static void storeSAincurrentPGL(GtArray *pgls, GtUword currentPGLindex,
GthSA *sa)
{
GtUword leftgenomicexonborder, rightgenomicexonborder;
GthPGL *currentPGL;
/* the current PGL index is defined */
gt_assert(currentPGLindex != GT_UNDEF_UWORD);
currentPGL = *(GthPGL**) gt_array_get(pgls, currentPGLindex);
/* update maxrange */
leftgenomicexonborder = gth_sa_get_exon(sa, 0)
->leftgenomicexonborder;
rightgenomicexonborder = gth_sa_get_exon(sa,
gth_sa_num_of_exons(sa)-1)
->rightgenomicexonborder;
if (leftgenomicexonborder < currentPGL->maxrange.start)
currentPGL->maxrange.start = leftgenomicexonborder;
if (rightgenomicexonborder > currentPGL->maxrange.end)
currentPGL->maxrange.end = rightgenomicexonborder;
/* save SA */
gth_pgl_add_sa(currentPGL, sa);
}
GtUword gth_sa_intron_length(const GthSA *sa, GtUword intron)
{
Exoninfo *left_exon, *right_exon;
gt_assert(sa);
left_exon = gth_sa_get_exon(sa, intron);
right_exon = gth_sa_get_exon(sa, intron + 1);
return right_exon->leftgenomicexonborder - left_exon->rightgenomicexonborder
- 1;
}
static void addSAtointrondistribution(GtDiscDistri *introndistribution,
GthSA *sa)
{
unsigned long i;
/* add values to introndistribution */
for (i = 0; i < gth_sa_num_of_introns(sa); i++) {
gt_disc_distri_add(introndistribution,
gth_sa_get_exon(sa, i+1)
->leftgenomicexonborder -
gth_sa_get_exon(sa, i)
->rightgenomicexonborder - 1);
}
}
static void storeSAinnewPGL(GtArray *pgls, GtUword *currentPGLindex,
GthSA *sa)
{
GthPGL *pgl;
pgl = gth_pgl_new(gth_sa_gen_strand_forward(sa));
pgl->maxrange.start = gth_sa_get_exon(sa, 0)->leftgenomicexonborder;
pgl->maxrange.end = gth_sa_get_exon(sa,gth_sa_num_of_exons(sa)-1)
->rightgenomicexonborder;
gth_pgl_add_sa(pgl, sa);
gt_array_add(pgls, pgl);
/* set the current PGL index */
*currentPGLindex = gt_array_size(pgls) - 1;
}
GtUword gth_sa_genomic_exon_length(const GthSA *sa, GtUword exon)
{
Exoninfo *exoninfo;
gt_assert(sa);
exoninfo = gth_sa_get_exon(sa, exon);
return exoninfo->rightgenomicexonborder - exoninfo->leftgenomicexonborder + 1;
}
double gth_sa_exon_score(const GthSA *sa, GtUword exon)
{
Exoninfo *exoninfo;
gt_assert(sa);
exoninfo = gth_sa_get_exon(sa, exon);
return exoninfo->exonscore;
}
GtUword gth_sa_right_intron_border(const GthSA *sa, GtUword intron)
{
Exoninfo *exoninfo;
gt_assert(sa);
exoninfo = gth_sa_get_exon(sa, intron + 1);
return SHOWGENPOS(sa->gen_strand_forward,
sa->gen_total_length,
sa->gen_offset,
exoninfo->leftgenomicexonborder - 1);
}
static void addSAtoexondistribution(GtDiscDistri *exondistribution,
GthSA *sa)
{
Exoninfo *exoninfo;
unsigned long i;
/* add values to exondistribution */
for (i = 0; i < gth_sa_num_of_exons(sa); i++) {
exoninfo = gth_sa_get_exon(sa, i);
gt_disc_distri_add(exondistribution, exoninfo->rightgenomicexonborder -
exoninfo->leftgenomicexonborder + 1);
}
}
static void showexons(GthSA *sa, unsigned int indentlevel,
GtFile *outfp)
{
Exoninfo *exoninfo;
GtUword i;
for (i = 0; i < gth_sa_num_of_exons(sa); i++) {
exoninfo = gth_sa_get_exon(sa, i);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exoninfo>\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<leftgenomicexonborder>"GT_WU"</leftgenomicexonborder>\n",
exoninfo->leftgenomicexonborder);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<rightgenomicexonborder>"GT_WU
"</rightgenomicexonborder>\n",
exoninfo->rightgenomicexonborder);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<leftreferenceexonborder>"GT_WU
