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");
}
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);
}
}
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;
}
static int callsahmt(bool call_dna_dp,
GthSA *sa,
bool forward,
GtUword gen_file_num,
GtUword ref_file_num,
GthChain *raw_chain,
GtUword gen_total_length,
GtUword gen_offset,
const GtRange *gen_seq_bounds,
const GtRange *gen_seq_bounds_rc,
const unsigned char *ref_seq_tran,
const unsigned char *ref_seq_orig,
GtUword ref_total_length,
GtUword ref_offset,
GthInput *input,
Introncutoutinfo *introncutoutinfo,
GthStat *stat,
GtUword chainctr,
GtUword num_of_chains,
GtUword translationtable,
bool directmatches,
bool proteinexonpenal,
GthSpliceSiteModel *splice_site_model,
GthDPOptionsCore *dp_options_core,
GthDPOptionsEST *dp_options_est,
GthDPOptionsPostpro *dp_options_postpro,
GthDNACompletePathMatrixJT dna_complete_path_matrix_jt,
GthProteinCompletePathMatrixJT
protein_complete_path_matrix_jt,
GthOutput *out)
{
int rval;
GthChain *actual_chain, *contracted_chain, *used_chain;
GtUword icdelta = introncutoutinfo->icinitialdelta,
iciterations = introncutoutinfo->iciterations;
bool useintroncutout = introncutoutinfo->introncutout;
/* initially useintron is set to the value of introncutoutinfo->introncutout,
if the automatic intron cutotu technique is acitvated it can be set to
true if an matrix allocation error (ERROR_MATRIX_ALLOCATION_FAILED) occurs
*/
gt_assert(sa);
actual_chain = gth_chain_new();
contracted_chain = gth_chain_new();
for (;;) {
/* reset actualDPrange; */
gt_array_set_size(actual_chain->forwardranges, 0);
gt_array_set_size(actual_chain->reverseranges, 0);
/* copy raw chain to actual chain */
gth_chain_copy(actual_chain, raw_chain);
/* shorten potential introns and compute spliced sequence, if the intron
cutout technique is used */
if (useintroncutout) {
/* shorten potential introns */
gth_chain_shorten_introns(actual_chain, icdelta,
introncutoutinfo->icminremintronlength,
gen_total_length, gen_offset, out->comments,
out->outfp);
}
else
gth_chain_contract(contracted_chain, actual_chain);
if (out->showverbose) {
show_matrix_calculation_status(out->showverbose, forward,
gth_sa_ref_strand_forward(sa),
useintroncutout, chainctr, num_of_chains,
icdelta, gen_file_num,
gth_input_num_of_gen_files(input),
ref_file_num,
gth_input_num_of_ref_files(input),
directmatches, out->verboseseqs,
gth_sa_gen_id(sa), gth_sa_ref_id(sa));
}
/* allocate space for DP parameter */
if (out->comments) {
gt_file_xprintf(out->outfp, "%c alloc space for DP param "
"(genomicid=%s, referenceid=%s)\n", COMMENTCHAR,
gth_sa_gen_id(sa), gth_sa_ref_id(sa));
}
used_chain = useintroncutout ? actual_chain : contracted_chain;
/* The variable 'forward' denotes the genomic strand on which the DP is
applied. */
if (forward) {
if (call_dna_dp) {
rval = gth_align_dna(sa, used_chain->forwardranges,
gth_input_current_gen_seq_tran(input),
gth_input_current_gen_seq_orig(input),
ref_seq_tran, ref_seq_orig, ref_total_length,
gth_input_current_gen_alphabet(input),
gth_input_current_ref_alphabet(input),
useintroncutout,
introncutoutinfo->autoicmaxmatrixsize,
out->showeops, out->comments, out->gs2out,
gen_seq_bounds, splice_site_model, dp_options_core,
dp_options_est, dp_options_postpro,
dna_complete_path_matrix_jt,
raw_chain->forward_jump_table, ref_offset, stat,
out->outfp);
}
else { /* call_protein_dp */
rval = gth_align_protein(sa, used_chain->forwardranges,
gth_input_current_gen_seq_tran(input),
ref_seq_tran, ref_seq_orig, ref_total_length,
gth_input_current_gen_alphabet(input),
gth_input_current_ref_alphabet(input),
