assign_I(b, e);
exons[i].dir = dir[0];
strcpy(exons[i].chr, chr_name);
i++;
}
}
}
if( i != num_cds ) {
fatalf("%s counting error: %d - %d\n", annot_name, num_cds, i);
}
fclose(f);
if(!(f = ckopen(argv[5], "r"))) {
fatalf("%s file not found\n", argv[5]);
}
rmsk = 0;
while(fgets(buf, 10000, f))
{
if( (buf[0] == '#') || (buf[0] == '>') ) {}
else if( sscanf(buf, "%*s %*s %s %d %d %*s", annot, &b, &e) != 3 ) {
fatalf("line in wrong gff format: %s\n", buf);
}
else {
num_rmsk++;
}
}
if( num_rmsk > 0 ) rmsk = (struct exons_list *) ckalloc(num_rmsk * sizeof(struct exons_list));
initialize_exons_list(rmsk, 0, num_rmsk);
fseek(f, 0, SEEK_SET);
i = 0;
while(fgets(buf, 10000, f))
{
if( (buf[0] == '#') || (buf[0] == '>') ) {}
else if( sscanf(buf, "%s %*s %s %d %d %*s %s %*s %s", chr_name, annot, &b, &e, dir, cur) != 6 ) {
fatalf("line in wrong gff format: %s\n", buf);
}
else {
strcpy(rmsk[i].name, annot);
rmsk[i].reg = assign_I(b, e);
rmsk[i].dir = dir[0];
strcpy(rmsk[i].chr, chr_name);
i++;
}
}
if( i != num_rmsk ) {
fatalf("%s counting error: %d - %d\n", annot_name, num_cds, i);
}
fclose(f);
if(!(f = ckopen(argv[1], "r"))) {
printf("no file %s exists\n", argv[1]);
return EXIT_FAILURE;
}
i = 0;
while(fgets(buf, 10000, f))
{
if( buf[0] != '#' ) {
num_snps++;
if( sscanf(buf, "%s %d %*s %s %s %f %s %*s", chr_name, &b, ref, alt, &qual, filter) != 6 ) {
fatalf("bad format in %s\n", buf);
}
else
{
if( strstr(filter, "PASS") == 0 ) {
num_pass++;
}
else if( strstr(filter, "filter") == 0 ) {
num_filter++;
}
rid = -1;
rid = find_overlap_gene(chr_name, b, rmsk, num_rmsk);
if( rid != -1 ) {
num_repeats++;
if( strstr(filter, "filter") == 0 ) {}
else if( strstr(filter, "PASS") == 0 ) {
num_repeats1++;
}
else {
fatalf("unexpected filter option: %s\n", filter);
}
}
if( (gid = find_overlap_gene(chr_name, b, exons, num_cds)) != -1 ) {
num_coding++;
if( strstr(filter, "PASS") == 0 ) {
num_coding1++;
}
if( ref[0] != s[b] ) {
fatalf("nucleotides not match: %c - %c\n", alt, s[b]);
}
if( exons[gid].dir == '+' ) {
rest = (b - exons[gid].reg.lower)%3;
if( rest == 0 ) {
sprintf(codon, "%c%c%c", s[b], s[b+1], s[b+2]);
sprintf(alt_codon, "%c%c%c", alt[0], s[b+1], s[b+2]);
}
else if( rest == 1 ) {
sprintf(codon, "%c%c%c", s[b-1], s[b], s[b+1]);
sprintf(alt_codon, "%c%c%c", s[b-1], alt[0], s[b+1]);
}
else {
sprintf(codon, "%c%c%c", s[b-2], s[b-1], s[b]);
sprintf(alt_codon, "%c%c%c", s[b-2], s[b-1], alt[0]);
}
}
else if( exons[gid].dir == '-' ) {
rest = (b - exons[gid].reg.upper)%3;
if( rest == 0 ) {
sprintf(codon, "%c%c%c", compl[s[b]], compl[s[b-1]], compl[s[b-2]]);
sprintf(alt_codon, "%c%c%c", compl[alt[0]], compl[s[b-1]], compl[s[b-2]]);
}
else if( rest == 1 ) {
sprintf(codon, "%c%c%c", compl[s[b+1]], compl[s[b]], compl[s[b-1]]);
sprintf(alt_codon, "%c%c%c", compl[s[b+1]], compl[alt[0]], compl[s[b-1]]);
}
else {
sprintf(codon, "%c%c%c", compl[s[b+2]], compl[s[b+1]], compl[s[b]]);
sprintf(alt_codon, "%c%c%c", compl[s[b+2]], compl[s[b+1]], compl[alt[0]]);
}
}
else {
fatalf("%c unsupported\n", exons[gid].dir);
}
aa1 = dna2oneaa(codon);
aa2 = dna2oneaa(alt_codon);
if( aa1 == aa2 ) {
num_syn++;
if( strstr(filter, "filter") == 0) {
}
else if( strstr(filter, "PASS") == 0 ) {
num_syn1++;
}
else {
fatalf("unexpected filter option: %s\n", filter);
}
}
else {
num_non++;
if( strstr(filter, "filter") == 0) {
}
else if( strstr(filter, "PASS") == 0 ) {
num_non1++;
}
else {
fatalf("unexpected filter option: %s\n", filter);
