Skip to content

C implementation of IBD-finding code. #679

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Sep 18, 2020
Merged

C implementation of IBD-finding code. #679

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
0068597
Basics of IBD-C procedure now implemented.
gtsambos Sep 17, 2020
fa237ca
Add low-level Python-C interface.
jeromekelleher Sep 2, 2020
5f882f8
Various changes suggested by Jerome to IBD-C code
gtsambos Sep 3, 2020
556e8b3
Tweaked C IBD API.
jeromekelleher Sep 11, 2020
b75d8c9
Simplified Python IBD tests and added a compare_lib option.
gtsambos Sep 13, 2020
732e37c
C and Python tests for algorithm pass now.
gtsambos Sep 14, 2020
f0679b0
Set default for max_time to DBL_MAX
jeromekelleher Sep 18, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
371 changes: 371 additions & 0 deletions c/tests/test_tables.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@
#include "testlib.h"
#include <tskit/tables.h>

#include <float.h>
#include <unistd.h>
#include <stdlib.h>

Expand Down Expand Up @@ -2851,6 +2852,367 @@ test_link_ancestors_multiple_to_single_tree(void)
tsk_treeseq_free(&ts);
}

/* Helper method for running IBD tests */
static int TSK_WARN_UNUSED
ibd_finder_init_and_run(tsk_ibd_finder_t *ibd_finder, tsk_table_collection_t *tables,
tsk_id_t *samples, tsk_size_t num_samples, double min_length, double max_time)
{
int ret = 0;

ret = tsk_ibd_finder_init(ibd_finder, tables, samples, num_samples);
if (ret != 0) {
goto out;
}
ret = tsk_ibd_finder_set_min_length(ibd_finder, min_length);
if (ret != 0) {
goto out;
}
ret = tsk_ibd_finder_set_max_time(ibd_finder, max_time);
if (ret != 0) {
goto out;
}
ret = tsk_ibd_finder_run(ibd_finder);
if (ret != 0) {
goto out;
}

out:
return ret;
}

static void
test_ibd_finder(void)
{
int ret;
int j, k;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1 };
tsk_ibd_finder_t ibd_finder;
double true_left[] = { 0.0 };
double true_right[] = { 1.0 };
tsk_id_t true_node[] = { 4 };
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL, NULL,
NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 1, 0.0, DBL_MAX);

// Check the output.
CU_ASSERT_EQUAL_FATAL(ret, 0);

for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_EQUAL_FATAL(ret, 0);
k = 0;
while (seg != NULL) {
CU_ASSERT_EQUAL_FATAL(seg->left, true_left[k]);
CU_ASSERT_EQUAL_FATAL(seg->right, true_right[k]);
CU_ASSERT_EQUAL_FATAL(seg->node, true_node[k]);
k++;
seg = seg->next;
}
}
tsk_ibd_finder_print_state(&ibd_finder, _devnull);

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_multiple_trees(void)
{
int ret;
int j, k;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1, 0, 2 };
tsk_ibd_finder_t ibd_finder;
double true_left[2][2] = { { 0.0, 0.7 }, { 0.7, 0.0 } };
double true_right[2][2] = { { 0.7, 1.0 }, { 1.0, 0.7 } };
double true_node[2][2] = { { 4, 5 }, { 5, 6 } };
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 2, multiple_tree_ex_nodes, multiple_tree_ex_edges, NULL,
NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Run ibd_finder.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 2, 0.0, DBL_MAX);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Check the output.
for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
ret = tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_EQUAL_FATAL(ret, 0);
k = 0;
while (seg != NULL) {
CU_ASSERT_EQUAL_FATAL(seg->left, true_left[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->right, true_right[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->node, true_node[j][k]);
k++;
seg = seg->next;
}
}

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_empty_result(void)
{
int ret;
int j;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1 };
tsk_ibd_finder_t ibd_finder;
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL, NULL,
NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Run ibd_finder.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 1, 0.0, 0.5);

// Check the output.
CU_ASSERT_EQUAL_FATAL(ret, 0);

for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_EQUAL_FATAL(ret, 0);
CU_ASSERT_EQUAL_FATAL(seg, NULL);
}
tsk_ibd_finder_print_state(&ibd_finder, _devnull);

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_min_length_max_time(void)
{
int ret;
int j, k;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1, 1, 2, 2, 0 };
tsk_ibd_finder_t ibd_finder;
double true_left[3][1] = { { 0.0 }, { -1 }, { -1 } };
double true_right[3][1] = { { 0.7 }, { -1 }, { -1 } };
double true_node[3][1] = { { 4 }, { -1 }, { -1 } };
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 2, multiple_tree_ex_nodes, multiple_tree_ex_edges, NULL,
NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Run ibd_finder.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 3, 0.5, 3.0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Check the output.
for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
ret = tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_TRUE_FATAL((ret == 0) || (ret == -1));
if (ret == -1) {
continue;
}
k = 0;
while (seg != NULL) {
CU_ASSERT_EQUAL_FATAL(seg->left, true_left[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->right, true_right[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->node, true_node[j][k]);
k++;
seg = seg->next;
}
}

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_errors(void)
{
int ret;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1, 2, 0 };
tsk_id_t samples2[] = { -1, 1 };
tsk_id_t samples3[] = { 0 };
tsk_ibd_finder_t ibd_finder;

tsk_treeseq_from_text(&ts, 2, multiple_tree_ex_nodes, multiple_tree_ex_edges, NULL,
NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Invalid sample IDs
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples2, 1, 0.0, DBL_MAX);
CU_ASSERT_EQUAL_FATAL(ret, TSK_ERR_NODE_OUT_OF_BOUNDS);
tsk_ibd_finder_free(&ibd_finder);

