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");
}
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 showreferencefilename(GthSA *sa,
GthInput *input,
unsigned int indentlevel, GtFile *outfp)
{
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referencefile>\n");
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referencefilename>%s</referencefilename>\n",
gth_input_get_reference_filename(input,
gth_sa_ref_file_num(sa)));
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<referencefilehash>%s</referencefilehash>\n",
GTH_UNDEFINED_HASH);
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</referencefile>\n");
}
void gth_chaining(GthChainCollection *chain_collection,
GtUword gen_file_num,
GtUword ref_file_num,
GthCallInfo *call_info,
GthInput *input,
GthStat *stat,
bool directmatches,
const GthPlugins *plugins)
{
GtUword i, numofsequences = 0;
GtArray *matches;
GthChainingInfo chaining_info;
void *matcher_arguments;
GtFile *outfp = call_info->out->outfp;
GthMatchProcessorInfo match_processor_info;
bool refseqisdna = gth_input_ref_file_is_dna(input, ref_file_num);
/* make sure matcher is defined */
gt_assert(plugins);
gt_assert(plugins->matcher_arguments_new);
gt_assert(plugins->matcher_arguments_delete);
gt_assert(plugins->matcher_runner);
/* init */
matches = gt_array_new(sizeof (GthMatch));
chaining_info_init(&chaining_info, directmatches, refseqisdna, call_info,
input, stat, gen_file_num, ref_file_num);
matcher_arguments =
plugins->matcher_arguments_new(true,
input,
call_info->simfilterparam.inverse || !refseqisdna
? gth_input_get_genomic_filename(input, gen_file_num)
: gth_input_get_reference_filename(input,
ref_file_num),
call_info->simfilterparam.inverse || !refseqisdna
? gth_input_get_reference_filename(input,
ref_file_num)
: gth_input_get_genomic_filename(input, gen_file_num),
directmatches,
refseqisdna,
call_info->progname,
gt_str_get(gth_input_proteinsmap(input)),
call_info->simfilterparam.exact,
call_info->simfilterparam.edist,
false,
0,
call_info->simfilterparam.minmatchlength,
call_info->simfilterparam.seedlength,
call_info->simfilterparam.exdrop,
call_info->simfilterparam.prminmatchlen,
call_info->simfilterparam.prseedlength,
call_info->simfilterparam.prhdist,
call_info->translationtable,
call_info->simfilterparam.online,
call_info->simfilterparam.noautoindex,
call_info->simfilterparam.maskpolyAtails,
false);
match_processor_info_init(&match_processor_info, matches, chain_collection,
directmatches, refseqisdna,
call_info->simfilterparam.online,
call_info->simfilterparam.inverse, stat,
&chaining_info,
call_info->simfilterparam.maxnumofmatches,
call_info->simfilterparam.rare,
call_info->fragweightfactor,
plugins->jump_table_new,
plugins->jump_table_new_reverse,
plugins->jump_table_delete);
if (call_info->simfilterparam.maxnumofmatches > 0 ||
gth_stat_get_matchnumdistri(stat)) {
/* alloc space of match number counter */
numofsequences = gth_input_num_of_ref_seqs(input, ref_file_num);
match_processor_info.matchnumcounter = gt_malloc(sizeof (GtUword) *
numofsequences);
/* init match number counter to 0 */
memset(match_processor_info.matchnumcounter, 0,
(size_t) numofsequences * sizeof (GtUword));
}
/* free input, which contains the virtual trees.
because vmatch loads the virtual trees into memory, too.
this prevents that the virtual trees are loaded twice. */
gth_input_delete_current(input);
/* call matcher */
if (call_info->out->showverbose)
call_info->out->showverbose("call vmatch to compute matches");
plugins->matcher_runner(matcher_arguments, call_info->out->showverbose,
call_info->out->showverboseVM, &match_processor_info);
/* free matcher stuff here, because otherwise the reference file is mapped
twice below */
plugins->matcher_arguments_delete(matcher_arguments);
/* free sequence collections (if they have been filled by the matcher) */
gth_seq_con_delete(match_processor_info.gen_seq_con);
gth_seq_con_delete(match_processor_info.ref_seq_con);
/* save match numbers of match number distribution, if necessary */
if (gth_stat_get_matchnumdistri(stat)) {
for (i = 0; i < numofsequences; i++) {
if (match_processor_info.matchnumcounter[i] > 0) {
gth_stat_add_to_matchnumdistri(stat,
match_processor_info.matchnumcounter[i]);
}
}
}
/* free match number counter */
gt_free(match_processor_info.matchnumcounter);
/* return if no match has been found */
if (!gt_array_size(matches)) {
if (call_info->out->comments)
gt_file_xprintf(outfp, "%c no match has been found\n", COMMENTCHAR);
gt_array_delete(matches);
return;
}
/* load genomic file back into memory */
gth_input_load_genomic_file(input, gen_file_num, true);
/* load reference file back into memory */
