int gt_parseargsandcallmkfmindex(int argc,const char **argv,GtError *err)
{
Mkfmcallinfo mkfmcallinfo;
int retval;
bool haserr = false;
retval = mkfmindexoptions(&mkfmcallinfo,argc,argv,err);
if (retval == 0)
{
GtLogger *logger = gt_logger_new(false, GT_LOGGER_DEFLT_PREFIX, stdout);
if (runmkfmindex(&mkfmcallinfo,logger,err) < 0)
{
haserr = true;
}
gt_logger_delete(logger);
logger = NULL;
} else
{
if (retval < 0)
{
haserr = true;
}
}
freemkfmcallinfo(&mkfmcallinfo);
return haserr ? -1 : 0;
}
static void infer_cds_visitor_test_data(GtQueue *queue)
{
GtError *error = gt_error_new();
const char *file = "data/gff3/grape-codons.gff3";
GtNodeStream *gff3in = gt_gff3_in_stream_new_unsorted(1, &file);
gt_gff3_in_stream_check_id_attributes((GtGFF3InStream *)gff3in);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream *)gff3in);
GtLogger *logger = gt_logger_new(true, "", stderr);
GtNodeStream *icv_stream = agn_infer_cds_stream_new(gff3in, NULL, logger);
GtArray *feats = gt_array_new( sizeof(GtFeatureNode *) );
GtNodeStream *arraystream = gt_array_out_stream_new(icv_stream, feats, error);
int pullresult = gt_node_stream_pull(arraystream, error);
if(pullresult == -1)
{
fprintf(stderr, "[AgnInferCDSVisitor::infer_cds_visitor_test_data] error "
"processing features: %s\n", gt_error_get(error));
}
gt_node_stream_delete(gff3in);
gt_node_stream_delete(icv_stream);
gt_node_stream_delete(arraystream);
gt_logger_delete(logger);
gt_array_sort(feats, (GtCompare)agn_genome_node_compare);
gt_array_reverse(feats);
while(gt_array_size(feats) > 0)
{
GtFeatureNode *fn = *(GtFeatureNode **)gt_array_pop(feats);
gt_queue_add(queue, fn);
}
gt_array_delete(feats);
gt_error_delete(error);
}
static int encode_sequence_files(GtStrArray *infiles, GtEncseqOptions *opts,
const char *indexname, bool verbose,
bool esq_no_header,
GtError *err)
{
GtEncseqEncoder *encseq_encoder;
GtLogger *logger;
int had_err = 0;
gt_error_check(err);
gt_assert(infiles && gt_str_array_size(infiles) > 0 && opts);
logger = gt_logger_new(verbose, "# ", stderr);
encseq_encoder = gt_encseq_encoder_new_from_options(opts, err);
if (!encseq_encoder)
had_err = -1;
if (!had_err) {
gt_encseq_encoder_set_logger(encseq_encoder, logger);
if (esq_no_header)
{
gt_encseq_encoder_disable_esq_header(encseq_encoder);
}
had_err = gt_encseq_encoder_encode(encseq_encoder, infiles, indexname, err);
}
gt_encseq_encoder_delete(encseq_encoder);
gt_logger_delete(logger);
return had_err;
}
static void gv_test_calc_integrity(AgnUnitTest *test)
{
const char *filename = "data/gff3/gaeval-stream-unit-test-2.gff3";
GtNodeStream *align_in = gt_gff3_in_stream_new_unsorted(1, &filename);
AgnGaevalParams params = { 0.6, 0.3, 0.05, 0.05, 400, 200, 100 };
GtNodeVisitor *nv = agn_gaeval_visitor_new(align_in, params);
AgnGaevalVisitor *gv = gaeval_visitor_cast(nv);
gt_node_stream_delete(align_in);
GtNodeStream *gff3in = gt_gff3_in_stream_new_unsorted(1, &filename);
GtHashmap *typestokeep = gt_hashmap_new(GT_HASH_STRING, NULL, NULL);
gt_hashmap_add(typestokeep, "mRNA", "mRNA");
GtNodeStream *filtstream = agn_filter_stream_new(gff3in, typestokeep);