input, useintroncutout,
introncutoutinfo->autoicmaxmatrixsize,
proteinexonpenal, out->showeops, out->comments,
out->gs2out, translationtable, gen_seq_bounds,
splice_site_model, dp_options_core,
dp_options_postpro,
protein_complete_path_matrix_jt,
raw_chain->forward_jump_table, ref_offset,
stat, out->outfp);
}
}
else {
/* the DP is called with the revers positions specifiers */
if (call_dna_dp) {
rval = gth_align_dna(sa, used_chain->reverseranges,
gth_input_current_gen_seq_tran_rc(input),
gth_input_current_gen_seq_orig_rc(input),
ref_seq_tran, ref_seq_orig, ref_total_length,
gth_input_current_gen_alphabet(input),
gth_input_current_ref_alphabet(input),
useintroncutout,
introncutoutinfo->autoicmaxmatrixsize,
out->showeops, out->comments, out->gs2out,
gen_seq_bounds_rc, splice_site_model,
dp_options_core, dp_options_est,
dp_options_postpro, dna_complete_path_matrix_jt,
raw_chain->reverse_jump_table, ref_offset, stat,
out->outfp);
}
else { /* call_protein_dp */
rval = gth_align_protein(sa, used_chain->reverseranges,
gth_input_current_gen_seq_tran_rc(input),
ref_seq_tran, ref_seq_orig, ref_total_length,
gth_input_current_gen_alphabet(input),
gth_input_current_ref_alphabet(input),
input, useintroncutout,
introncutoutinfo->autoicmaxmatrixsize,
proteinexonpenal, out->showeops, out->comments,
out->gs2out, translationtable, gen_seq_bounds,
splice_site_model, dp_options_core,
dp_options_postpro,
protein_complete_path_matrix_jt,
raw_chain->reverse_jump_table, ref_offset,
stat, out->outfp);
}
}
if (rval == GTH_ERROR_DP_PARAMETER_ALLOCATION_FAILED)
return GTH_ERROR_DP_PARAMETER_ALLOCATION_FAILED;
/* handling of special error codes ERROR_CUTOUT_NOT_IN_INTRON and
ERROR_MATRIX_ALLOCATION_FAILED from DP
the only possible special error code given back by this function is
ERROR_SA_COULD_NOT_BE_DETERMINED */
#ifndef NDEBUG
if (!useintroncutout) gt_assert(rval != GTH_ERROR_CUTOUT_NOT_IN_INTRON);
#endif
if (useintroncutout && rval == GTH_ERROR_CUTOUT_NOT_IN_INTRON) {
/* the intron cutout technique failed -> increase counter */
gth_stat_increment_numofunsuccessfulintroncutoutDPs(stat);
if (--iciterations > 0) {
/* if an iterations is left, increase icdelta, decrease the remaining
iterations, and continue the while-loop */
icdelta += introncutoutinfo->icdeltaincrease;
continue;
}
else {
/* no iteration left, discard SA */
gth_stat_increment_numofundeterminedSAs(stat);
gth_chain_delete(actual_chain);
gth_chain_delete(contracted_chain);
return GTH_ERROR_SA_COULD_NOT_BE_DETERMINED;
}
}
else if (rval == GTH_ERROR_MATRIX_ALLOCATION_FAILED) {
if (introncutoutinfo->autoicmaxmatrixsize > 0 && !useintroncutout) {
/* if the automatic intron cutout technique is enabled and a ``normal''
DP returned with the matrix allocation error, set useintroncutout,
increase counter, and continue */
if (out->showverbose) {
out->showverbose("matrix allocation failed, use intron cutout "
"technique");
}
gth_stat_increment_numofautointroncutoutcalls(stat);
useintroncutout = true;
continue;
}
else {
/* otherwise increase relevant statistics, free space and return with
error */
gth_stat_increment_numoffailedmatrixallocations(stat);
gth_stat_increment_numofundeterminedSAs(stat);
gth_chain_delete(actual_chain);
gth_chain_delete(contracted_chain);
return GTH_ERROR_SA_COULD_NOT_BE_DETERMINED;
}
}
else if (rval) /* ``normal'' DP */
return -1;
break;
}
#if 0
if (out->comments) {
gt_file_xprintf(out->outfp, "%c this SA has been computed:\n", COMMENTCHAR);
gth_sa_show(sa, input, out->outfp);
}
#endif
/* free */
gth_chain_delete(actual_chain);
gth_chain_delete(contracted_chain);
return 0;
}
