Skip to content

32bit genotypes #2108

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 1 commit into from
Mar 7, 2022
Merged

32bit genotypes #2108

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
9 changes: 8 additions & 1 deletion c/CHANGELOG.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,14 @@
[0.99.16] - 2022-0X-XX
----------------------

- Make dumping of tables and tree seqences to disk a zero-copy operation.
**Breaking changes**

- Change the type of genotypes to ``int32_t``, removing the TSK_16_BIT_GENOTYPES flag option.
(:user:`benjeffery`, :issue:`463`, :pr:`2108`)

**Features**

- Make dumping of tables and tree sequences to disk a zero-copy operation.
(:user:`benjeffery`, :issue:`2111`, :pr:`2124`)

----------------------
Expand Down
370 changes: 123 additions & 247 deletions c/tests/test_genotypes.c
View file
Open in desktop

Large diffs are not rendered by default.

20 changes: 10 additions & 10 deletions c/tests/test_haplotype_matching.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ tsk_ls_hmm_next_probability_test(tsk_ls_hmm_t *TSK_UNUSED(self),
}

static int
run_test_hmm(tsk_ls_hmm_t *hmm, int8_t *haplotype, tsk_compressed_matrix_t *output)
run_test_hmm(tsk_ls_hmm_t *hmm, int32_t *haplotype, tsk_compressed_matrix_t *output)
{
int ret = 0;

Expand Down Expand Up @@ -79,7 +79,7 @@ test_single_tree_missing_alleles(void)

double rho[] = { 0, 0.25, 0.25 };
double mu[] = { 0.125, 0.125, 0.125 };
int8_t h[] = { 0, 0, 0, 0 };
int32_t h[] = { 0, 0, 0, 0 };

tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL,
single_tree_ex_sites, single_tree_ex_mutations, NULL, NULL, 0);
Expand Down Expand Up @@ -108,7 +108,7 @@ test_single_tree_exact_match(void)

double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
int8_t h[] = { 1, 1, 1 };
int32_t h[] = { 1, 1, 1 };
tsk_id_t path[3];
double decoded_compressed_matrix[12];
unsigned int precision;
Expand Down Expand Up @@ -167,7 +167,7 @@ test_single_tree_missing_haplotype_data(void)

double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
int8_t h[] = { 1, TSK_MISSING_DATA, 1 };
int32_t h[] = { 1, TSK_MISSING_DATA, 1 };
tsk_id_t path[3];
double decoded_compressed_matrix[12];

Expand Down Expand Up @@ -211,7 +211,7 @@ test_single_tree_match_impossible(void)
double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
/* This haplotype can't happen with a mutation rate of 0 */
int8_t h[] = { 0, 0, 0 };
int32_t h[] = { 0, 0, 0 };

tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL,
single_tree_ex_sites, single_tree_ex_mutations, NULL, NULL, 0);
Expand Down Expand Up @@ -247,7 +247,7 @@ test_single_tree_errors(void)

double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
int8_t h[] = { 0, 0, 0 };
int32_t h[] = { 0, 0, 0 };

tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL,
single_tree_ex_sites, single_tree_ex_mutations, NULL, NULL, 0);
Expand Down Expand Up @@ -309,7 +309,7 @@ test_single_tree_compressed_matrix(void)

double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0.1, 0.1, 0.1 };
int8_t h[] = { 0, 0, 0 };
int32_t h[] = { 0, 0, 0 };

tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL,
single_tree_ex_sites, single_tree_ex_mutations, NULL, NULL, 0);
Expand Down Expand Up @@ -374,7 +374,7 @@ test_single_tree_viterbi_matrix(void)
tsk_ls_hmm_t ls_hmm;
double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
int8_t h[] = { 1, 1, 1 };
int32_t h[] = { 1, 1, 1 };
tsk_id_t path[3];
tsk_value_transition_t T[2];
int j;
Expand Down Expand Up @@ -448,7 +448,7 @@ test_multi_tree_exact_match(void)

