static void showgthreferenceinformation(GthSA *sa, GthInput *input,
bool showseqnums,
GtFile *outfp)
{
gt_assert(gth_sa_ref_file_num(sa) != GT_UNDEF_UWORD);
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
gt_file_xprintf(outfp,
"EST Sequence: file=%s, strand=%c, description=",
gth_input_get_reference_filename(input,
gth_sa_ref_file_num(sa)),
gth_sa_ref_strand_char(sa));
break;
case PROTEIN_ALPHA:
gt_file_xprintf(outfp, "Protein Sequence: file=%s, description=",
gth_input_get_reference_filename(input,
gth_sa_ref_file_num(sa)));
break;
default: gt_assert(0);
}
gth_sa_echo_reference_description(sa, input, outfp);
if (showseqnums)
gt_file_xprintf(outfp, ", seqnum="GT_WU"", gth_sa_ref_seq_num(sa));
gt_file_xfputc('\n', outfp);
gt_file_xfputc('\n', outfp);
}
static void xml_showgthreferenceinformation(GthSA *sa,
GthInput *input,
unsigned int indentlevel,
GtFile *outfp)
{
gt_assert(gth_sa_ref_file_num(sa) != GT_UNDEF_ULONG);
gth_indent(outfp, indentlevel);
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
gt_file_xprintf(outfp, "<reference ref_file=\"%s\" ref_id=\"%s\" "
"ref_strand=\"%c\" ref_description=\"",
gth_input_get_reference_filename(input,
gth_sa_ref_file_num(sa)),
gth_sa_ref_id(sa),
gth_sa_ref_strand_char(sa));
break;
case PROTEIN_ALPHA:
gt_file_xprintf(outfp, "<reference ref_file=\"%s\" ref_id=\"%s\" "
"ref_description=\"",
gth_input_get_reference_filename(input,
gth_sa_ref_file_num(sa)),
gth_sa_ref_id(sa));
break;
default: gt_assert(0);
}
gth_input_echo_reference_description(input, gth_sa_ref_file_num(sa),
gth_sa_ref_seq_num(sa), outfp);
gt_file_xprintf(outfp, "\">\n");
}
/* The following function prints the "classic" GeneSeqer2 MATCH line */
static void xml_showmatchline(GthSA *sa, unsigned int indentlevel,
GtFile *outfp)
{
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<MATCH_line gen_id=\"%s\" gen_strand=\"%c\" ",
gth_sa_gen_id(sa),
gth_sa_gen_strand_char(sa));
if (gth_sa_alphatype(sa) == DNA_ALPHA) {
gt_file_xprintf(outfp, "ref_id=\"%s\" ref_strand=\"%c\">\n",
gth_sa_ref_id(sa),
gth_sa_ref_strand_char(sa));
}
else
gt_file_xprintf(outfp, "ref_id=\"%s\">\n", gth_sa_ref_id(sa));
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<total_alignment_score>%.3f</total_alignment_score>\n",
gth_sa_score(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<cumulative_length_of_scored_exons>%lu"
"</cumulative_length_of_scored_exons>\n",
gth_sa_cumlen_scored_exons(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<coverage percentage=\"%.3f\" high_type=\"",
gth_sa_coverage(sa));
gt_file_xfputc(gth_sa_coverage_char(sa), outfp);
gt_file_xprintf(outfp, "\"/>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</MATCH_line>\n");
}
char gth_sa_coverage_char(const GthSA *sa)
{
gt_assert(sa);
if (gth_sa_genomic_cov_is_highest(sa))
return 'G';
else {
if (gth_sa_alphatype(sa) == DNA_ALPHA)
return 'C';
else
return 'P';
}
}
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 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 end_element_handler(void *info, const XML_Char *name)
{
Parseinfo *parseinfo = (Parseinfo*) info;
GthSA *sa = parseinfo->currentSA;
GtUword datalength;
double retdouble;
GtWord ret;
char *data;
/* save data and data length */
data = gt_str_get(parseinfo->databuf);
datalength = gt_str_length(parseinfo->databuf);
/* perform actions depending on end tag */
