/**
* Returns the value of the named attribute as nucleotide identified
*/
unsigned char seqfeat_get_attrib_nuc(const SeqFeature *sf, const char *name) {
char *str;
str = seqfeat_get_attrib_str(sf, name);
if(strlen(str) != 1) {
g_error("%s:%d: attribute '%s' value '%s' does not look like nucleotide",
__FILE__, __LINE__, name, str);
}
return nuc_char_to_id(str[0]);
}
/**
* Returns the degeneracy of the provided position in the provided
* codon. The provided codon should be an array of length 3 containing
* NUC identifiers representing nucleotides (defined in nuc.h) The idx
* (codon position) argument must be either 0,1, or 2. The degeneracy
* is how many of the 3 possible nucleotide changes at the given
* position would result in a synonymous change (either 0, 1, 2, or
* 3). 0 is returned if the codon contains an ambiguity nucleotide
* (e.g. an N).
*/
int aa_codon_degeneracy(const unsigned char *codon, const int idx) {
int change_count;
int nuc_id, aa_id, new_aa_id, i;
unsigned char new_codon[3];
if((idx < 0) || (idx > 2)) {
g_error("%s:%d: codon idx must be 0, 1 or 2", __FILE__, __LINE__);
}
nuc_id = nuc_char_to_id(codon[idx]);
aa_id = aa_codon_to_id(codon);
/* this is an ambiguous amino acid, just return 0 */
if(aa_id == AA_X) {
return 0;
}
/* Try every other nucleotide at the specified position
* Count when amino acid is different.
*/
new_codon[0] = codon[0];
new_codon[1] = codon[1];
new_codon[2] = codon[2];
change_count = 0;
for(i = 0; i < NUM_REAL_NUCS; i++) {
if(i == nuc_id) {
continue;
}
new_codon[idx] = i;
new_aa_id = aa_codon_to_id(new_codon);
if(new_aa_id == AA_X) {
/* too complicated to handle codons with ambiguity symbols */
return 0;
}
if(aa_id != new_aa_id) {
change_count++;
}
}
return 3 - change_count;
}