}
}
if( rid != -1 ) {
num_coding_repeats++;
if( strstr(filter, "PASS") == 0 ) {
num_coding_repeats1++;
}
}
if( is_num_print == false ) {
if( rid == -1 ) {
printf("%s\t%d\t%s\t%s\t%f\t%s\t%s\t%d\t%d\t%c\t%c\t%c\t.\n", chr_name, b, ref, alt, qual, filter, exons[gid].name, exons[gid].reg.lower, exons[gid].reg.upper, exons[gid].dir, aa1, aa2);
}
else {
printf("%s\t%d\t%s\t%s\t%f\t%s\t%s\t%d\t%d\t%c\t%c\t%c\t%s\n", chr_name, b, ref, alt, qual, filter, exons[gid].name, exons[gid].reg.lower, exons[gid].reg.upper, exons[gid].dir, aa1, aa2, rmsk[rid].name);
}
}
}
else {
}
}
}
}
if( is_num_print == true ) {
printf("%s\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\n", chr_name, num_snps, num_pass, num_filter, num_coding, num_coding1, num_non, num_syn, num_non1, num_syn1, num_repeats, num_repeats1, num_coding_repeats, num_coding_repeats1);
}
if( num_cds > 0 ) {
free(exons);
}
fclose(f);
return EXIT_SUCCESS;
}
int driver (int t_num)
{
int i, j;
double current_seconds = 0;
long current_transactions = 0;
double current_tps = 0;
struct timeval currentTV;
/* Actually, WaitTimes are needed... */
while( activate_transaction ){
// tmc : add rate limiting
if (t_num == 0)
{
gettimeofday(¤tTV, NULL);
current_seconds = currentTV.tv_sec + (currentTV.tv_usec/1000000);
if (current_seconds > (rate_last_seconds + transaction_interval))
{
rate_last_seconds = current_seconds;
rate_last_transactions = success[0];
}
}
// check current rate and tarpit if necessary
gettimeofday(¤tTV, NULL);
current_seconds = currentTV.tv_sec + (currentTV.tv_usec/1000000);
current_transactions = success[0];
current_tps = ((current_transactions - rate_last_transactions) / (current_seconds - rate_last_seconds) * transaction_interval);
// if (t_num == 1)
// {
// printf("thread %d : tps = %f\n",t_num,current_tps);
// printf("thread %d : tps = %f : cur_secs = %f : last_secs = %f : cur_txn = %f : last_txn = %f\n",t_num,current_tps,current_seconds,rate_last_seconds,current_transactions,rate_last_transactions);
// }
while (current_tps > num_neword_per_10_sec)
{
usleep(100000);
gettimeofday(¤tTV, NULL);
current_seconds = currentTV.tv_sec + (currentTV.tv_usec/1000000);
current_transactions = success[0];
current_tps = ((current_transactions - rate_last_transactions) / (current_seconds - rate_last_seconds) * transaction_interval);
}
switch(seq_get()){
case 0:
do_neword(t_num);
break;
case 1:
do_payment(t_num);
break;
case 2:
do_ordstat(t_num);
break;
case 3:
do_delivery(t_num);
break;
case 4:
do_slev(t_num);
break;
default:
printf("Error - Unknown sequence.\n");
}
}
return(0);
}
bool c_seq::seq_exists(uint32 keyId)
{
string value;
return seq_get(keyId, value);
}
int main(int argc, char **argv) {
SEQ *sf;
uchar *s;
FILE *f;
char chr_name[100], info[1000], dir;
int i = 0, j = 0, k = 0, B = 0, E = 0;
int max_len = 0;
char *cur_seq;
int seq_len = 0;
bool is_correct_splicing = false;
int num_genes = 0;
int num_exons = 0;
int num = 0;
struct g_list *genes;
struct exons_list *exons;
bool no_branchpoint = false;
if( argc == 5 ) {
}
else if( argc == 4 ) {
if( strcmp( argv[3], "NO_BRANCHPOINT") == 0 ) {
no_branchpoint = true;
}
else {
fatalf("args: fasta gff (NO_BRANCHPOINT)");
}
}
else if (argc != 3)
fatalf("args: fasta gff (NO_BRANCHPOINT)");
if((f = ckopen(argv[2], "r")) == NULL )
{
fatalf("Cannot open file %s\n", argv[1]);
}
else {
num_genes = count_genes_in_gff(f, &num_exons);
if( num_genes > 0 ) {
