nt main(int argc, char *argv[]) {
FILE *genome_f;
int read_size=0; //the size of the reads
int table_factor; //arbitrary
int num_of_reads;
int bf_table_size = table_factor*num_of_reads;
int num_of_hash_func;
BloomFilter* bf_unique; //BF for the unique tries
// char* output_label=(char *)malloc(50); //label name for the output files
hattrie_t* trie_unique; //hattrie that holds the unique reads
hattrie_t* trie_repeat; //hattrie that holds the repetetive reads, and the one that has N inside of them.
hattrie_t* trie_genome_unique; //put 'accepts' (everything that uniqe BF says yes that it's in genome) into a trie
hattrie_t* trie_fp; //triw that holds false negatives set
hattrie_t* trie_fn;//trie that holds false positives set
// f = fopen(argv[1], "r");
trie_repeat = hattrie_create();
trie_unique = hattrie_create();
make_repeat_and_unique_tries((argv[1], "r"), trie_unique, trie_repeat);
table_factor = 10; //arbitrary
num_of_reads = line_number/2;
bf_table_size = table_factor*num_of_reads;
num_of_hash_func = (int) ceil(table_factor*0.69314);
//print the keys of the uniqe and repaet tries
hattrie_iteration(trie_unique, "unique", argv[3]);
hattrie_iteration(trie_repeat, "repeat", argv[3]);
//hashing uniqe reads trie using bloom filter
bf_unique = bloom_filter_new(bf_table_size, string_hash, num_of_hash_func);
hash_trie_into_bf(trie_unique, bf_unique);
check_if_trie_in_bf(trie_unique, bf_unique);
//create trie for all of the sliding windows in the genome reference which are in the unique reads according to the bf_unique
read_size = size-2;
trie_genome_unique = hattrie_create();
query_bf_with_genome(bf_unique, genome_f ,trie_genome_unique, read_size);
check_if_trie_in_bf(trie_unique, bf_unique);
fclose(genome_f);
hattrie_iteration(trie_genome_unique, "genome_unique", argv[3]);
trie_fp = hattrie_create();
trie_fn = hattrie_create();
printf("start checking for false positive \n");
check_fp(trie_unique,trie_genome_unique, trie_fp);
hattrie_iteration(trie_fp, "fp_unique", argv[3]);
printf("start checking for false negative \n");
check_fn(trie_unique,trie_genome_unique, trie_fn);
hattrie_iteration(trie_fn, "fn_unique", argv[3]);
bloom_filter_free(bf_unique);
free(buffer);
hattrie_free(trie_unique);
hattrie_free(trie_repeat);
hattrie_free(trie_genome_unique);
hattrie_free(trie_fn);
hattrie_free(trie_fp);
return 0;
}
int main()
{
hattrie_t* T = hattrie_create();
const size_t n = 1000000; // how many strings
const size_t m_low = 50; // minimum length of each string
const size_t m_high = 500; // maximum length of each string
char x[501];
size_t i, m;
for (i = 0; i < n; ++i) {
m = m_low + rand() % (m_high - m_low);
randstr(x, m);
*hattrie_get(T, x, m) = 1;
}
hattrie_iter_t* it;
clock_t t0, t;
const size_t repetitions = 100;
size_t r;
/* iterate in unsorted order */
fprintf(stderr, "iterating out of order ... ");
t0 = clock();
for (r = 0; r < repetitions; ++r) {
it = hattrie_iter_begin(T, false);
while (!hattrie_iter_finished(it)) {
hattrie_iter_next(it);
}
hattrie_iter_free(it);
}
t = clock();
fprintf(stderr, "finished. (%0.2f seconds)\n", (double) (t - t0) / (double) CLOCKS_PER_SEC);
/* iterate in sorted order */
fprintf(stderr, "iterating in order ... ");
t0 = clock();
for (r = 0; r < repetitions; ++r) {
it = hattrie_iter_begin(T, true);
while (!hattrie_iter_finished(it)) {
hattrie_iter_next(it);
}
hattrie_iter_free(it);
}
t = clock();
fprintf(stderr, "finished. (%0.2f seconds)\n", (double) (t - t0) / (double) CLOCKS_PER_SEC);
hattrie_free(T);
return 0;
}
void test_trie_non_ascii()
{
fprintf(stderr, "checking non-ascii... \n");
value_t* u;
hattrie_t* T = hattrie_create();
char* txt = "\x81\x70";