GtLogger *logger = gt_logger_new(true, "", stderr);
GtNodeStream *ics = agn_infer_cds_stream_new(filtstream, NULL, logger);
GtNodeStream *ies = agn_infer_exons_stream_new(ics, NULL, logger);
GtError *error = gt_error_new();
GtArray *feats = gt_array_new( sizeof(GtFeatureNode *) );
GtNodeStream *featstream = gt_array_out_stream_new(ies, feats, error);
int result = gt_node_stream_pull(featstream, error);
if(result == -1)
{
fprintf(stderr, "[AgnGaevalVisitor::gv_test_calc_integrity] error "
"processing GFF3: %s\n", gt_error_get(error));
return;
}
gt_node_stream_delete(gff3in);
gt_node_stream_delete(filtstream);
gt_node_stream_delete(featstream);
gt_node_stream_delete(ics);
gt_node_stream_delete(ies);
gt_logger_delete(logger);
gt_hashmap_delete(typestokeep);
agn_assert(gt_array_size(feats) == 2);
GtFeatureNode *g1 = *(GtFeatureNode **)gt_array_get(feats, 0);
GtFeatureNode *g2 = *(GtFeatureNode **)gt_array_get(feats, 1);
double cov1 = gaeval_visitor_calculate_coverage(gv, g1, error);
double cov2 = gaeval_visitor_calculate_coverage(gv, g2, error);
double int1 = gaeval_visitor_calculate_integrity(gv, g1, cov1, NULL, error);
double int2 = gaeval_visitor_calculate_integrity(gv, g2, cov2, NULL, error);
bool test1 = fabs(cov1 - 1.000) < 0.001 &&
fabs(cov2 - 0.997) < 0.001 &&
fabs(int1 - 0.850) < 0.001 &&
fabs(int2 - 0.863) < 0.001;
agn_unit_test_result(test, "calculate integrity", test1);
gt_error_delete(error);
gt_array_delete(feats);
gt_genome_node_delete((GtGenomeNode *)g1);
gt_genome_node_delete((GtGenomeNode *)g2);
gt_node_visitor_delete(nv);
}
static int gt_condenseq_hmmsearch_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtCondenseqHmmsearchArguments *arguments = tool_arguments;
GtCondenseq *ces = NULL;
GtStr *table_filename = NULL;
GtLogger *logger = NULL;
int had_err = 0;
logger = gt_logger_new(gt_condenseq_search_arguments_verbose(arguments->csa),
GT_LOGGER_DEFLT_PREFIX, stderr);
gt_error_check(err);
gt_assert(arguments);
table_filename = gt_condenseq_search_arguments_db_filename(arguments->csa,
"_tabout.tsv");
if (!had_err) {
struct stat buf;
if (stat(gt_str_get(table_filename), &buf) == 0 && !arguments->force_ow) {
gt_error_set(err, "file %s already exists, use option -force_ow to "
"overwrite or delete", gt_str_get(table_filename));
had_err = -1;
}
}
if (!had_err) {
ces = gt_condenseq_search_arguments_read_condenseq(arguments->csa,
logger, err);
if (ces == NULL)
had_err = -1;
}
if (!had_err) {
had_err =
hmmsearch_call_coarse_search(ces,
gt_str_get(arguments->hmmsearch_path),
gt_str_get(table_filename),
gt_str_get(arguments->hmm),
logger, err);
}
if (!had_err) {
had_err = hmmsearch_process_coarse_hits(gt_str_get(table_filename),
ces, arguments,
logger, err);
}
gt_condenseq_delete(ces);
gt_logger_delete(logger);
gt_str_delete(table_filename);
return had_err;
}
int gt_mergeesa(int argc, const char **argv, GtError *err)
{
GtStr *storeindex;
GtStrArray *indexnametab;
bool haserr = false;
int parsed_args;
gt_error_check(err);
storeindex = gt_str_new();
indexnametab = gt_str_array_new();
switch (parse_options(storeindex, indexnametab, &parsed_args, argc, argv,