"</leftreferenceexonborder>\n",
exoninfo->leftreferenceexonborder);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<rightreferenceexonborder>"GT_WU
"</rightreferenceexonborder>\n",
exoninfo->rightreferenceexonborder);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exonscore>%.*f</exonscore>\n", PRECISION,
exoninfo->exonscore);
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</exoninfo>\n");
}
}
static void xml_showalignmentheader(GthSA *sa,
unsigned long minintronlength,
unsigned int indentlevel, GtFile *outfp)
{
unsigned long i, leftreferenceexonborder, rightreferenceexonborder,
referenceexonlength;
GthDbl exonscore, donorsitescore, acceptorsitescore;
GthFlt donorsiteprobability, acceptorsiteprobability;
Exoninfo *exoninfo;
Introninfo *introninfo;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<predicted_gene_structure>\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exon-intron_info>\n");
indentlevel++;
for (i = 0; i < gth_sa_num_of_exons(sa); i++) {
exoninfo = gth_sa_get_exon(sa, i);
leftreferenceexonborder = exoninfo->leftreferenceexonborder;
rightreferenceexonborder = exoninfo->rightreferenceexonborder;
referenceexonlength = rightreferenceexonborder
- leftreferenceexonborder + 1;
exonscore = exoninfo->exonscore;
if (i > 0) {
introninfo = gth_sa_get_intron(sa, i-1);
donorsiteprobability = introninfo->donorsiteprobability;
donorsitescore = introninfo->donorsitescore;
acceptorsiteprobability = introninfo->acceptorsiteprobability;
acceptorsitescore = introninfo->acceptorsitescore;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<intron i_serial=\"%lu\">\n",
i - 1 + OUTPUTOFFSET);
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<gDNA_intron_boundary i_start=\"%lu\" i_stop=\"%lu\" "
"i_length=\"%lu\">\n",
gth_sa_left_intron_border(sa, i-1),
gth_sa_right_intron_border(sa, i-1),
gth_sa_intron_length(sa, i-1));
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<donor d_prob=\"%.3f\"", donorsiteprobability);
if (gth_sa_alphatype(sa) == DNA_ALPHA)
gt_file_xprintf(outfp, " d_score=\"%.2f\"", donorsitescore);
gt_file_xprintf(outfp, "/>\n");
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<acceptor a_prob=\"%.3f\"",
acceptorsiteprobability);
if (gth_sa_alphatype(sa) == DNA_ALPHA)
gt_file_xprintf(outfp, " a_score=\"%.2f\"", acceptorsitescore);
gt_file_xprintf(outfp, "/>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</gDNA_intron_boundary>\n");
/* if the intron is shorter or equal than the minimal intron length an
additional tag is shown */
if (gth_sa_intron_length(sa, i-1) <= minintronlength) {
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<shorter_than_min_intron_len/>\n");
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</intron>\n");
}
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exon e_serial=\"%lu\">\n", i + OUTPUTOFFSET);
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<gDNA_exon_boundary g_start=\"%lu\" g_stop="
"\"%lu\" g_length=\"%lu\"/>\n",
gth_sa_left_genomic_exon_border(sa, i),
gth_sa_right_genomic_exon_border(sa, i),
gth_sa_genomic_exon_length(sa, i));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<reference_exon_boundary r_type=\"%s\" r_start=\"%lu\" "
"r_stop=\"%lu\" r_length=\"%lu\" r_score=\"%5.3f\"/>\n",
gth_sa_alphastring(sa),
leftreferenceexonborder + OUTPUTOFFSET ,
rightreferenceexonborder + OUTPUTOFFSET ,
