/*
* Get position after current segment. A segement is a stretch of kmers aligned
* to the graph with no gaps.
* @param aln alignment of read to the graph
* @param start offset in the alignment that starts this segments
* @param missing_edge set to 1 if the segment was ended due to a missing edge
* @return index of node after next gap or aln->nodes.len if no more gaps
*/
size_t db_alignment_next_gap(const dBAlignment *aln, size_t start,
bool *missing_edge, const dBGraph *db_graph)
{
size_t i, end = aln->rpos.len;
const int32_t *rpos = aln->rpos.b;
const dBNode *nodes = aln->nodes.b;
int colour = aln->colour;
Edges edges;
Nucleotide nuc;
*missing_edge = false;
if(end == 0) return 0;
// Set upper bound on position of the next gap as within this read
if(aln->used_r1 && aln->used_r2 && start < aln->r2strtidx)
end = aln->r2strtidx;
for(i = start+1; i < end && rpos[i-1]+1 == rpos[i]; i++)
{
// Check for a missing edge
edges = colour < 0 ? db_node_get_edges_union(db_graph, nodes[i-1].key)
: db_node_get_edges(db_graph, nodes[i-1].key, colour);
nuc = db_node_get_last_nuc(nodes[i], db_graph);
if(!edges_has_edge(edges, nuc, nodes[i-1].orient)) {
*missing_edge = true;
break;
}
}
// Return position after gap
return i;
}
// Do not print first k-1 bases => 3 nodes gives 3bp instead of 3+k-1
void db_nodes_gzprint_cont(const dBNode *nodes, size_t num,
const dBGraph *db_graph, gzFile out)
{
size_t i;
Nucleotide nuc;
for(i = 0; i < num; i++) {
nuc = db_node_get_last_nuc(nodes[i], db_graph);
gzputc(out, dna_nuc_to_char(nuc));
}
}
void db_nodes_gzprint(const dBNode *nodes, size_t num,
const dBGraph *db_graph, gzFile out)
{
size_t i, kmer_size = db_graph->kmer_size;
Nucleotide nuc;
BinaryKmer bkmer;
char tmp[MAX_KMER_SIZE+1];
bkmer = db_node_oriented_bkmer(db_graph, nodes[0]);
binary_kmer_to_str(bkmer, kmer_size, tmp);
gzputs(out, tmp);
for(i = 1; i < num; i++) {
nuc = db_node_get_last_nuc(nodes[i], db_graph);
gzputc(out, dna_nuc_to_char(nuc));
}
}
void db_nodes_print(const dBNode *nodes, size_t num,
const dBGraph *db_graph, FILE *out)
{
const size_t kmer_size = db_graph->kmer_size;
size_t i;
Nucleotide nuc;
BinaryKmer bkmer;
char tmp[MAX_KMER_SIZE+1];
bkmer = db_node_oriented_bkmer(db_graph, nodes[0]);
binary_kmer_to_str(bkmer, kmer_size, tmp);
fputs(tmp, out);
for(i = 1; i < num; i++) {
nuc = db_node_get_last_nuc(nodes[i], db_graph);
fputc(dna_nuc_to_char(nuc), out);
}
}
// Returns number of bytes added
size_t db_nodes_to_str(const dBNode *nodes, size_t num,
const dBGraph *db_graph, char *str)
{
if(num == 0) return 0;
size_t i;
size_t kmer_size = db_graph->kmer_size;
BinaryKmer bkmer = db_node_get_bkmer(db_graph, nodes[0].key);
Nucleotide nuc;
binary_kmer_to_str(bkmer, kmer_size, str);
if(nodes[0].orient == REVERSE) dna_reverse_complement_str(str, kmer_size);
for(i = 1; i < num; i++) {
nuc = db_node_get_last_nuc(nodes[i], db_graph);
str[kmer_size+i-1] = dna_nuc_to_char(nuc);
}
str[kmer_size+num-1] = '\0';
return kmer_size+num-1;
}
static void branch_to_str(const dBNode *nodes, size_t len, bool print_first_kmer,
StrBuf *sbuf, const dBGraph *db_graph)
{
size_t i = print_first_kmer, kmer_size = db_graph->kmer_size;
Nucleotide nuc;
BinaryKmer bkmer;
if(print_first_kmer) {
strbuf_ensure_capacity(sbuf, sbuf->end + kmer_size);
bkmer = db_node_oriented_bkmer(db_graph, nodes[0]);
binary_kmer_to_str(bkmer, kmer_size, sbuf->b+sbuf->end);
sbuf->end += kmer_size;
}
// i == 1 if print_first_kmer, otherwise 0
strbuf_ensure_capacity(sbuf, sbuf->end + len + 1); // +1 for '\n'
for(; i < len; i++) {
nuc = db_node_get_last_nuc(nodes[i], db_graph);
sbuf->b[sbuf->end++] = dna_nuc_to_char(nuc);
}
sbuf->b[sbuf->end++] = '\n';
sbuf->b[sbuf->end] = '\0';
}