err)) {
case GT_OPTION_PARSER_OK:
break;
case GT_OPTION_PARSER_ERROR:
haserr = true;
break;
case GT_OPTION_PARSER_REQUESTS_EXIT:
return 0;
}
if (!haserr)
{
GtUword i;
GtLogger *logger;
printf("# storeindex=%s\n",gt_str_get(storeindex));
for (i=0; i<gt_str_array_size(indexnametab); i++)
{
printf("# input=%s\n",gt_str_array_get(indexnametab,i));
}
logger = gt_logger_new(false, GT_LOGGER_DEFLT_PREFIX, stdout);
if (gt_performtheindexmerging(storeindex,
indexnametab,
logger,
err) != 0)
{
haserr = true;
}
gt_logger_delete(logger);
}
gt_str_delete(storeindex);
gt_str_array_delete(indexnametab);
return haserr ? -1 : 0;
}
static int gt_cge_spacedseed_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
Cge_spacedseed_options *arguments = tool_arguments;
GtLogger *logger = NULL;
bool haserr = false;
gt_assert(parsed_args == argc);
logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX, stdout);
if (arguments->verbose)
{
unsigned long idx;
printf("# %sindex=%s\n",arguments->withesa ? "esa" : "pck",
gt_str_get(arguments->str_inputindex));
for (idx = 0; idx < gt_str_array_size(arguments->queryfilenames); idx++)
{
printf("# queryfile=%s\n",
gt_str_array_get(arguments->queryfilenames,idx));
}
}
if (gt_matchspacedseed(arguments->withesa,
arguments->docompare,
gt_str_get(arguments->str_inputindex),
arguments->queryfilenames,
arguments->verbose,
err) != 0)
{
haserr = true;
}
gt_logger_delete(logger);
return haserr ? - 1 : 0;
}
extern int
gt_packedindex_chk_search(int argc, const char *argv[], GtError *err)
{
struct chkSearchOptions params;
Suffixarray suffixarray;
Enumpatterniterator *epi = NULL;
bool saIsLoaded = false;
BWTSeq *bwtSeq = NULL;
GtStr *inputProject = NULL;
int parsedArgs;
bool had_err = false;
BWTSeqExactMatchesIterator EMIter;
bool EMIterInitialized = false;
GtLogger *logger = NULL;
inputProject = gt_str_new();
do {
gt_error_check(err);
{
bool exitNow = false;
switch (parseChkBWTOptions(&parsedArgs, argc, argv, ¶ms,
inputProject, err))
{
case GT_OPTION_PARSER_OK:
break;
case GT_OPTION_PARSER_ERROR:
had_err = true;
exitNow = true;
break;
case GT_OPTION_PARSER_REQUESTS_EXIT:
exitNow = true;
break;
}
if (exitNow)
break;
}
gt_str_set(inputProject, argv[parsedArgs]);
logger = gt_logger_new(params.verboseOutput,
GT_LOGGER_DEFLT_PREFIX, stdout);
bwtSeq = gt_availBWTSeq(¶ms.idx.final, logger, err);
if ((had_err = bwtSeq == NULL))
break;
{
enum verifyBWTSeqErrCode retval =
gt_BWTSeqVerifyIntegrity(bwtSeq, gt_str_get(inputProject), params.flags,
params.progressInterval, stderr, logger, err);
if ((had_err = (retval != VERIFY_BWTSEQ_NO_ERROR)))
{
fprintf(stderr, "index integrity check failed: %s\n",
gt_error_get(err));
gt_error_set(err, "aborted because of index integrity check fail");
break;
}
}
if (BWTSeqHasLocateInformation(bwtSeq))
{
if ((had_err = !gt_initEmptyEMIterator(&EMIter, bwtSeq)))
{
gt_error_set(err, "Cannot create matches iterator for sequence index.");
break;
}
EMIterInitialized = true;
}
{
unsigned long totalLen, dbstart;
unsigned long trial, patternLen;
if ((had_err =
gt_mapsuffixarray(&suffixarray, SARR_SUFTAB | SARR_ESQTAB,