double rho[] = { 0.0, 0.25, 0.25 };
double mu[] = { 0, 0, 0 };
int8_t h[] = { 1, 1, 1 };
int32_t h[] = { 1, 1, 1 };
tsk_id_t path[3];
double decoded_compressed_matrix[12];
unsigned int precision;
Expand Down Expand Up @@ -530,7 +530,7 @@ test_caterpillar_tree_many_values(void)
tsk_ls_hmm_t ls_hmm;
tsk_compressed_matrix_t matrix;
double unused[] = { 0, 0, 0, 0, 0 };
int8_t h[] = { 0, 0, 0, 0, 0 };
int32_t h[] = { 0, 0, 0, 0, 0 };
tsk_size_t n[] = {
8,
16,
Expand Down
64 changes: 32 additions & 32 deletions c/tests/test_trees.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -486,7 +486,7 @@ verify_simplify_genotypes(tsk_treeseq_t *ts, tsk_treeseq_t *subset,
tsk_vargen_t vargen, subset_vargen;
tsk_variant_t *variant, *subset_variant;
tsk_size_t j, k;
int8_t a1, a2;
int32_t a1, a2;
const tsk_id_t *sample_index_map;

sample_index_map = tsk_treeseq_get_sample_index_map(ts);
Expand All @@ -511,8 +511,8 @@ verify_simplify_genotypes(tsk_treeseq_t *ts, tsk_treeseq_t *subset,
CU_ASSERT_EQUAL(variant->site->position, subset_variant->site->position);
for (k = 0; k < num_samples; k++) {
CU_ASSERT_FATAL(sample_index_map[samples[k]] < (tsk_id_t) ts->num_samples);
a1 = variant->genotypes.i8[sample_index_map[samples[k]]];
a2 = subset_variant->genotypes.i8[k];
a1 = variant->genotypes[sample_index_map[samples[k]]];
a2 = subset_variant->genotypes[k];
/* printf("a1 = %d, a2 = %d\n", a1, a2); */
/* printf("k = %d original node = %d " */
/* "original_index = %d a1=%.*s a2=%.*s\n", */
Expand Down Expand Up @@ -1330,29 +1330,29 @@ test_simplest_non_sample_leaf_records(void)
CU_ASSERT_EQUAL_FATAL(ret, 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[0], "0", 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[1], "1", 1);
CU_ASSERT_EQUAL(var->genotypes.i8[0], 1);
CU_ASSERT_EQUAL(var->genotypes.i8[1], 0);
CU_ASSERT_EQUAL(var->genotypes[0], 1);
CU_ASSERT_EQUAL(var->genotypes[1], 0);

ret = tsk_vargen_next(&vargen, &var);
CU_ASSERT_EQUAL_FATAL(ret, 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[0], "0", 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[1], "1", 1);
CU_ASSERT_EQUAL(var->genotypes.i8[0], 0);
CU_ASSERT_EQUAL(var->genotypes.i8[1], 1);
CU_ASSERT_EQUAL(var->genotypes[0], 0);
CU_ASSERT_EQUAL(var->genotypes[1], 1);

ret = tsk_vargen_next(&vargen, &var);
CU_ASSERT_EQUAL_FATAL(ret, 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[0], "0", 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[1], "1", 1);
CU_ASSERT_EQUAL(var->genotypes.i8[0], 0);
CU_ASSERT_EQUAL(var->genotypes.i8[1], 0);
CU_ASSERT_EQUAL(var->genotypes[0], 0);
CU_ASSERT_EQUAL(var->genotypes[1], 0);

ret = tsk_vargen_next(&vargen, &var);
CU_ASSERT_EQUAL_FATAL(ret, 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[0], "0", 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[1], "1", 1);
CU_ASSERT_EQUAL(var->genotypes.i8[0], 0);
CU_ASSERT_EQUAL(var->genotypes.i8[1], 0);
CU_ASSERT_EQUAL(var->genotypes[0], 0);
CU_ASSERT_EQUAL(var->genotypes[1], 0);