if (strcmp(name, SPLICEDALIGNMENT_TAG) == 0) {
/* before we store the spliced alignment we have to reverse its edit
operations */
gt_assert(sa && gth_sa_backtrace_path(sa));
gth_backtrace_path_reverse(gth_sa_backtrace_path(sa));
/* ensure that before an intron which is not in phase the edit operation
has length 1 (only for protein spliced alignments) */
gth_backtrace_path_ensure_length_1_before_introns(
gth_sa_backtrace_path(sa));
if (parseinfo->saprocessfunc(parseinfo->data , sa,
parseinfo->outputfilename, parseinfo->err)) {
/* XXX */
fprintf(stderr, "error: %s\n", gt_error_get(parseinfo->err));
exit(EXIT_FAILURE);
}
/* reset current spliced alignment */
parseinfo->currentSA = NULL;
}
else if (strcmp(name, REFERENCEALPHATYPE_TAG) == 0) {
if (strcmp(data, "DNA_ALPHA") == 0)
gth_sa_set_alphatype(sa, DNA_ALPHA);
else if (strcmp(data, "PROTEIN_ALPHA") == 0) {
gth_sa_set_alphatype(sa, PROTEIN_ALPHA);
}
else {
ILLEGAL_DATA;
}
}
else if (strcmp(name, DNA_EOP_TYPE_TAG) == 0) {
if (strcmp(data, "match") == 0)
parseinfo->eoptype = EOP_TYPE_MATCH;
else if (strcmp(data, "deletion") == 0)
parseinfo->eoptype = EOP_TYPE_DELETION;
else if (strcmp(data, "insertion") == 0)
parseinfo->eoptype = EOP_TYPE_INSERTION;
else if (strcmp(data, "mismatch") == 0)
parseinfo->eoptype = EOP_TYPE_MISMATCH;
else if (strcmp(data, "intron") == 0)
parseinfo->eoptype = EOP_TYPE_INTRON;
else {
ILLEGAL_DATA;
}
}
else if (strcmp(name, DNA_EOP_LENGTH_TAG) == 0) {
SCANUINT;
gth_backtrace_path_add_eop(gth_sa_backtrace_path(sa), parseinfo->eoptype,
ret);
}
else if (strcmp(name, PROTEIN_EOP_TYPE_TAG) == 0) {
if (strcmp(data, "match") == 0)
parseinfo->eoptype = EOP_TYPE_MATCH;
else if (strcmp(data, "deletion") == 0)
parseinfo->eoptype = EOP_TYPE_DELETION;
else if (strcmp(data, "insertion") == 0)
parseinfo->eoptype = EOP_TYPE_INSERTION;
else if (strcmp(data, "mismatch") == 0)
parseinfo->eoptype = EOP_TYPE_MISMATCH;
else if (strcmp(data, "intron") == 0)
parseinfo->eoptype = EOP_TYPE_INTRON;
else if (strcmp(data, "mismatch_with_1_gap") == 0)
parseinfo->eoptype = EOP_TYPE_MISMATCH_WITH_1_GAP;
else if (strcmp(data, "mismatch_with_2_gaps") == 0)
parseinfo->eoptype = EOP_TYPE_MISMATCH_WITH_2_GAPS;
else if (strcmp(data, "deletion_with_1_gap") == 0)
parseinfo->eoptype = EOP_TYPE_DELETION_WITH_1_GAP;
else if (strcmp(data, "deletion_with_2_gaps") == 0)
parseinfo->eoptype = EOP_TYPE_DELETION_WITH_2_GAPS;
else if (strcmp(data, "intron_with_1_base_left") == 0)
parseinfo->eoptype = EOP_TYPE_INTRON_WITH_1_BASE_LEFT;
else if (strcmp(data, "intron_with_2_bases_left") == 0)
parseinfo->eoptype = EOP_TYPE_INTRON_WITH_2_BASES_LEFT;
else {
ILLEGAL_DATA;
}
}
else if (strcmp(name, PROTEIN_EOP_LENGTH_TAG) == 0) {
SCANUINT;
gth_backtrace_path_add_eop(gth_sa_backtrace_path(sa), parseinfo->eoptype,
ret);
}
else if (strcmp(name, INDELCOUNT_TAG) == 0) {
SCANUINT;
/* ignore indelcount, gets recomputed anyway */
}
else if (strcmp(name, GENOMICLENGTHDP_TAG) == 0) {
SCANUINT;
gth_sa_set_gen_dp_length(sa, ret);
}
else if (strcmp(name, GENOMICLENGTHTOTAL_TAG) == 0) {
SCANUINT;
gth_sa_set_gen_total_length(sa, ret);
}
else if (strcmp(name, GENOMICOFFSET_TAG) == 0) {
SCANUINT;
gth_sa_set_gen_offset(sa, ret);
}
else if (strcmp(name, REFERENCELENGTH_TAG) == 0) {
SCANUINT;
gth_sa_set_ref_total_length(sa, ret);
}
else if (strcmp(name, DPSTARTPOS_TAG) == 0) {