gt_str_get(inputProject), NULL, err) != 0))
{
gt_error_set(err, "Can't load suffix array project with"
" demand for encoded sequence and suffix table files\n");
break;
}
totalLen = gt_encseq_total_length(suffixarray.encseq);
saIsLoaded = true;
if ((had_err = (params.minPatLen >= 0L && params.maxPatLen >= 0L
&& params.minPatLen > params.maxPatLen)))
{
gt_error_set(err, "Invalid pattern lengths selected: min=%ld, max=%ld;"
" min <= max is required.", params.minPatLen,
params.maxPatLen);
break;
}
if (params.minPatLen < 0 || params.maxPatLen < 0)
{
unsigned int numofchars
= gt_alphabet_num_of_chars(
gt_encseq_alphabet(suffixarray.encseq));
if (params.minPatLen < 0)
params.minPatLen
= gt_recommendedprefixlength(numofchars,
totalLen,
GT_RECOMMENDED_MULTIPLIER_DEFAULT,
true);
if (params.maxPatLen < 0)
params.maxPatLen =
MAX(params.minPatLen,
125 * gt_recommendedprefixlength(numofchars,totalLen,
GT_RECOMMENDED_MULTIPLIER_DEFAULT,
true)/100);
else
params.maxPatLen = MAX(params.maxPatLen, params.minPatLen);
}
fprintf(stderr, "Using patterns of lengths %lu to %lu\n",
params.minPatLen, params.maxPatLen);
if ((had_err = totalLen + 1 != BWTSeqLength(bwtSeq)))
{
gt_error_set(err, "base suffix array and index have diferrent lengths!"
"%lu vs. %lu", totalLen + 1,
BWTSeqLength(bwtSeq));
break;
}
if ((had_err =
(epi = gt_newenumpatterniterator(params.minPatLen, params.maxPatLen,
suffixarray.encseq,
err)) == NULL))
{
fputs("Creation of pattern iterator failed!\n", stderr);
break;
}
for (trial = 0; !had_err && trial < params.numOfSamples; ++trial)
{
const GtUchar *pptr = gt_nextEnumpatterniterator(&patternLen, epi);
GtMMsearchiterator *mmsi =
gt_mmsearchiterator_new_complete_olain(suffixarray.encseq,
suffixarray.suftab,
0, /* leftbound */
totalLen, /* rightbound */
0, /* offset */
suffixarray.readmode,
pptr,
patternLen);
if (BWTSeqHasLocateInformation(bwtSeq))
{
if ((had_err = !gt_reinitEMIterator(&EMIter, bwtSeq, pptr, patternLen,
false)))
{
fputs("Internal error: failed to reinitialize pattern match"
" iterator", stderr);
abort();
}
gt_assert(gt_EMINumMatchesTotal(&EMIter) ==
gt_BWTSeqMatchCount(bwtSeq, pptr, patternLen,
false));
gt_assert(gt_EMINumMatchesTotal(&EMIter)
== gt_mmsearchiterator_count(mmsi));
while (gt_mmsearchiterator_next(&dbstart,mmsi))
{
unsigned long matchPos = 0;
bool match = EMIGetNextMatch(&EMIter, &matchPos, bwtSeq);
if ((had_err = !match))
{
gt_error_set(err,
"matches of packedindex expired before mmsearch!");
break;
}
if ((had_err = matchPos != dbstart))
{
gt_error_set(err, "packedindex match doesn't equal mmsearch "
"match result!\n%lu vs. %lu\n",
matchPos, dbstart);
}
}
if (!had_err)
{
unsigned long matchPos;
bool trailingMatch = EMIGetNextMatch(&EMIter, &matchPos, bwtSeq);
if ((had_err = trailingMatch))
{
gt_error_set(err, "matches of mmsearch expired before fmindex!");
break;
}
}
}
else
{
unsigned long numFMIMatches = gt_BWTSeqMatchCount(bwtSeq, pptr,
patternLen,
false),
numMMSearchMatches = gt_mmsearchiterator_count(mmsi);
if ((had_err = numFMIMatches != numMMSearchMatches))
{
gt_error_set(err, "Number of matches not equal for suffix array ("