ret = tsk_vargen_next(&vargen, &var);
CU_ASSERT_EQUAL_FATAL(ret, 0);
Expand Down Expand Up @@ -1659,9 +1659,9 @@ test_simplest_back_mutations(void)
CU_ASSERT_EQUAL(var->num_alleles, 2);
CU_ASSERT_NSTRING_EQUAL(var->alleles[0], "0", 1);
CU_ASSERT_NSTRING_EQUAL(var->alleles[1], "1", 1);
CU_ASSERT_EQUAL(var->genotypes.i8[0], 0);
CU_ASSERT_EQUAL(var->genotypes.i8[1], 1);
CU_ASSERT_EQUAL(var->genotypes.i8[2], 0);
CU_ASSERT_EQUAL(var->genotypes[0], 0);
CU_ASSERT_EQUAL(var->genotypes[1], 1);
CU_ASSERT_EQUAL(var->genotypes[2], 0);
CU_ASSERT_EQUAL(var->site->id, 0);
CU_ASSERT_EQUAL(var->site->mutations_length, 2);
tsk_vargen_free(&vargen);
Expand Down Expand Up @@ -3069,10 +3069,10 @@ test_simplest_map_mutations(void)
const char *edges = "0 1 2 0,1\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0 };
int32_t genotypes[] = { 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3151,10 +3151,10 @@ test_simplest_nonbinary_map_mutations(void)
const char *edges = "0 1 4 0,1,2,3\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0, 0, 0 };
int32_t genotypes[] = { 0, 0, 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3197,10 +3197,10 @@ test_simplest_unary_map_mutations(void)
"0 1 4 2,3\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0 };
int32_t genotypes[] = { 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3241,10 +3241,10 @@ test_simplest_non_sample_leaf_map_mutations(void)
const char *edges = "0 1 2 0,1,3,4\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0 };
int32_t genotypes[] = { 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3283,10 +3283,10 @@ test_simplest_internal_sample_map_mutations(void)
const char *edges = "0 1 2 0,1\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0, 0 };
int32_t genotypes[] = { 0, 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3340,10 +3340,10 @@ test_simplest_multiple_root_map_mutations(void)
"0 1 5 2,3\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0, 0, 0 };
int32_t genotypes[] = { 0, 0, 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -3397,10 +3397,10 @@ test_simplest_chained_map_mutations(void)
"0 1 4 2,3\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 0, 0, 0 };
int32_t genotypes[] = { 0, 0, 0, 0 };
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;
int ret;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -4543,11 +4543,11 @@ test_single_tree_map_mutations(void)
{
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 1, 1, 1 };
int32_t genotypes[] = { 0, 1, 1, 1 };
int ret = 0;
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state, j;
int32_t ancestral_state, j;

tsk_treeseq_from_text(&ts, 1, single_tree_ex_nodes, single_tree_ex_edges, NULL, NULL,
NULL, NULL, NULL, 0);
Expand Down Expand Up @@ -4689,11 +4689,11 @@ test_single_tree_map_mutations_internal_samples(void)
"0.00000000 1.00000000 8 7\n";
tsk_treeseq_t ts;
tsk_tree_t t;
int8_t genotypes[] = { 0, 2, 2, 1, 0 };
int32_t genotypes[] = { 0, 2, 2, 1, 0 };
int ret = 0;
tsk_size_t num_transitions;
tsk_state_transition_t *transitions;
int8_t ancestral_state;
int32_t ancestral_state;

tsk_treeseq_from_text(&ts, 1, nodes, edges, NULL, NULL, NULL, NULL, NULL, 0);
CU_ASSERT_EQUAL(tsk_treeseq_get_num_samples(&ts), 5);
Expand Down
Loading