SCANUINT;
gth_sa_set_gen_dp_start(sa, ret);
}
else if (strcmp(name, DPENDPOS_TAG) == 0) {
SCANUINT;
/* ignore DP end pos, gets recomputed from gen_dp_length anyway */
gt_assert(gth_sa_gen_dp_end(sa) == ret);
}
else if (strcmp(name, GENOMICFILENAME_TAG) == 0) {
/* save genomic file name */
gt_str_append_cstr_nt(parseinfo->genomicfilename, data, datalength);
}
else if (strcmp(name, GENOMICFILEHASH_TAG) == 0) {
gth_sa_set_gen_file_num(sa, process_file(parseinfo->input,
gt_str_get(parseinfo->genomicfilename), data, false,
UNDEF_ALPHA));
/* reset genomic filename */
gt_str_reset(parseinfo->genomicfilename);
}
else if (strcmp(name, GENOMICSEQNUM_TAG) == 0) {
SCANUINT;
gth_sa_set_gen_seq_num(sa, ret);
}
else if (strcmp(name, REFERENCEFILENAME_TAG) == 0) {
/* save reference file name */
gt_str_append_cstr_nt(parseinfo->referencefilename, data, datalength);
}
else if (strcmp(name, REFERENCEFILEHASH_TAG) == 0) {
gth_sa_set_ref_file_num(sa, process_file(parseinfo->input,
gt_str_get(parseinfo->referencefilename),
data, true,
gth_sa_alphatype(sa)));
/* reset reference filename */
gt_str_reset(parseinfo->referencefilename);
}
else if (strcmp(name, REFERENCESEQNUM_TAG) == 0) {
SCANUINT;
gth_sa_set_ref_seq_num(sa, ret);
}
else if (strcmp(name, GENOMICID_TAG) == 0)
gth_sa_set_gen_id(sa, data);
else if (strcmp(name, REFERENCEID_TAG) == 0)
gth_sa_set_ref_id(sa, data);
else if (strcmp(name, GENOMICSTRANDISFORWARD_TAG) == 0)
gth_sa_set_gen_strand(sa, parse_boolean(data, parseinfo));
else if (strcmp(name, REFERENCESTRANDISFORWARD_TAG) == 0)
gth_sa_set_ref_strand(sa, parse_boolean(data, parseinfo));
else if (strcmp(name, GENOMICCUTOFF_TAG) == 0) {
SCANUINT;
parseinfo->cutoffs.genomiccutoff = ret;
}
else if (strcmp(name, REFERENCECUTOFF_TAG) == 0) {
SCANUINT;
parseinfo->cutoffs.referencecutoff = ret;
}
else if (strcmp(name, EOPCUTOFF_TAG) == 0) {
SCANUINT;
parseinfo->cutoffs.eopcutoff = ret;
}
else if (strcmp(name, CUTOFFSSTART_TAG) == 0)
gth_sa_set_cutoffs_start(sa, &parseinfo->cutoffs);
else if (strcmp(name, CUTOFFSEND_TAG) == 0)
gth_sa_set_cutoffs_end(sa, &parseinfo->cutoffs);
else if (strcmp(name, LEFTGENOMICEXONBORDER_TAG) == 0) {
SCANUINT;
parseinfo->exoninfo.leftgenomicexonborder = ret;
}
else if (strcmp(name, RIGHTGENOMICEXONBORDER_TAG) == 0) {
SCANUINT;
parseinfo->exoninfo.rightgenomicexonborder = ret;
}
else if (strcmp(name, LEFTREFERENCEEXONBORDER_TAG) == 0) {
SCANUINT;
parseinfo->exoninfo.leftreferenceexonborder = ret;
}
else if (strcmp(name, RIGHTREFERENCEEXONBORDER_TAG) == 0) {
SCANUINT;
parseinfo->exoninfo.rightreferenceexonborder = ret;
}
else if (strcmp(name, EXONSCORE_TAG) == 0) {
SCANDOUBLE;
parseinfo->exoninfo.exonscore = retdouble;
}
else if (strcmp(name, EXONINFO_TAG) == 0)
gth_sa_add_exon(sa, &parseinfo->exoninfo);
else if (strcmp(name, DONORSITEPROBABILITY_TAG) == 0) {
SCANDOUBLE;
parseinfo->introninfo.donorsiteprobability = (GthFlt) retdouble;
}
else if (strcmp(name, ACCEPTORSITEPROBABILITY_TAG) == 0) {
SCANDOUBLE;
parseinfo->introninfo.acceptorsiteprobability = (GthFlt) retdouble;
}
else if (strcmp(name, DONORSITESCORE_TAG) == 0) {
SCANDOUBLE;
parseinfo->introninfo.donorsitescore = retdouble;
}
else if (strcmp(name, ACCEPTORSITESCORE_TAG) == 0) {
SCANDOUBLE;
parseinfo->introninfo.acceptorsitescore = retdouble;
}
else if (strcmp(name, INTRONINFO_TAG) == 0)