"%lu) and fmindex (%lu).\n",
numFMIMatches, numMMSearchMatches);
}
}
gt_mmsearchiterator_delete(mmsi);
mmsi = NULL;
if (params.progressInterval && !((trial + 1) % params.progressInterval))
putc('.', stderr);
}
if (params.progressInterval)
putc('\n', stderr);
fprintf(stderr, "Finished %lu of %lu matchings successfully.\n",
trial, params.numOfSamples);
}
} while (0);
if (EMIterInitialized) gt_destructEMIterator(&EMIter);
if (saIsLoaded) gt_freesuffixarray(&suffixarray);
gt_freeEnumpatterniterator(epi);
if (bwtSeq) gt_deleteBWTSeq(bwtSeq);
if (logger) gt_logger_delete(logger);
if (inputProject) gt_str_delete(inputProject);
return had_err?-1:0;
}
static int gt_readjoiner_assembly_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv, GT_UNUSED int parsed_args,
void *tool_arguments, GtError *err)
{
GtReadjoinerAssemblyArguments *arguments = tool_arguments;
GtLogger *verbose_logger, *default_logger;
GtEncseqLoader *el;
GtEncseq *reads;
GtTimer *timer = NULL;
GtStrgraph *strgraph = NULL;
GtBitsequence *contained = NULL;
const char *readset = gt_str_get(arguments->readset);
bool eqlen = true;
GtUword nreads, tlen, rlen;
int had_err = 0;
gt_assert(arguments);
gt_error_check(err);
default_logger = gt_logger_new(!arguments->quiet, GT_LOGGER_DEFLT_PREFIX,
stdout);
gt_logger_log(default_logger,
"gt readjoiner assembly (version "GT_READJOINER_VERSION")");
verbose_logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX,
stdout);
gt_logger_log(verbose_logger, "verbose output activated");
gt_logger_log(verbose_logger, "readset name = %s", readset);
if (gt_showtime_enabled())
{
timer = gt_timer_new_with_progress_description(
GT_READJOINER_ASSEMBLY_MSG_COUNTSPM);
gt_timer_start(timer);
gt_timer_show_cpu_time_by_progress(timer);
}
if (!arguments->paths2seq)
{
el = gt_encseq_loader_new();
gt_encseq_loader_drop_description_support(el);
gt_encseq_loader_disable_autosupport(el);
reads = gt_encseq_loader_load(el, readset, err);
if (reads == NULL)
{
had_err = -1;
}
if (had_err == 0)
{
eqlen = gt_encseq_accesstype_get(reads) == GT_ACCESS_TYPE_EQUALLENGTH;
nreads = gt_encseq_num_of_sequences(reads);
gt_logger_log(default_logger, "number of reads in filtered readset = "
GT_WU, nreads);
tlen = gt_encseq_total_length(reads) - nreads + 1;
gt_logger_log(verbose_logger, "total length of filtered readset = " GT_WU,
tlen);
if (eqlen)
{
rlen = gt_encseq_seqlength(reads, 0);
gt_logger_log(verbose_logger, "read length = " GT_WU, rlen);
gt_encseq_delete(reads);
reads = NULL;
}
else
{
had_err = gt_readjoiner_assembly_build_contained_reads_list(
arguments, &contained, err);
rlen = 0;
gt_logger_log(verbose_logger, "read length = variable");
gt_assert(reads != NULL);
}
}
if (had_err == 0)
{
if (!arguments->load)
{
had_err = gt_readjoiner_assembly_build_graph(arguments, &strgraph,
reads, readset, eqlen, rlen, nreads, contained, default_logger,
verbose_logger, timer, err);
}
else
{
gt_readjoiner_assembly_load_graph(&strgraph, reads, readset, rlen,
default_logger, timer);
}
}
if (!eqlen && reads != NULL && !arguments->errors)
{
gt_encseq_delete(reads);
reads = NULL;
if (had_err == 0)
