static void xml_outputPGSlines(GtArray *alignments, unsigned int indentlevel,
GtFile *outfp)
{
unsigned long i, j;
GthSA *sa;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<supporting_evidence xmlns=\""
"http://www.genomethreader.org/"
"GTH_output/PGL_module/predicted_gene_location/"
"AGS_information/supporting_evidence/\">\n");
indentlevel++;
for (i = 0; i < gt_array_size(alignments); i++) {
sa = *(GthSA**) gt_array_get(alignments, i);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<PGS_line>\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<gDNA_exon_coordinates>\n");
indentlevel++;
for (j = 0; j < gth_sa_num_of_exons(sa); j++) {
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exon start=\"%lu\" stop=\"%lu\"/>\n",
gth_sa_left_genomic_exon_border(sa, j),
gth_sa_right_genomic_exon_border(sa, j));
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</gDNA_exon_coordinates>\n");
gth_indent(outfp, indentlevel);
if (gth_sa_alphatype(sa) == DNA_ALPHA) {
gt_file_xprintf(outfp, "<referenceDNA id=\"%s\" strand=\"%c\"/>\n",
gth_sa_ref_id(sa),
gth_sa_ref_strand_char(sa));
}
else {
gt_file_xprintf(outfp, "<referenceProtein id=\"%s\"/>\n",
gth_sa_ref_id(sa));
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</PGS_line>\n");
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</supporting_evidence>\n");
}
/*
The following function prints the "classic" GeneSeqer2 PGS line
*/
static void showpgsline(GthSA *sa, GtFile *outfp)
{
GtUword i, numofexons;
gt_assert(sa);
numofexons = gth_sa_num_of_exons(sa);
gt_file_xprintf(outfp, "PGS_%s%c_%s%c\t(",
gth_sa_gen_id(sa),
gth_sa_gen_strand_char(sa),
gth_sa_ref_id(sa),
gth_sa_ref_strand_char(sa));
for (i = 0; i < numofexons; i++) {
gt_file_xprintf(outfp, ""GT_WU" "GT_WU"",
gth_sa_left_genomic_exon_border(sa, i),
gth_sa_right_genomic_exon_border(sa, i));
if (i == numofexons - 1)
gt_file_xprintf(outfp, ")\n\n");
else
gt_file_xfputc(',', outfp);
}
}
static void outputPGSlines(GtArray *alignments, GtFile *outfp)
{
GtUword i, j;
GthSA *sa;
for (i = 0; i < gt_array_size(alignments); i++) {
sa = *(GthSA**) gt_array_get(alignments, i);
gt_file_xprintf(outfp, " PGS (");
for (j = 0; j < gth_sa_num_of_exons(sa); j++) {
if (j > 0)
gt_file_xfputc(',', outfp);
gt_file_xprintf(outfp, GT_WU " " GT_WU ,
gth_sa_left_genomic_exon_border(sa, j),
gth_sa_right_genomic_exon_border(sa, j));
}
gt_file_xprintf(outfp, ")\t%s%c\n", gth_sa_ref_id(sa),
gth_sa_ref_strand_char(sa));
}
gt_file_xfputc('\n', outfp);
}
/*
The following function prints the "classic" GeneSeqer2 PGS line
*/
static void xml_showpgsline(GthSA *sa, unsigned int indentlevel,
GtFile *outfp)
{
unsigned long i, numofexons;
gt_assert(sa);
numofexons = gth_sa_num_of_exons(sa);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<PGS_line>\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<gDNA gen_id=\"%s\" gen_strand=\"%c\"/>\n",
gth_sa_gen_id(sa), gth_sa_gen_strand_char(sa));
gth_indent(outfp, indentlevel);
if (gth_sa_alphatype(sa) == DNA_ALPHA) {
gt_file_xprintf(outfp, "<rDNA rDNA_id=\"%s\" rDNA_strand=\"%c\"/>\n",
gth_sa_ref_id(sa), gth_sa_ref_strand_char(sa));
}
else {
gt_file_xprintf(outfp, "<rProt rProt_id=\"%s\"/>\n",
gth_sa_ref_id(sa));
}
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<gDNA_exon_coordinates>\n");
indentlevel++;
for (i = 0; i < numofexons; i++) {
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<exon e_start=\"%lu\" e_stop=\"%lu\"/>\n",
gth_sa_left_genomic_exon_border(sa, i),
gth_sa_right_genomic_exon_border(sa, i));
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</gDNA_exon_coordinates>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</PGS_line>\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);
}