gth_sa_add_intron(sa, &parseinfo->introninfo);
else if (strcmp(name, POLYASTART_TAG) == 0) {
SCANUINT;
gth_sa_set_polyAtail_start(sa, ret);
}
else if (strcmp(name, POLYAEND_TAG) == 0) {
SCANUINT;
gth_sa_set_polyAtail_stop(sa, ret);
}
else if (strcmp(name, ALIGNMENTSCORE_TAG) == 0) {
SCANDOUBLE;
gth_sa_set_score(sa, retdouble);
}
else if (strcmp(name, COVERAGE_TAG) == 0) {
SCANDOUBLE;
gth_sa_set_coverage(sa, retdouble);
}
else if (strcmp(name, COVERAGEOFGENOMICSEGMENTISHIGHEST_TAG) == 0) {
gth_sa_set_highest_cov(sa, parse_boolean(data, parseinfo));
}
else if (strcmp(name, CUMULATIVELENGTHOFSCOREDEXONS_TAG) == 0) {
SCANUINT;
gth_sa_set_cumlen_scored_exons(sa, ret);
}
}
static void xml_inter_show_spliced_alignment(GthSA *sa, GthInput *input,
unsigned int indentlevel,
GtFile *outfp)
{
bool dnaalpha = true;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<spliced_alignment xmlns=\"http://www.GenomeThreader.org/"
"SplicedAlignment/spliced_alignment/\">\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referencealphatype>");
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
gt_file_xprintf(outfp, "DNA_ALPHA");
break;
case PROTEIN_ALPHA:
gt_file_xprintf(outfp, "PROTEIN_ALPHA");
dnaalpha = false;
break;
default: gt_assert(0);
}
gt_file_xprintf(outfp, "</referencealphatype>\n");
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<editoperations>\n");
indentlevel++;
gth_backtrace_path_show_complete(gth_sa_backtrace_path(sa), true, indentlevel,
outfp);
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</editoperations>\n");
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<indelcount>"GT_WU"</indelcount>\n",
gth_sa_indelcount(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genomiclengthDP>"GT_WU"</genomiclengthDP>\n",
gth_sa_gen_dp_length(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genomiclengthtotal>"GT_WU"</genomiclengthtotal>\n",
gth_sa_gen_total_length(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genomicoffset>"GT_WU"</genomicoffset>\n",
gth_sa_gen_offset(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referencelength>"GT_WU"</referencelength>\n",
gth_sa_ref_total_length(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<dpstartpos>"GT_WU"</dpstartpos>\n",
gth_sa_gen_dp_start(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<dpendpos>"GT_WU"</dpendpos>\n",
gth_sa_gen_dp_end(sa));
showgenomicfilename(sa, input, indentlevel, outfp);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genomicseqnum>"GT_WU"</genomicseqnum>\n",
gth_sa_gen_seq_num(sa));
showreferencefilename(sa, input, indentlevel, outfp);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referenceseqnum>"GT_WU"</referenceseqnum>\n",
gth_sa_ref_seq_num(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genomicid>%s</genomicid>\n", gth_sa_gen_id(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referenceid>%s</referenceid>\n",
gth_sa_ref_id(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<genomicstrandisforward>%s</genomicstrandisforward>\n",
GTH_SHOWBOOL(gth_sa_gen_strand_forward(sa)));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<referencestrandisforward>%s</referencestrandisforward>\n",
GTH_SHOWBOOL(gth_sa_ref_strand_forward(sa)));
showalignmentcutoffs(sa, indentlevel, outfp);
showexons(sa, indentlevel, outfp);
showintrons(sa, dnaalpha, indentlevel, outfp);