u = hattrie_get(T, txt, strlen(txt));
*u = 10;
u = hattrie_tryget(T, txt, strlen(txt));
if (*u != 10) {
fprintf(stderr, "can't store non-ascii strings\n");
}
hattrie_free(T);
fprintf(stderr, "done.\n");
}
void setup()
{
fprintf(stderr, "generating %zu keys ... ", n);
xs = malloc(n * sizeof(char*));
ds = malloc(d * sizeof(char*));
size_t i;
size_t m;
for (i = 0; i < n; ++i) {
m = m_low + rand() % (m_high - m_low);
xs[i] = malloc(m + 1);
randstr(xs[i], m);
}
for (i = 0; i < d; ++i) {
m = rand()%n;
ds[i] = xs[m];
}
T = hattrie_create();
M = str_map_create();
fprintf(stderr, "done.\n");
}
zone_tree_t* zone_tree_create()
{
return hattrie_create();
}
int main(int argc, char *argv[])
{
plan_lazy();
/* Random keys. */
srand(time(NULL));
unsigned key_count = 100000;
char **keys = malloc(sizeof(char*) * key_count);
for (unsigned i = 0; i < key_count; ++i) {
keys[i] = str_key_rand(KEY_MAXLEN);
}
/* Sort random keys. */
str_key_sort(keys, key_count);
/* Create trie */
value_t *val = NULL;
hattrie_t *trie = hattrie_create();
ok(trie != NULL, "hattrie: create");
/* Insert keys */
bool passed = true;
size_t inserted = 0;
for (unsigned i = 0; i < key_count; ++i) {
val = hattrie_get(trie, keys[i], strlen(keys[i]) + 1);
if (!val) {
passed = false;
break;
}
if (*val == NULL) {
*val = keys[i];
++inserted;
}
}
ok(passed, "hattrie: insert");
/* Check total insertions against trie weight. */
is_int(hattrie_weight(trie), inserted, "hattrie: trie weight matches insertions");
/* Build order-index. */
hattrie_build_index(trie);
/* Lookup all keys */
passed = true;
for (unsigned i = 0; i < key_count; ++i) {
val = hattrie_tryget(trie, keys[i], strlen(keys[i]) + 1);
if (val && (*val == keys[i] || strcmp(*val, keys[i]) == 0)) {
continue;
} else {
diag("hattrie: mismatch on element '%u'", i);
passed = false;
break;
}
}
ok(passed, "hattrie: lookup all keys");
/* Lesser or equal lookup. */
passed = true;
for (unsigned i = 0; i < key_count; ++i) {
if (!str_key_find_leq(trie, keys, i, key_count)) {
passed = false;
for (int off = -10; off < 10; ++off) {
int k = (int)i + off;
if (k < 0 || k >= key_count) {
continue;
}
diag("[%u/%d]: %s%s", i, off, off == 0?">":"",keys[k]);
}
break;
}
}
ok(passed, "hattrie: find lesser or equal for all keys");
/* Next lookup. */
passed = true;
for (unsigned i = 0; i < key_count - 1 && passed; ++i) {
value_t *val;
hattrie_find_next(trie, keys[i], strlen(keys[i]), &val);
passed = val && *val == (void *)keys[(i + 1)];
}
ok(passed, "hattrie: find next for all keys");
/* Unsorted iteration */
size_t iterated = 0;
hattrie_iter_t *it = hattrie_iter_begin(trie, false);
while (!hattrie_iter_finished(it)) {
++iterated;
hattrie_iter_next(it);
}
is_int(inserted, iterated, "hattrie: unsorted iteration");
hattrie_iter_free(it);
/* Sorted iteration. */
char key_buf[KEY_MAXLEN] = {'\0'};
iterated = 0;
it = hattrie_iter_begin(trie, true);
while (!hattrie_iter_finished(it)) {
size_t cur_key_len = 0;
const char *cur_key = hattrie_iter_key(it, &cur_key_len);
if (iterated > 0) { /* Only if previous exists. */
if (strcmp(key_buf, cur_key) > 0) {
diag("'%s' <= '%s' FAIL\n", key_buf, cur_key);
break;
}
}
++iterated;
memcpy(key_buf, cur_key, cur_key_len);
hattrie_iter_next(it);
}
is_int(inserted, iterated, "hattrie: sorted iteration");
hattrie_iter_free(it);
/* Cleanup */
for (unsigned i = 0; i < key_count; ++i) {
free(keys[i]);
}
free(keys);
hattrie_free(trie);
return 0;
}
int main(int argc, char* argv[])
{
if (argc < 2) {
fprintf(stderr, "Usage: bam-summarize reads.bam\n");
exit(EXIT_FAILURE);
}
samfile_t* f = samopen(argv[1], "rb", NULL);
if (f == NULL) {
fprintf(stderr, "can't open bam file %s\n", argv[1]);
exit(1);
}