gt_strgraph_set_encseq(strgraph, NULL);
}
if (had_err == 0 && arguments->redtrans)
{
if (gt_showtime_enabled())
gt_timer_show_progress(timer, GT_READJOINER_ASSEMBLY_MSG_REDTRANS,
stdout);
gt_strgraph_sort_edges_by_len(strgraph, false);
(void)gt_strgraph_redtrans(strgraph, false);
(void)gt_strgraph_redself(strgraph, false);
(void)gt_strgraph_redwithrc(strgraph, false);
gt_strgraph_log_stats(strgraph, verbose_logger);
}
if (had_err == 0 && arguments->errors)
{
if (gt_showtime_enabled())
gt_timer_show_progress(timer, GT_READJOINER_ASSEMBLY_MSG_CLEANSG,
stdout);
gt_logger_log(default_logger, GT_READJOINER_ASSEMBLY_MSG_CLEANSG);
had_err = gt_readjoiner_assembly_error_correction(strgraph,
arguments->bubble, arguments->deadend, arguments->deadend_depth,
verbose_logger);
}
if (had_err == 0 && arguments->save)
{
if (gt_showtime_enabled())
gt_timer_show_progress(timer, GT_READJOINER_ASSEMBLY_MSG_SAVESG,
stdout);
gt_logger_log(default_logger, GT_READJOINER_ASSEMBLY_MSG_SAVESG);
gt_strgraph_show(strgraph, GT_STRGRAPH_BIN,
gt_str_get(arguments->readset), GT_READJOINER_SUFFIX_SG, false);
}
if (!eqlen && reads != NULL)
{
gt_encseq_delete(reads);
reads = NULL;
if (had_err == 0)
gt_strgraph_set_encseq(strgraph, NULL);
}
if (had_err == 0)
{
if (gt_showtime_enabled())
gt_timer_show_progress(timer, GT_READJOINER_ASSEMBLY_MSG_TRAVERSESG,
stdout);
gt_logger_log(default_logger, GT_READJOINER_ASSEMBLY_MSG_TRAVERSESG);
gt_readjoiner_assembly_show_current_space("(before traversal)");
gt_strgraph_spell(strgraph, (GtUword)arguments->depthcutoff,
(GtUword)arguments->lengthcutoff, arguments->vd, readset,
GT_READJOINER_SUFFIX_CONTIG_PATHS, NULL, true,
arguments->show_contigs_info, false, verbose_logger);
}
if (contained != NULL)
gt_free(contained);
gt_strgraph_delete(strgraph);
strgraph = NULL;
gt_assert(reads == NULL);
gt_encseq_loader_delete(el);
}
if (had_err == 0)
{
gt_readjoiner_assembly_show_current_space("(before paths2seq)");
had_err = gt_readjoiner_assembly_paths2seq(readset,
(GtUword)arguments->lengthcutoff, arguments->vd,
arguments->astat, arguments->coverage, arguments->copynum,
arguments->buffersize, default_logger, &timer, err);
}
if (gt_showtime_enabled())
{
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
gt_logger_delete(default_logger);
gt_logger_delete(verbose_logger);
return had_err;
}
static int gt_ltrdigest_runner(GT_UNUSED int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
GtLTRdigestOptions *arguments = tool_arguments;
GtNodeStream *gff3_in_stream = NULL,
*gff3_out_stream = NULL,
*ltrdigest_stream = NULL,
*tab_out_stream = NULL,
*last_stream = NULL;
int had_err = 0,
tests_to_run = 0,
arg = parsed_args;
const char *indexname = argv[arg+1];
GtLogger *logger = gt_logger_new(arguments->verbose,
GT_LOGGER_DEFLT_PREFIX, stdout);
GtEncseqLoader *el;
GtEncseq *encseq;
gt_error_check(err);
gt_assert(arguments);
/* Set sequence encoder options. Defaults are ok. */
el = gt_encseq_loader_new();
gt_encseq_loader_set_logger(el, logger);
/* Open sequence file */
encseq = gt_encseq_loader_load(el, indexname, err);
if (!encseq)
had_err = -1;
/* Always search for PPT. */