showpolyAtailpos(sa, indentlevel, outfp);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<alignmentscore>%.*f</alignmentscore>\n",
PRECISION, gth_sa_score(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<coverage>%.*f</coverage>\n", PRECISION,
gth_sa_coverage(sa));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<coverageofgenomicsegmentishighest>%s"
"</coverageofgenomicsegmentishighest>\n",
GTH_SHOWBOOL(gth_sa_genomic_cov_is_highest(sa)));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<cumulativelengthofscoredexons>"GT_WU""
"</cumulativelengthofscoredexons>\n",
gth_sa_cumlen_scored_exons(sa));
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</spliced_alignment>\n");
}
static void xml_final_show_spliced_alignment(GthSA *sa, GthInput *input,
unsigned long minintronlength,
unsigned long translationtable,
unsigned int indentlevel,
GtFile *outfp)
{
unsigned char *first_line, *second_line, *third_line;
GT_UNUSED bool reverse_subject_pos = false;
unsigned long cols;
gt_assert(sa && input);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp,
"<spliced_alignment xmlns=\"http://www.genomethreader.org/"
"GTH_output/alignment_module/spliced_alignment/\">\n");
indentlevel++;
/* If the reverse complement of the genomic DNA is considered, this
opition is needed for correct output of the genomic sequence positions
by the function showalignmentgeneric() */
if (!gth_sa_gen_strand_forward(sa))
reverse_subject_pos = true;
xml_showgthreferenceinformation(sa, input, indentlevel, outfp);
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<seq>");
gth_sa_echo_reference_sequence(sa, input, false, outfp);
gt_file_xprintf(outfp, "</seq>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</reference>\n");
xml_showgthgenomicinformation(sa, input, indentlevel, outfp);
xml_showalignmentheader(sa, minintronlength, indentlevel, outfp);
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<alignment>\n");
/* compute the alignment lines */
cols = gth_sa_get_alignment_lines(sa, &first_line, &second_line, &third_line,
translationtable, input);
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<genome_strand>");
showconcreteline(first_line, cols, outfp);
gt_file_xprintf(outfp, "</genome_strand>\n");
gth_indent(outfp, indentlevel);
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
gt_file_xprintf(outfp, "<mrna_strand>");
showconcreteline(second_line, cols, outfp);
gt_file_xprintf(outfp, "</mrna_strand>\n");
break;
case PROTEIN_ALPHA:
gt_file_xprintf(outfp, "<genomeProt>");
showconcreteline(second_line, cols, outfp);
gt_file_xprintf(outfp, "</genomeProt>\n");
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<queryProt>");
showconcreteline(third_line, cols, outfp);
gt_file_xprintf(outfp, "</queryProt>\n");
break;
default: gt_assert(0);
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</alignment>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</predicted_gene_structure>\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</spliced_alignment>\n");
/* free */
gt_free(first_line);
gt_free(second_line);
gt_free(third_line);
}
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);
}
const char* gth_sa_alphastring(const GthSA *sa)
{
gt_assert(sa);
return gth_sa_alphatype(sa) == DNA_ALPHA ? "cDNA" : "Protein";
}
bool gth_sas_are_equal(const GthSA *saA, const GthSA *saB)