bam1_t* b = bam_init1();
hattrie_t* T = hattrie_create();
char* qname = NULL;
size_t qname_size = 0;
size_t j, n = 0;
uint32_t* cigar;
uint32_t cigar_op, cigar_len;
read_stat_t** val;
while (samread(f, b) >= 0) {
if (++n % 1000000 == 0) {
fprintf(stderr, "\t%zu alignments\n", n);
}
bool perfect = true;
bool spliced = false;
bool gapped = false;
cigar = bam1_cigar(b);
for (j = 0; j < b->core.n_cigar; ++j) {
cigar_op = cigar[j] & BAM_CIGAR_MASK;
cigar_len = cigar[j] >> BAM_CIGAR_SHIFT;
if (cigar_op == BAM_CREF_SKIP) {
if (cigar_len < min_splice_length) gapped = true;
else spliced = true;
}
else if (cigar_op != BAM_CMATCH) perfect = false;
if (cigar_op == BAM_CSOFT_CLIP || cigar_op == BAM_CHARD_CLIP) break;
}
/* Skip any clipped alignments. We don't want your kind! */
if (cigar_op == BAM_CSOFT_CLIP || cigar_op == BAM_CHARD_CLIP) continue;
/* Hack the read to include mate information. */
if (b->core.flag & BAM_FPAIRED) {
if (qname_size < b->core.l_qname + 3) {
qname_size = b->core.l_qname + 3;
qname = realloc(qname, qname_size);
}
memcpy(qname, bam1_qname(b), b->core.l_qname);
if (b->core.flag & BAM_FREAD1) {
qname[b->core.l_qname] = '/';
qname[b->core.l_qname + 1] = '2';
qname[b->core.l_qname + 2] = '\0';
}
else {
qname[b->core.l_qname] = '/';
qname[b->core.l_qname + 1] = '1';
qname[b->core.l_qname + 2] = '\0';
}
val = (read_stat_t**) hattrie_get(T, qname, b->core.l_qname + 2);
}
else {
val = (read_stat_t**) hattrie_get(T, bam1_qname(b), b->core.l_qname);
}
if (*val == NULL) {
*val = malloc(sizeof(read_stat_t));
memset(*val, 0, sizeof(read_stat_t));
}
(*val)->aln_count++;
if (perfect) {
if (spliced) (*val)->spliced_perfect_cnt++;
else (*val)->unspliced_perfect_cnt++;
}
if (spliced) (*val)->spliced_cnt++;
if (gapped) (*val)->gapped_cnt++;
}
printf("alignment_count\t%zu\n", n);
printf("read_count\t%zu\n", hattrie_size(T));
/* print stats from the table */
uint32_t multi_count = 0;
uint32_t unspliced_perfect_cnt = 0;
uint32_t spliced_perfect_cnt = 0;
uint32_t spliced_cnt = 0;
uint32_t gapped_cnt = 0;
/* excluding multireads */
uint32_t unique_unspliced_perfect_cnt = 0;
uint32_t unique_spliced_perfect_cnt = 0;
uint32_t unique_spliced_cnt = 0;
uint32_t unique_gapped_cnt = 0;
hattrie_iter_t* i;
for (i = hattrie_iter_begin(T);
!hattrie_iter_finished(i);
hattrie_iter_next(i))
{
val = (read_stat_t**) hattrie_iter_val(i);
if ((*val)->aln_count == 1) {
unique_unspliced_perfect_cnt += (*val)->unspliced_perfect_cnt;
unique_spliced_perfect_cnt += (*val)->spliced_perfect_cnt;
unique_spliced_cnt += (*val)->spliced_cnt;
unique_gapped_cnt += (*val)->gapped_cnt;
}
else multi_count++;
unspliced_perfect_cnt += (*val)->unspliced_perfect_cnt;
spliced_perfect_cnt += (*val)->spliced_perfect_cnt;
spliced_cnt += (*val)->spliced_cnt;
gapped_cnt += (*val)->gapped_cnt;
}
hattrie_iter_free(i);
printf("multi_count\t%u\n", multi_count);
printf("unspliced_perfect_cnt\t%u\n", unspliced_perfect_cnt);
printf("spliced_perfect_cnt\t%u\n", spliced_perfect_cnt);
printf("spliced_cnt\t%u\n", spliced_cnt);
printf("gapped_cnt\t%u\n", gapped_cnt);
printf("unique_unspliced_perfect_cnt\t%u\n", unique_unspliced_perfect_cnt);
printf("unique_spliced_perfect_cnt\t%u\n", unique_spliced_perfect_cnt);
printf("unique_spliced_cnt\t%u\n", unique_spliced_cnt);
printf("unique_gapped_cnt\t%u\n", unique_gapped_cnt);
/* free the table */
for (i = hattrie_iter_begin(T);
!hattrie_iter_finished(i);
hattrie_iter_next(i))
{
free(* (read_stat_t**) hattrie_iter_val(i));
}
hattrie_iter_free(i);
hattrie_free(T);
free(qname);
bam_destroy1(b);
return 0;
}
AlnIndex::AlnIndex()
{
t = hattrie_create();
}
knot_zone_tree_t* knot_zone_tree_create()
{
return hattrie_create();
}