gth_input_load_reference_file(input, ref_file_num, true);
/* compute chains from matches */
calc_chains_from_matches(chain_collection, matches, &chaining_info,
gth_input_current_gen_seq_con(input),
gth_input_current_ref_seq_con(input),
call_info->simfilterparam.rare,
call_info->fragweightfactor,
plugins->jump_table_new,
plugins->jump_table_new_reverse,
plugins->jump_table_delete);
if (call_info->out->showverbose) {
call_info->out->showverbose("sort global chains according to reference "
"sequence coverage");
}
/* sort chains */
gth_chain_collection_sort(chain_collection);
/* free */
gt_array_delete(matches);
}
static void show_xml_run_header(GthCallInfo *call_info, GthInput *input,
const char *timestring, const char *gth_version,
unsigned int indentlevel, const char **args)
{
GtFile *outfp = call_info->out->outfp;
GtUword i;
gth_indent(outfp, indentlevel);
if (call_info->intermediate) {
gt_file_xprintf(outfp, "<header xmlns=\"http://www.GenomeThreader.org/"
"SplicedAlignment/header/\">\n");
}
else {
gt_file_xprintf(outfp,
"<header xmlns=\"http://www.genomethreader.org/GTH_output/"
"header/\">\n");
}
/* at least one genomic file defined */
gt_assert(gth_input_num_of_gen_files(input));
/* at least one reference file defined */
gt_assert(gth_input_num_of_ref_files(input));
/* show a readable version of GthCallInfo. That is, it is shown with wich
parameters the program was called */
indentlevel++;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<source program=\"GenomeThreader\" version=\"%s\" "
"build_date=\"%s\" run_date=\"%s\"/>\n", gth_version,
GT_BUILT, timestring);
/* show genomic file names */
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<gDNA_template_files>\n");
indentlevel++;
for (i = 0; i < gth_input_num_of_gen_files(input); i++) {
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<temp_name>%s</temp_name>\n",
gth_input_get_genomic_filename(input, i));
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</gDNA_template_files>\n");
/* show reference file names */
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<reference_files>\n");
indentlevel++;
for (i = 0; i < gth_input_num_of_ref_files(input); i++) {
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<file ref_name=\"%s\" type=\"%s\"/>\n",
gth_input_get_reference_filename(input, i),
gth_input_get_alphatype(input, i) == DNA_ALPHA
? "ESTcDNA" : "Protein");
}
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</reference_files>\n");
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<splice_site_parameters parameter_type=\"%s\" "
"species=\"%s\"/>\n", SPLICE_SITE_MODEL_NAME,
call_info->speciesnum == NUMOFSPECIES
? GENERIC_SPECIES_NAME
: speciestab[call_info->speciesnum]);
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<parameters>\n");
indentlevel++;
/* output name of BSSM file */
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<parameter name=\"bssmfile\" value=\"%s\"/>\n",
gth_input_bssmfilename(input));
/* output name of scorematrix */
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<parameter name=\"scorematrixfile\" value=\"%s\"/>\n",
gt_str_get(call_info->scorematrixfile));
/* output searchmode */
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "<parameter name=\"searchmode\" "
"value=\"forward=%s,reverse=%s)\"/>\n",
GTH_SHOWBOOL(gth_input_forward(input)),
GTH_SHOWBOOL(gth_input_reverse(input)));
/* output arguments as comment */
gt_file_xprintf(outfp, "<!--\n%c Arguments: ", COMMENTCHAR);
gt_cstr_array_show_genfile(args, outfp);
gt_file_xprintf(outfp, "-->\n");
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</parameters>\n");
show_overall_reference_type(gth_input_overall_alphatype(input),
indentlevel, outfp);
indentlevel--;
gth_indent(outfp, indentlevel);
gt_file_xprintf(outfp, "</header>\n");
}
static int calc_spliced_alignments(GthSACollection *sa_collection,
GthChainCollection *chain_collection,
GthCallInfo *call_info,
GthInput *input,
GthStat *stat,
GtUword gen_file_num,
GtUword ref_file_num,
bool directmatches,
GthMatchInfo *match_info,
GthDNACompletePathMatrixJT
dna_complete_path_matrix_jt,
GthProteinCompletePathMatrixJT
protein_complete_path_matrix_jt)
{
const unsigned char *ref_seq_tran, *ref_seq_orig, *ref_seq_tran_rc = NULL,
*ref_seq_orig_rc = NULL;
GtUword chainctr, gen_offset = GT_UNDEF_UWORD, gen_total_length,
ref_total_length;
GtFile *outfp = call_info->out->outfp;
GtRange gen_seq_bounds, gen_seq_bounds_rc;
bool refseqisdna;
GthChain *chain;
GtRange range;
GthSA *saA;
int rval;
gt_assert(sa_collection && chain_collection);
refseqisdna = gth_input_ref_file_is_dna(input, ref_file_num);
for (chainctr = 0;
chainctr < gth_chain_collection_size(chain_collection);
chainctr++) {