// Only 1 sample
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples3, 0, 0.0, DBL_MAX);
CU_ASSERT_EQUAL_FATAL(ret, TSK_ERR_NO_SAMPLE_PAIRS);
tsk_ibd_finder_free(&ibd_finder);

// Bad length or time
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 2, 0.0, -1);
CU_ASSERT_EQUAL_FATAL(ret, TSK_ERR_BAD_PARAM_VALUE);
tsk_ibd_finder_free(&ibd_finder);
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 2, -1, 0.0);
CU_ASSERT_EQUAL_FATAL(ret, TSK_ERR_BAD_PARAM_VALUE);
tsk_ibd_finder_free(&ibd_finder);

tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_samples_are_descendants(void)
{
int ret;
int j, k;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 2, 0, 4, 2, 4, 1, 3, 1, 5, 3, 5 };
tsk_ibd_finder_t ibd_finder;
double true_left[6][1] = { { 0.0 }, { 0.0 }, { 0.0 }, { 0.0 }, { 0.0 }, { 0.0 } };
double true_right[6][1] = { { 1.0 }, { 1.0 }, { 1.0 }, { 1.0 }, { 1.0 }, { 1.0 } };
tsk_id_t true_node[6][1] = { { 2 }, { 4 }, { 4 }, { 3 }, { 5 }, { 5 } };
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 1, multi_root_tree_ex_nodes, multi_root_tree_ex_edges,
NULL, NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Run ibd_finder.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 6, 0.0, DBL_MAX);

// Check the output.
CU_ASSERT_EQUAL_FATAL(ret, 0);

for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_EQUAL_FATAL(ret, 0);
k = 0;
while (seg != NULL) {
CU_ASSERT_EQUAL_FATAL(seg->left, true_left[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->right, true_right[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->node, true_node[j][k]);
k++;
seg = seg->next;
}
}
tsk_ibd_finder_print_state(&ibd_finder, _devnull);

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_multiple_ibd_paths(void)
{
int ret;
int j, k;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1, 0, 2, 1, 2 };
tsk_ibd_finder_t ibd_finder;
double true_left[3][2] = { { 0.2, 0.0 }, { 0.2, 0.0 }, { 0.0, 0.2 } };
double true_right[3][2] = { { 1.0, 0.2 }, { 1.0, 0.2 }, { 0.2, 1.0 } };
double true_node[3][2] = { { 4, 5 }, { 3, 5 }, { 4, 4 } };
tsk_segment_t *seg = NULL;

// Read in the tree sequence.
tsk_treeseq_from_text(&ts, 2, multi_path_tree_ex_nodes, multi_path_tree_ex_edges,
NULL, NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Run ibd_finder.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 3, 0.0, 0.0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Check the output.
CU_ASSERT_EQUAL_FATAL(ret, 0);

for (j = 0; j < (int) ibd_finder.num_pairs; j++) {
tsk_ibd_finder_get_ibd_segments(&ibd_finder, j, &seg);
CU_ASSERT_EQUAL_FATAL(ret, 0);
k = 0;
while (seg != NULL) {
CU_ASSERT_EQUAL_FATAL(seg->left, true_left[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->right, true_right[j][k]);
CU_ASSERT_EQUAL_FATAL(seg->node, true_node[j][k]);
k++;
seg = seg->next;
}
}
tsk_ibd_finder_print_state(&ibd_finder, _devnull);

// Free.
tsk_ibd_finder_free(&ibd_finder);
tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_ibd_finder_odd_topologies(void)
{
int ret;
// int j;
tsk_treeseq_t ts;
tsk_table_collection_t tables;
tsk_id_t samples[] = { 0, 1 };
tsk_id_t samples1[] = { 0, 2 };
tsk_ibd_finder_t ibd_finder;

tsk_treeseq_from_text(
&ts, 1, odd_tree1_ex_nodes, odd_tree1_ex_edges, NULL, NULL, NULL, NULL, NULL, 0);
ret = tsk_treeseq_copy_tables(&ts, &tables, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);

// Multiple roots.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples, 1, 0, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);
tsk_ibd_finder_free(&ibd_finder);

// // Parent is a sample.
ret = ibd_finder_init_and_run(&ibd_finder, &tables, samples1, 1, 0, 0);
CU_ASSERT_EQUAL_FATAL(ret, 0);
tsk_ibd_finder_free(&ibd_finder);

tsk_table_collection_free(&tables);
tsk_treeseq_free(&ts);
}

static void
test_simplify_tables_drops_indexes(void)
{
Expand Down Expand Up @@ -4627,6 +4989,15 @@ main(int argc, char **argv)
{ "test_link_ancestors_multiple_to_single_tree",
test_link_ancestors_multiple_to_single_tree },
{ "test_sort_tables_offsets", test_sort_tables_offsets },
{ "test_ibd_finder", test_ibd_finder },
{ "test_ibd_finder_multiple_trees", test_ibd_finder_multiple_trees },
{ "test_ibd_finder_empty_result", test_ibd_finder_empty_result },
{ "test_ibd_finder_min_length_max_time", test_ibd_finder_min_length_max_time },
{ "test_ibd_finder_samples_are_descendants",
test_ibd_finder_samples_are_descendants },
{ "test_ibd_finder_multiple_ibd_paths", test_ibd_finder_multiple_ibd_paths },
{ "test_ibd_finder_odd_topologies", test_ibd_finder_odd_topologies },
{ "test_ibd_finder_errors", test_ibd_finder_errors },
{ "test_sort_tables_drops_indexes", test_sort_tables_drops_indexes },
{ "test_sort_tables_edge_metadata", test_sort_tables_edge_metadata },
{ "test_sort_tables_no_edge_metadata", test_sort_tables_no_edge_metadata },
Expand Down
Loading