tests_to_run |= GT_LTRDIGEST_RUN_PPT;
/* Open tRNA library if given. */
if (!had_err && arguments->trna_lib
&& gt_str_length(arguments->trna_lib) > 0)
{
tests_to_run |= GT_LTRDIGEST_RUN_PBS;
arguments->pbs_opts.trna_lib = gt_bioseq_new(gt_str_get(arguments->trna_lib),
err);
if (gt_error_is_set(err))
had_err = -1;
}
#ifdef HAVE_HMMER
/* Open HMMER files if given. */
if (!had_err && gt_str_array_size(arguments->pdom_opts.hmm_files) > 0)
{
tests_to_run |= GT_LTRDIGEST_RUN_PDOM;
if (!strcmp(gt_str_get(arguments->cutoffs), "GA")) {
arguments->pdom_opts.cutoff = GT_PHMM_CUTOFF_GA;
} else if (!strcmp(gt_str_get(arguments->cutoffs), "TC")) {
arguments->pdom_opts.cutoff = GT_PHMM_CUTOFF_TC;
} else if (!strcmp(gt_str_get(arguments->cutoffs), "NONE")) {
arguments->pdom_opts.cutoff = GT_PHMM_CUTOFF_NONE;
} else {
gt_error_set(err, "invalid cutoff setting!");
had_err = -1;
}
}
#endif
if (!had_err)
{
/* set up stream flow
* ------------------*/
last_stream = gff3_in_stream = gt_gff3_in_stream_new_sorted(argv[arg]);
last_stream = ltrdigest_stream = gt_ltrdigest_stream_new(last_stream,
tests_to_run,
encseq,
&arguments->pbs_opts,
&arguments->ppt_opts,
#ifdef HAVE_HMMER
&arguments->pdom_opts,
#endif
err);
if (!ltrdigest_stream)
had_err = -1;
}
if (!had_err)
{
/* attach tabular output stream, if requested */
if (gt_str_length(arguments->prefix) > 0)
{
last_stream = tab_out_stream = gt_ltr_fileout_stream_new(last_stream,
tests_to_run,
encseq,
gt_str_get(arguments->prefix),
&arguments->ppt_opts,
&arguments->pbs_opts,
#ifdef HAVE_HMMER
&arguments->pdom_opts,
#endif
gt_str_get(arguments->trna_lib),
argv[arg+1],
argv[arg],
arguments->seqnamelen,
err);
#ifdef HAVE_HMMER
if (&arguments->pdom_opts.write_alignments)
gt_ltr_fileout_stream_enable_pdom_alignment_output(tab_out_stream);
if (&arguments->pdom_opts.write_aaseqs)
gt_ltr_fileout_stream_enable_aa_sequence_output(tab_out_stream);
#endif
}
last_stream = gff3_out_stream = gt_gff3_out_stream_new(last_stream,
arguments->outfp);
/* pull the features through the stream and free them afterwards */
had_err = gt_node_stream_pull(last_stream, err);
}
gt_node_stream_delete(gff3_out_stream);
gt_node_stream_delete(ltrdigest_stream);
if (tab_out_stream != NULL)
gt_node_stream_delete(tab_out_stream);
gt_node_stream_delete(gff3_in_stream);
gt_encseq_loader_delete(el);
gt_encseq_delete(encseq);
encseq = NULL;
gt_bioseq_delete(arguments->pbs_opts.trna_lib);
gt_logger_delete(logger);
return had_err;
}
static int gt_condenser_search_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtCondenserSearchArguments *arguments = tool_arguments;
int i, had_err = 0;
char *querypath = gt_str_get(arguments->querypath);
GtStr* coarse_fname = gt_str_new_cstr("coarse_");
char *db_basename = NULL;
char *suffix_ptr = NULL;
GtTimer *timer = NULL;
GtLogger *logger = NULL;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX, stderr);
db_basename = gt_basename(gt_str_get(arguments->dbpath));
/* if first char is '.' this might be a hidden file */
if (strlen(db_basename) > (size_t) 1 &&
(suffix_ptr = strrchr(db_basename + 1, '.')) != NULL) {
/* remove suffix */
*suffix_ptr = '\0';