chain = gth_chain_collection_get(chain_collection, chainctr);
if (++match_info->call_number > call_info->firstalshown &&
call_info->firstalshown > 0) {
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "<!--\n");
if (!call_info->out->gff3out) {
gt_file_xprintf(outfp, "Maximal matching %s count (%u) reached.\n",
refseqisdna ? "EST" : "protein",
call_info->firstalshown);
gt_file_xprintf(outfp, "Only the first %u matches will be "
"displayed.\n", call_info->firstalshown);
}
if (!(call_info->out->xmlout || call_info->out->gff3out))
gt_file_xfputc('\n', outfp);
else if (call_info->out->xmlout)
gt_file_xprintf(outfp, "-->\n");
match_info->max_call_number_reached = true;
break; /* break out of loop */
}
/* compute considered genomic regions if not set by -frompos */
if (!gth_input_use_substring_spec(input)) {
gen_seq_bounds = gth_input_get_genomic_range(input, chain->gen_file_num,
chain->gen_seq_num);
gen_total_length = gt_range_length(&gen_seq_bounds);
gen_offset = gen_seq_bounds.start;
gen_seq_bounds_rc = gen_seq_bounds;
}
else {
/* genomic multiseq contains exactly one sequence */
gt_assert(gth_input_num_of_gen_seqs(input, chain->gen_file_num) == 1);
gen_total_length = gth_input_genomic_file_total_length(input,
chain
->gen_file_num);
gen_seq_bounds.start = gth_input_genomic_substring_from(input);
gen_seq_bounds.end = gth_input_genomic_substring_to(input);
gen_offset = 0;
gen_seq_bounds_rc.start = gen_total_length - 1 - gen_seq_bounds.end;
gen_seq_bounds_rc.end = gen_total_length - 1 - gen_seq_bounds.start;
}
/* "retrieving" the reference sequence */
range = gth_input_get_reference_range(input, chain->ref_file_num,
chain->ref_seq_num);
ref_seq_tran = gth_input_current_ref_seq_tran(input) + range.start;
ref_seq_orig = gth_input_current_ref_seq_orig(input) + range.start;
if (refseqisdna) {
ref_seq_tran_rc = gth_input_current_ref_seq_tran_rc(input) + range.start;
ref_seq_orig_rc = gth_input_current_ref_seq_orig_rc(input) + range.start;
}
ref_total_length = range.end - range.start + 1;
/* check if protein sequences have a stop amino acid */
if (!refseqisdna && !match_info->stop_amino_acid_warning &&
ref_seq_orig[ref_total_length - 1] != GT_STOP_AMINO) {
GtStr *ref_id = gt_str_new();
gth_input_save_ref_id(input, ref_id, chain->ref_file_num,
chain->ref_seq_num);
gt_warning("protein sequence '%s' (#" GT_WU " in file %s) does not end "
"with a stop amino acid ('%c'). If it is not a protein "
"fragment you should add a stop amino acid to improve the "
"prediction. For example with `gt seqtransform "
"-addstopaminos` (see http://genometools.org for details).",
gt_str_get(ref_id), chain->ref_seq_num,
gth_input_get_reference_filename(input, chain->ref_file_num),
GT_STOP_AMINO);
match_info->stop_amino_acid_warning = true;
gt_str_delete(ref_id);
}
/* allocating space for alignment */
saA = gth_sa_new_and_set(directmatches, true, input, chain->gen_file_num,
chain->gen_seq_num, chain->ref_file_num,
chain->ref_seq_num, match_info->call_number,
gen_total_length, gen_offset, ref_total_length);
/* extend the DP borders to the left and to the right */
gth_chain_extend_borders(chain, &gen_seq_bounds, &gen_seq_bounds_rc,
gen_total_length, gen_offset);
/* From here on the dp positions always refer to the forward strand of the
genomic DNA. */
/* call the Dynamic Programming */
if (refseqisdna) {
rval = call_dna_DP(directmatches, call_info, input, stat,
sa_collection, saA, gen_file_num, ref_file_num,
gen_total_length, gen_offset, &gen_seq_bounds,
&gen_seq_bounds_rc, ref_total_length, range.start,
chainctr, gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig,
ref_seq_tran_rc, ref_seq_orig_rc, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
else {
rval = call_protein_DP(directmatches, call_info, input,
stat, sa_collection, saA, gen_file_num,
ref_file_num, gen_total_length, gen_offset,
&gen_seq_bounds, &gen_seq_bounds_rc,
ref_total_length, range.start, chainctr,
gth_chain_collection_size(chain_collection),
match_info, ref_seq_tran, ref_seq_orig, chain,
dna_complete_path_matrix_jt,
protein_complete_path_matrix_jt);
}
/* check return value */
if (rval == GTH_ERROR_DP_PARAMETER_ALLOCATION_FAILED) {
/* statistics bookkeeping */
gth_stat_increment_numoffailedDPparameterallocations(stat);
gth_stat_increment_numofundeterminedSAs(stat);
/* free space */
gth_sa_delete(saA);
match_info->call_number--;
continue; /* continue with the next DP range */
}
else if (rval)
return -1;
}
if (!call_info->out->xmlout && !call_info->out->gff3out && !directmatches &&
!match_info->significant_match_found &&
match_info->call_number <= call_info->firstalshown) {
show_no_match_line(gth_input_get_alphatype(input, ref_file_num), outfp);
}
return 0;
}