{
Exoninfo *exoninfoA, *exoninfoB;
Introninfo *introninfoA, *introninfoB;
GtUword i;
/* compare element 0 */
if (gth_sa_alphatype(saA) != gth_sa_alphatype(saB))
return false;
/* compare element 1 */
if (gth_backtrace_path_length(saA->backtrace_path) !=
gth_backtrace_path_length(saB->backtrace_path)) {
return false;
}
for (i = 0; i < gth_backtrace_path_length(saA->backtrace_path); i++) {
if (((Editoperation*) gth_backtrace_path_get(saA->backtrace_path))[i] !=
((Editoperation*) gth_backtrace_path_get(saB->backtrace_path))[i]) {
return false;
}
}
/* element 2 has been removed (indelcount) */
/* compare element 3 */
if (gth_sa_gen_dp_length(saA) != gth_sa_gen_dp_length(saB))
return false;
/* compare element 4 */
if (saA->gen_total_length != saB->gen_total_length)
return false;
/* compare element 5 */
if (saA->gen_offset != saB->gen_offset)
return false;
/* compare element 6 */
if (gth_sa_ref_total_length(saA) != gth_sa_ref_total_length(saB))
return false;
/* compare element 7 */
if (gth_sa_gen_dp_start(saA) != gth_sa_gen_dp_start(saB))
return false;
/* element 8 has been removed (gen_dp_end) */
/* compare element 9 */
if (saA->gen_file_num != saB->gen_file_num)
return false;
/* compare element 10 */
if (saA->gen_seq_num != saB->gen_seq_num)
return false;
/* compare element 11 */
if (saA->ref_file_num != saB->ref_file_num)
return false;
/* compare element 12 */
if (saA->ref_seq_num != saB->ref_seq_num)
return false;
/* compare element 13 */
if (gt_str_cmp(saA->gen_id, saB->gen_id))
return false;
/* compare element 14 */
if (gt_str_cmp(saA->ref_id, saB->ref_id))
return false;
/* compare element 15 */
if (saA->gen_strand_forward != saB->gen_strand_forward)
return false;
/* compare element 16 */
if (saA->ref_strand_forward != saB->ref_strand_forward)
return false;
/* compare element 17 */
if (gth_sa_genomiccutoff_start(saA) != gth_sa_genomiccutoff_start(saB))
return false;
if (gth_sa_referencecutoff_start(saA) != gth_sa_referencecutoff_start(saB))
return false;
if (gth_sa_eopcutoff_start(saA) != gth_sa_eopcutoff_start(saB))
return false;
if (gth_sa_genomiccutoff_end(saA) != gth_sa_genomiccutoff_end(saB))
return false;
if (gth_sa_referencecutoff_end(saA) != gth_sa_referencecutoff_end(saB))
return false;
if (gth_sa_eopcutoff_end(saA) != gth_sa_eopcutoff_end(saB))
return false;
/* compare element 18 */
if (gt_array_size(saA->exons) != gt_array_size(saB->exons))
return false;
for (i = 0; i < gt_array_size(saA->exons); i++) {
exoninfoA = (Exoninfo*) gt_array_get(saA->exons, i);
exoninfoB = (Exoninfo*) gt_array_get(saB->exons, i);
if (exoninfoA->leftgenomicexonborder != exoninfoB->leftgenomicexonborder)
return false;
if (exoninfoA->rightgenomicexonborder != exoninfoB->rightgenomicexonborder)
return false;
if (exoninfoA->leftreferenceexonborder !=
exoninfoB->leftreferenceexonborder) {
return false;
}
if (exoninfoA->rightreferenceexonborder !=
exoninfoB->rightreferenceexonborder) {
return false;
}
if (!gt_double_equals_double(exoninfoA->exonscore, exoninfoB->exonscore)) {
return false;
}
}
/* compare element 19 */
if (gt_array_size(saA->introns) != gt_array_size(saB->introns))
return false;
for (i = 0; i < gt_array_size(saA->introns); i++) {
introninfoA = (Introninfo*) gt_array_get(saA->introns, i);
introninfoB = (Introninfo*) gt_array_get(saB->introns, i);