referenceexonlength, exonscore);
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</exon>\n");
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</exon-intron_info>\n");
/* showing PPA line (if an poly-A tail was determined) */
if (gth_sa_alphatype(sa) == DNA_ALPHA)
xml_showppaline(sa, indentlevel, outfp);
/* showing MATCH line */
xml_showmatchline(sa, indentlevel, outfp);
/* showing PGS line */
xml_showpgsline(sa, indentlevel, outfp);
}
static void showalignmentheader(GthSA *sa, bool gs2out, int widthforgenpos,
GtUword minintronlength, GtFile *outfp)
{
GtUword i, leftreferenceexonborder, rightreferenceexonborder,
referenceexonlength;
GthDbl exonscore, donorsitescore, acceptorsitescore;
GthFlt donorsiteprobability, acceptorsiteprobability;
Exoninfo *exoninfo;
Introninfo *introninfo;
gt_file_xprintf(outfp, "Predicted gene structure");
if (gs2out) {
gt_file_xprintf(outfp, " (within gDNA segment "GT_WU" to "GT_WU"):\n",
gth_sa_gen_dp_start_show(sa),
gth_sa_gen_dp_end_show(sa));
}
else
gt_file_xprintf(outfp, ":\n");
gt_file_xfputc('\n', outfp);
for (i = 0; i < gth_sa_num_of_exons(sa); i++) {
exoninfo = gth_sa_get_exon(sa, i);
leftreferenceexonborder = exoninfo->leftreferenceexonborder;
rightreferenceexonborder = exoninfo->rightreferenceexonborder;
referenceexonlength = rightreferenceexonborder
- leftreferenceexonborder + 1;
exonscore = exoninfo->exonscore;
if (i > 0) {
introninfo = gth_sa_get_intron(sa, i-1);
donorsiteprobability = introninfo->donorsiteprobability;
donorsitescore = introninfo->donorsitescore;
acceptorsiteprobability = introninfo->acceptorsiteprobability;
acceptorsitescore = introninfo->acceptorsitescore;
gt_file_xprintf(outfp, " Intron %2" GT_WUS " %*" GT_WUS " %*" GT_WUS
" (%4" GT_WUS " n); ",
i - 1 + OUTPUTOFFSET, widthforgenpos,
gth_sa_left_intron_border(sa, i-1), widthforgenpos,
gth_sa_right_intron_border(sa, i-1),
gth_sa_intron_length(sa, i-1));
gt_file_xprintf(outfp, "Pd: %5.3f ", donorsiteprobability);
if (gth_sa_alphatype(sa) == DNA_ALPHA) {
if (donorsitescore == 0.0)
gt_file_xprintf(outfp, "(s: 0), ");
else
gt_file_xprintf(outfp, "(s: %4.2f), ", donorsitescore);
}
else
gt_file_xprintf(outfp, " ");
gt_file_xprintf(outfp, "Pa: %5.3f ", acceptorsiteprobability);
if (gth_sa_alphatype(sa) == DNA_ALPHA) {
if (acceptorsitescore == 0.0)
gt_file_xprintf(outfp, "(s: 0)");
else
gt_file_xprintf(outfp, "(s: %4.2f)", acceptorsitescore);
}
/* if the intron is shorter or equal than the minimum intron length two
question marks are shown at the end of the line */
if (gth_sa_intron_length(sa, i-1) <= minintronlength)
gt_file_xprintf(outfp, " ??");
gt_file_xfputc('\n', outfp);
}
gt_file_xprintf(outfp,
" Exon %2" GT_WUS " %*" GT_WUS " %*" GT_WUS " (%4" GT_WUS
" n); %s %6" GT_WUS " %6" GT_WUS " (%4" GT_WUS " %s); "
"score: %5.3f\n", i + OUTPUTOFFSET, widthforgenpos,
gth_sa_left_genomic_exon_border(sa, i), widthforgenpos,
gth_sa_right_genomic_exon_border(sa, i),
gth_sa_genomic_exon_length(sa, i), gth_sa_alphastring(sa),
leftreferenceexonborder + OUTPUTOFFSET,
rightreferenceexonborder + OUTPUTOFFSET,
referenceexonlength,
gth_sa_alphatype(sa) == DNA_ALPHA ? "n" : "aa", exonscore);
}
/* showing PPA line (if an poly-A tail was determined) */
if (gth_sa_alphatype(sa) == DNA_ALPHA)
showppaline(sa, outfp);
gt_file_xfputc('\n', outfp);
/* showing MATCH line */
showmatchline(sa, outfp);
/* showing PGS line */
showpgsline(sa, outfp);
}