genes = (struct g_list *) ckalloc(num_genes * sizeof(struct g_list));
if( num_exons < num_genes ) num_exons = num_genes;
exons = (struct exons_list *) ckalloc(num_exons * sizeof(struct exons_list));
initialize_genes(genes, num_genes);
initialize_exons(exons, num_exons);
}
}
fseek(f, 0, SEEK_SET);
branchpoints = (char **) ckalloc(sizeof(char *) * NUM_BP_SEQ);
for( i = 0; i < NUM_BP_SEQ; i++ ) {
branchpoints[i] = (char *) ckalloc(sizeof(char) * 8);
}
strcpy(branchpoints[0], "AACTAAC");
strcpy(branchpoints[1], "AATTAAC");
strcpy(branchpoints[2], "CACTAAC");
strcpy(branchpoints[3], "GACTAAC");
strcpy(branchpoints[4], "TACTAAC");
strcpy(branchpoints[5], "TACTAAT");
strcpy(branchpoints[6], "TATTAAC");
strcpy(branchpoints[7], "TGCTAAC");
strcpy(branchpoints[8], "GATTAAC");
num = input_genes_in_gff(f, genes, exons);
if( num != num_genes ) {
fatalf("gene counter error in %s\n", argv[1]);
}
if( num_genes > 0 ) {
quick_sort_inc_genes(genes, 0, num_genes-1, POS_BASE);
}
i = 0;
while( i < num_genes ) {
j = 0;
while( ((i+j) < num_genes) && (genes[i].txStart == genes[i+j].txStart )) j++;
quick_sort_dec_genes(genes, i, i+j-1, LEN_BASE);
i = i+j;
}
fclose(f);
compl['a'] = compl['A'] = 'T';
compl['c'] = compl['C'] = 'G';
compl['g'] = compl['G'] = 'C';
compl['t'] = compl['T'] = 'A';
sf = seq_get(argv[1]);
s = SEQ_CHARS(sf) - 1;
seq_len = SEQ_LEN(sf);
for( i = 0; i < num_genes; i++ ) {
B = genes[i].txStart;
E = genes[i].txEnd;
if( E > seq_len ) {
fatalf("gene boundary [%d,%d] over the sequence length %d\n", B, E, seq_len);
}
if( (E - B + 1) > max_len ) {
max_len = E - B + 1;
}
}
cur_seq = (char *) ckalloc(sizeof(char) * (max_len+1));
for( i = 0; i < num_genes; i++ ) {
if( genes[i].exonCount >= 2 ) {
strcpy(chr_name, genes[i].sname);
B = genes[i].txStart;
E = genes[i].txEnd;
dir = genes[i].strand;
strcpy(info, genes[i].gname);
k = 0;
if( dir == '+' ) {
for (j = B; j <= E; j++) {
cur_seq[k] = s[j];
k++;
}
cur_seq[k] = '\0';
}
else {
k = 0;
for (j = E; j >= B; j--) {
cur_seq[k] = compl[s[j]];
k++;
}
cur_seq[k] = '\0';
}
is_correct_splicing = true;
is_correct_splicing = check_introns(genes, i, exons, cur_seq, k);
if( is_correct_splicing == false ) {
if( no_branchpoint == false ) {
genes[i].type = REDUN;
}
}
else {
if( no_branchpoint == true ) {
genes[i].type = REDUN;
}
}
}
}
num_genes = rm_redun_genes(genes, 0, num_genes-1);
write_in_gff(genes, num_genes, exons, num_exons);
free(cur_seq);
for( i = 0; i < NUM_BP_SEQ; i++ ) free(branchpoints[i]);
free(branchpoints);
seq_close(sf);
return EXIT_SUCCESS;
}
int driver (int t_num)
{
int i, j;
#ifdef OUT_PERIOD
outputLog("d0|");
#endif
/* Actually, WaitTimes are needed... */
while( activate_transaction ){
switch(seq_get()){
case 0:
#ifdef OUT_PERIOD
outputLog("d1|");
#endif
do_neword(t_num);
#ifdef OUT_PERIOD
outputLog("d1e|");
#endif
break;
case 1:
#ifdef OUT_PERIOD
outputLog("d2|");
#endif
do_payment(t_num);
#ifdef OUT_PERIOD
outputLog("d2e|");
#endif
break;
case 2:
#ifdef OUT_PERIOD
outputLog("d3|");
#endif
do_ordstat(t_num);
#ifdef OUT_PERIOD
outputLog("d3e|");
#endif
break;
case 3:
#ifdef OUT_PERIOD
outputLog("d4|");
#endif
do_delivery(t_num);
#ifdef OUT_PERIOD
outputLog("d4e|");
#endif
break;
case 4:
#ifdef OUT_PERIOD
outputLog("d5|");
#endif
do_slev(t_num);
#ifdef OUT_PERIOD
outputLog("d5e|");
#endif
break;
default:
#ifdef OUT_PERIOD
outputLog("d6|");
#endif
printf("Error - Unknown sequence.\n");
}
}
return(0);
}