if (!gt_double_equals_double(introninfoA->donorsiteprobability,
introninfoB->donorsiteprobability)) {
return false;
}
if (!gt_double_equals_double(introninfoA->acceptorsiteprobability,
introninfoB->acceptorsiteprobability)) {
return false;
}
if (!gt_double_equals_double(introninfoA->donorsitescore,
introninfoB->donorsitescore)) {
return false;
}
if (!gt_double_equals_double(introninfoA->acceptorsitescore,
introninfoB->acceptorsitescore)) {
return false;
}
}
/* compare element 20 */
if (saA->polyAtailpos.start != saB->polyAtailpos.start)
return false;
if (saA->polyAtailpos.end != saB->polyAtailpos.end)
return false;
/* compare element 21 */
if (saA->alignmentscore != saB->alignmentscore)
return false;
/* compare element 22 */
if (saA->coverage != saB->coverage)
return false;
/* compare element 23 */
if (saA->genomic_cov_is_highest != saB->genomic_cov_is_highest)
return false;
/* compare element 24 */
if (saA->cumlen_scored_exons != saB->cumlen_scored_exons)
return false;
return true;
}
GtUword gth_sa_get_alignment_lines(const GthSA *sa,
unsigned char **first_line,
unsigned char **second_line,
unsigned char **third_line,
GtUword translationtable,
GthInput *input)
{
GtUword genomicstartcutoff, genomicendcutoff, genomictotalcutoff,
referencestartcutoff, referenceendcutoff, referencetotalcutoff;
GT_UNUSED bool reverse_subject_pos = false;
gt_assert(sa && first_line && second_line && third_line && input);
/* only for cosmetic reasons */
genomicstartcutoff = gth_sa_genomiccutoff_start(sa);
genomicendcutoff = gth_sa_genomiccutoff_end(sa);
genomictotalcutoff = genomicstartcutoff + genomicendcutoff;
referencestartcutoff = gth_sa_referencecutoff_start(sa);
referenceendcutoff = gth_sa_referencecutoff_end(sa);
referencetotalcutoff = referencestartcutoff + referenceendcutoff;
/* sequences */
unsigned char *gen_seq_orig, *ref_seq_orig;
GtUword cols = 0;
GthSeqCon *ref_seq_con;
/* make sure that the correct files are loaded */
gth_input_load_reference_file(input, gth_sa_ref_file_num(sa), false);
ref_seq_con = gth_input_current_ref_seq_con(input);
/* If the reverse complement of the genomic DNA is considered, this
opition is needed for correct output of the genomic sequence positions
by the function showalignmentgeneric() */
if (!gth_sa_gen_strand_forward(sa))
reverse_subject_pos = true;
/* get genomic sequence */
gen_seq_orig = (unsigned char*)
gth_input_original_genomic_sequence(input, gth_sa_gen_file_num(sa),
gth_sa_gen_strand_forward(sa))
+ gth_sa_gen_dp_start(sa);
/* get reference sequence */
if (gth_sa_ref_strand_forward(sa)) {
ref_seq_orig =
gth_seq_con_get_orig_seq(ref_seq_con, gth_sa_ref_seq_num(sa));
}
else {
ref_seq_orig =
gth_seq_con_get_orig_seq_rc(ref_seq_con, gth_sa_ref_seq_num(sa));
}
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
/* compute the two alignment lines */
cols = gthfillthetwoalignmentlines(first_line,
second_line,
gen_seq_orig +
genomicstartcutoff,
gth_sa_gen_dp_length(sa) -
genomictotalcutoff,
ref_seq_orig +
referencestartcutoff,
gth_sa_ref_total_length(sa) -
referencetotalcutoff,
gth_sa_get_editoperations(sa),
gth_sa_get_editoperations_length(sa),
0, /* linewidth not important here */
0, /* no short introns here */
NULL,/* therefore no shortintroninfo */
gth_sa_indelcount(sa));
*third_line = NULL;
break;
case PROTEIN_ALPHA:
/* compute the three alignment lines */
