CodonFrequency * flat_CodonFrequency(CodonTable * ct)
{
int number[26];
CodonFrequency * out;
register int i;
out = CodonFrequency_alloc();
for(i=0;i<26;i++)
number[i] = 0;
for(i=0;i<64;i++)
out->freq[i]= 0.0;
for(i=0;i<125;i++)
if( has_random_bases(i) == FALSE && is_stop_codon(i,ct) == FALSE)
number[aminoacid_no_from_codon(ct,i)]++;
for(i=0;i<64;i++) {
if( is_stop_codon(codon_from_base4_codon(i),ct) == FALSE && number[aminoacid_no_from_codon(ct,codon_from_base4_codon(i))] != 0)
out->freq[i] = 1.0 / number[aminoacid_no_from_codon(ct,codon_from_base4_codon(i))];
}
return out;
}
RandomCodon * RandomCodon_from_cds_triplet(GeneFrequency21 * gf)
{
int i;
int a,b,c;
double total;
double lit;
RandomCodon * out;
out = RandomCodon_alloc();
for(i=0,total=0;i<64;i++)
total += gf->cds_triplet[i];
for(i=0;i<125;i++) {
if( has_random_bases(i) ) {
lit = 0.0;
for(a=0;a<4;a++)
for(b=0;b<4;b++)
for(c=0;c<4;c++)
lit += gf->cds_triplet[base4_codon_from_codon(permute_possible_random_bases(i,a,b,c))];
out->codon[i] = lit/(64*total);
} else {
out->codon[i] = gf->cds_triplet[base4_codon_from_codon(i)]/(total);
}
}
return out;
}
CodonMatrixScore * naive_CodonMatrixScore(CodonTable * ct,CompMat * comp)
{
int i;
int j;
CodonMatrixScore * out;
out = CodonMatrixScore_alloc();
for(i=0; i<125; i++)
for(j=i; j<125; j++) {
if( has_random_bases(i) == TRUE || has_random_bases(j) == TRUE )
out->score[i][j] = out->score[j][i] = 0;
else
out->score[i][j] = out->score[j][i] = fail_safe_CompMat_access(comp,aminoacid_no_from_codon(ct,i),aminoacid_no_from_codon(ct,j));
}
return out;
}
CodonMatrix * naive_CodonMatrix(CodonTable * ct,CompProb * comp)
{
int i;
int j;
CodonMatrix * out;
out = CodonMatrix_alloc();
for(i=0; i<125; i++)
for(j=i; j<125; j++) {
if( has_random_bases(i) == TRUE || has_random_bases(j) == TRUE ) {
out->prob[i][j] = out->prob[j][i] = 1.0;
} else if ( is_stop_codon(i,ct) == TRUE || is_stop_codon(j,ct) == TRUE ) {
out->prob[i][j] = out->prob[j][i] = 0.00001;
} else {
out->prob[i][j] = out->prob[j][i] = comp->comp[aminoacid_no_from_codon(ct,i)][aminoacid_no_from_codon(ct,j)];
}
}
return out;
}
RandomCodon * RandomCodon_from_raw_CodonFrequency(double codon[64],CodonTable *ct)
{
RandomCodon * out;
register int i;
double total = 0.0;
base one;
base two;
base three;
int o,t,r;
/** codon frequencies here *do* include protein amino acid freq...
ie, they are raw frequencies, not adjusted for a codon table **/
/** the only thing is that we have to figure out how to
deal with N'd codons, which will just be summed over... **/
out= RandomCodon_alloc();
for(i=0;i<64;i++) {
total += codon[i];
}
for(i=0;i<125;i++) {
if( has_random_bases(i) == FALSE ) {
out->codon[i] = codon[base4_codon_from_codon(i)]/total;
}
else {
all_bases_from_codon(i,&one,&two,&three);
if( one == BASE_N && two != BASE_N && three != BASE_N ) {
for(o=0;o<4;o++)
out->codon[i] += (codon[o*16+two*4+three]/total);
}
else if( one == BASE_N && two == BASE_N && three != BASE_N) {
for(o=0;o<4;o++)
for(t=0;t<4;t++)
out->codon[i] += (codon[o*16+t*4+three]/total);
}
else if( one == BASE_N && two == BASE_N && three == BASE_N) {
for(o=0;o<4;o++)
for(t=0;t<4;t++)
for(r=0;r<4;r++)
out->codon[i] += (codon[o*16+t*4+r]/total);
}
else if( one != BASE_N && two == BASE_N && three != BASE_N) {
for(t=0;t<4;t++)
out->codon[i] += (codon[one*16+t*4+three]/total);
}
else if( one != BASE_N && two == BASE_N && three == BASE_N) {
for(t=0;t<4;t++)
for(r=0;r<4;r++)
out->codon[i] += (codon[one*16+t*4+r]/total);
}
else if( one != BASE_N && two != BASE_N && three == BASE_N) {
for(r=0;r<4;r++)
out->codon[i] += (codon[one*16+two*4+r]/total);
}
}
}
out->codon[125] = 0.0;
return out;
}
CodonMapper * new_CodonMapper(CodonTable * ct,CodonFrequency * cf)
{
register int i;
register int j;
int k;
base one;
base two;
base three;
int base4;
int oi,ti,ri;
double total_freq;
CodonMapper * out;
out = CodonMapper_alloc();
out->ct = hard_link_CodonTable(ct);
for(i=0;i<125;i++) {
for(j=0;j<26;j++)
out->codon_map[i][j] =0.0;
if( has_random_bases(i) == FALSE ) {
if( is_stop_codon(i,ct) == TRUE ) {
for(k=0;k<26;k++)
out->codon_map[i][k] = (0.0);
}
else {
out->codon_map[i][aminoacid_no_from_codon(ct,i)] = cf->freq[base4_codon_from_codon(i)];
}
}
else { /*** is a random base ***/
/***
sneaky stuff. What we want to do is loop over all possible
codons, adding up their frequencies for the amino acids
they represent. This is done by looping over all possible
bases for each position and then letting through ones
which either have an N at this position or is the actual base.
***/
all_bases_from_codon(i,&one,&two,&three);
total_freq = 0.0;
for(oi=0;oi<4;oi++)
for(ti=0;ti<4;ti++)
for(ri=0;ri<4;ri++) {
if( (one == BASE_N || one == oi) && (two == BASE_N || two == ti) && (three == BASE_N || three == ri) ) {
base4 = codon_from_base4_codon(oi*16+ti*4+ri);
if( !is_stop_codon(base4,ct) ) {
out->codon_map[i][aminoacid_no_from_codon(ct,base4)] += cf->freq[base4_codon_from_codon(base4)];
}
} /* end of if one == BASE_N || one == oi */
} /* end of for oi,ti,ri */
} /* end of else */
}
return out;
}
