obitools/obitools3
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 66 Wiki Activity

Added the kmer filter to LCS alignments, and now obiblobs containing

Browse Source 
encoded sequences are directly put in int16_t arrays for the alignment
This commit is contained in:
Celine Mercier
2016-11-18 16:29:28 +01:00
parent 08e67a090f
commit 70c49e214a
17 changed files with 653 additions and 94 deletions
Download Patch File Download Diff File Expand all files Collapse all files

108
src/obi_align.c
View File

@ -19,6 +19,8 @@
#include "obitypes.h"
#include "obiview.h"
#include "sse_banded_LCS_alignment.h"
#include "upperband.h"
#include "obiblob.h"
#define DEBUG_LEVEL 0 // TODO has to be defined somewhere else (cython compil flag?)
@ -29,21 +31,22 @@
// option pour ecrire en stdint?
// check NUC_SEQS view type? and score type (int or float if normalize)
// what's with multiple sequences/line columns?
// make function that put blobs in int16
int obi_align_one_column(Obiview_p seq_view, OBIDMS_column_p seq_column,
Obiview_p score_view, OBIDMS_column_p id1_column, OBIDMS_column_p id2_column, OBIDMS_column_p score_column,
double threshold, bool normalize, int reference, bool similarity_mode)
{
index_t i, j, k;
index_t seq_count;
char* seq1;
char* seq2;
const char* id1;
const char* id2;
double score;
index_t i, j, k;
index_t seq_count;
const char* id1;
const char* id2;
double score;
OBIDMS_column_p id_column;
Kmer_table_p ktable;
Obi_blob_p blob1;
Obi_blob_p blob2;
int lcs_min;
k = 0;
@ -62,6 +65,15 @@ int obi_align_one_column(Obiview_p seq_view, OBIDMS_column_p seq_column,
return -1;
}
// Build kmer tables
ktable = hash_seq_column(seq_view, seq_column);
if (ktable == NULL)
{
obi_set_errno(OBI_ALIGN_ERROR);
obidebug(1, "\nError building kmer tables before aligning");
return -1;
}
// Get the ID column pointer
id_column = obi_view_get_column(seq_view, ID_COLUMN);
@ -69,66 +81,72 @@ int obi_align_one_column(Obiview_p seq_view, OBIDMS_column_p seq_column,
for (i=0; i < (seq_count - 1); i++)
{
if (i%100 == 0)
fprintf(stderr,"\rDone : %f %% ", (i / (float) seq_count)*100);
for (j=i+1; j < seq_count; j++)
{
//fprintf(stderr, "\ni=%lld, j=%lld, k=%lld", i, j, k);
blob1 = obi_get_blob_with_elt_idx_and_col_p_in_view(seq_view, seq_column, i, 0);
blob2 = obi_get_blob_with_elt_idx_and_col_p_in_view(seq_view, seq_column, j, 0);
seq1 = obi_get_seq_with_elt_idx_and_col_p_in_view(seq_view, seq_column, i, 0);
seq2 = obi_get_seq_with_elt_idx_and_col_p_in_view(seq_view, seq_column, j, 0);
if ((seq1 == NULL) || (seq2 == NULL))
if ((blob1 == NULL) || (blob2 == NULL))
{
obidebug(1, "\nError retrieving sequences to align");
return -1;
}
// TODO kmer filter
// kmer filter
align_filters(ktable, blob1, blob2, i, j, threshold, normalize, reference, similarity_mode, &score, &lcs_min);
// Compute alignment score
score = generic_sse_banded_lcs_align(seq1, seq2, threshold, normalize, reference, similarity_mode);
if ((threshold == 0) || (score == -1.0)) // no threshold or filter passed, and sequences not identical: align
score = obiblob_sse_banded_lcs_align(blob1, blob2, threshold, normalize, reference, similarity_mode);
// Get sequence ids
id1 = obi_get_str_with_elt_idx_and_col_p_in_view(seq_view, id_column, i, 0);
id2 = obi_get_str_with_elt_idx_and_col_p_in_view(seq_view, id_column, j, 0);
if ((score >= 0) && (((normalize || similarity_mode) && (score >= threshold)) || ((!similarity_mode && !normalize) && (score <= threshold))))
{ // Print result
// Write sequence ids in output view
if (obi_set_str_with_elt_idx_and_col_p_in_view(score_view, id1_column, k, 0, id1) < 0)
{
obidebug(1, "\nError writing id1 in a column");
return -1;
}
// Get sequence ids
id1 = obi_get_str_with_elt_idx_and_col_p_in_view(seq_view, id_column, i, 0);
id2 = obi_get_str_with_elt_idx_and_col_p_in_view(seq_view, id_column, j, 0);
if (obi_set_str_with_elt_idx_and_col_p_in_view(score_view, id2_column, k, 0, id2) < 0)
{
obidebug(1, "\nError writing id2 in a column");
return -1;
}
// Write score in output view
if (normalize)
{
if (obi_set_float_with_elt_idx_and_col_p_in_view(score_view, score_column, k, 0, (obifloat_t) score) < 0)
// Write sequence ids in output view
if (obi_set_str_with_elt_idx_and_col_p_in_view(score_view, id1_column, k, 0, id1) < 0)
{
obidebug(1, "\nError writing alignment score in a column");
obidebug(1, "\nError writing id1 in a column");
return -1;
}
}
else
{
if (obi_set_int_with_elt_idx_and_col_p_in_view(score_view, score_column, k, 0, (obiint_t) score) < 0)
if (obi_set_str_with_elt_idx_and_col_p_in_view(score_view, id2_column, k, 0, id2) < 0)
{
obidebug(1, "\nError writing alignment score in a column");
obidebug(1, "\nError writing id2 in a column");
return -1;
}
// Write score in output view
if (normalize)
{
if (obi_set_float_with_elt_idx_and_col_p_in_view(score_view, score_column, k, 0, (obifloat_t) score) < 0)
{
obidebug(1, "\nError writing alignment score in a column");
return -1;
}
}
else
{
if (obi_set_int_with_elt_idx_and_col_p_in_view(score_view, score_column, k, 0, (obiint_t) score) < 0)
{
obidebug(1, "\nError writing alignment score in a column");
return -1;
}
}
k++;
}
free(seq1);
free(seq2);
k++;
}
}
free_kmer_tables(ktable, seq_count);
return 0;
}