cols = gthfillthethreealignmentlines(first_line,
second_line,
third_line,
gth_sa_get_editoperations(sa),
gth_sa_get_editoperations_length(sa),
gth_sa_indelcount(sa),
gen_seq_orig +
genomicstartcutoff,
gth_sa_gen_dp_length(sa) -
genomictotalcutoff,
ref_seq_orig +
referencestartcutoff,
gth_sa_ref_total_length(sa) -
referencetotalcutoff,
translationtable);
break;
default: gt_assert(0);
}
return cols;
}
void gth_sa_echo_alignment(const GthSA *sa, GtUword showintronmaxlen,
GtUword translationtable,
bool wildcardimplosion, GthInput *input,
GtFile *outfp)
{
GtUword genomicstartcutoff, genomicendcutoff, genomictotalcutoff,
referencestartcutoff, referenceendcutoff, referencetotalcutoff;
bool reverse_subject_pos = false;
const unsigned char *gen_seq_orig, *ref_seq_orig;
GthSeqCon *ref_seq_con;
GtAlphabet *ref_alphabet;
gt_assert(sa && input);
/* only for cosmetic reasons */
genomicstartcutoff = gth_sa_genomiccutoff_start(sa);
genomicendcutoff = gth_sa_genomiccutoff_end(sa);
genomictotalcutoff = genomicstartcutoff + genomicendcutoff;
referencestartcutoff = gth_sa_referencecutoff_start(sa);
referenceendcutoff = gth_sa_referencecutoff_end(sa);
referencetotalcutoff = referencestartcutoff + referenceendcutoff;
/* make sure that the correct files are loaded */
gth_input_load_reference_file(input, gth_sa_ref_file_num(sa), false);
ref_seq_con = gth_input_current_ref_seq_con(input);
ref_alphabet = gth_input_current_ref_alphabet(input);
/* If the reverse complement of the genomic DNA is considered, this
opition is needed for correct output of the genomic sequence positions
by the function showalignmentgeneric() */
if (!gth_sa_gen_strand_forward(sa))
reverse_subject_pos = true;
/* get genomic sequence */
gen_seq_orig =
gth_input_original_genomic_sequence(input, sa->gen_file_num,
sa->gen_strand_forward)
+ gth_sa_gen_dp_start(sa);
/* get reference sequence */
if (gth_sa_ref_strand_forward(sa)) {
ref_seq_orig =
gth_seq_con_get_orig_seq(ref_seq_con, gth_sa_ref_seq_num(sa));
}
else {
ref_seq_orig =
gth_seq_con_get_orig_seq_rc(ref_seq_con, gth_sa_ref_seq_num(sa));
}
switch (gth_sa_alphatype(sa)) {
case DNA_ALPHA:
gthshowalignmentdna(outfp,ALIGNMENTLINEWIDTH,
gth_sa_get_editoperations(sa),
gth_sa_get_editoperations_length(sa),
gth_sa_indelcount(sa),
gen_seq_orig + genomicstartcutoff,
gth_sa_gen_dp_length(sa) - genomictotalcutoff,
ref_seq_orig + referencestartcutoff,
gth_sa_ref_total_length(sa) -
referencetotalcutoff,
gth_sa_gen_dp_start(sa) + genomicstartcutoff -
gth_sa_gen_offset(sa), referencestartcutoff,
gth_sa_gen_total_length(sa), showintronmaxlen,
ref_alphabet, reverse_subject_pos,
wildcardimplosion);
break;
case PROTEIN_ALPHA:
gthshowalignmentprotein(outfp, ALIGNMENTLINEWIDTH,
gth_sa_get_editoperations(sa),
gth_sa_get_editoperations_length(sa),
gth_sa_indelcount(sa),
gen_seq_orig + genomicstartcutoff,
gth_sa_gen_dp_length(sa) - genomictotalcutoff,
ref_seq_orig + referencestartcutoff,
gth_sa_ref_total_length(sa) -
referencetotalcutoff,
gth_sa_gen_dp_start(sa) + genomicstartcutoff -
gth_sa_gen_offset(sa), referencestartcutoff,
gth_sa_gen_total_length(sa), showintronmaxlen,
ref_alphabet, translationtable,
gth_input_score_matrix(input),
gth_input_score_matrix_alpha(input),
reverse_subject_pos, wildcardimplosion);
break;
default: gt_assert(0);
}
}
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);
}