}
gt_str_append_cstr(coarse_fname, db_basename);
gt_str_append_cstr(coarse_fname, ".fas");
gt_free(db_basename);
db_basename = NULL;
suffix_ptr = NULL;
if (arguments->blastn || arguments->blastp) {
GtMatch *match;
GtMatchIterator *mp = NULL;
GtNREncseq *nrencseq = NULL;
GtStr *fastaname = gt_str_clone(arguments->dbpath);
HitPosition *hits;
double eval,
raw_eval = 0.0;
GtUword coarse_db_len = 0;
GtMatchIteratorStatus status;
int curr_hits = 0,
max_hits = 100;
hits = gt_malloc(sizeof (*hits) * (size_t) max_hits);
gt_str_append_cstr(fastaname, ".fas");
for (i=0; i < max_hits; i++) {
hits[i].range = gt_malloc(sizeof (*hits[i].range) * (size_t) 1);
}
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("initialization");
gt_timer_start(timer);
}
/*extract sequences from compressed database*/
if (!had_err) {
nrencseq = gt_n_r_encseq_new_from_file(gt_str_get(arguments->dbpath),
logger, err);
if (nrencseq == NULL)
had_err = -1;
}
if (!had_err) {
if (arguments->ceval == GT_UNDEF_DOUBLE ||
arguments->feval == GT_UNDEF_DOUBLE) {
/* from NCBI BLAST tutorial:
E = Kmne^{-lambdaS}
calculates E-value for score S with natural scale parameters K for
search space size and lambda for the scoring system
E = mn2^-S'
m being the subject (total) length, n the length of ONE query
calculates E-value for bit-score S'
*/
GtFastaReader *reader;
GtCondenserSearchAvg avg = {0,0};
reader = gt_fasta_reader_rec_new(arguments->querypath);
had_err = gt_fasta_reader_run(reader, NULL, NULL,
gt_condenser_search_cum_moving_avg,
&avg,
err);
if (!had_err) {
GtUword S = arguments->bitscore;
gt_log_log(GT_WU " queries, avg query size: " GT_WU,
avg.count, avg.avg);
raw_eval = 1/pow(2.0, (double) S) * avg.avg;
gt_logger_log(logger, "Raw E-value set to %.4e", raw_eval);
gt_assert(avg.avg != 0);
}
gt_fasta_reader_delete(reader);
}
}
/*create BLAST database from compressed database fasta file*/
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "create coarse BLAST db", stderr);
if (arguments->blastn)
had_err = gt_condenser_search_create_nucl_blastdb(gt_str_get(fastaname),
err);
else
had_err = gt_condenser_search_create_prot_blastdb(gt_str_get(fastaname),
err);
}
if (!had_err) {
GtBlastProcessCall *call;
if (timer != NULL)
gt_timer_show_progress(timer, "coarse BLAST run", stderr);
if (arguments->blastp)
call = gt_blast_process_call_new_prot();
else
call = gt_blast_process_call_new_nucl();
gt_blast_process_call_set_db(call, gt_str_get(fastaname));
gt_blast_process_call_set_query(call, querypath);
gt_blast_process_call_set_evalue(call, arguments->ceval);
gt_blast_process_call_set_num_threads(call, arguments->blthreads);
mp = gt_match_iterator_blast_process_new(call, err);
if (!mp)
had_err = -1;
gt_blast_process_call_delete(call);
while (!had_err &&
(status = gt_match_iterator_next(mp, &match, err)) !=
GT_MATCHER_STATUS_END)
{
if (status == GT_MATCHER_STATUS_OK) {
GtUword hit_seq_id;
char string[7];
const char *dbseqid = gt_match_get_seqid2(match);
if (sscanf(dbseqid,"%6s" GT_WU, string, &hit_seq_id) == 2) {
gt_match_get_range_seq2(match, hits[curr_hits].range);
hits[curr_hits].idx = hit_seq_id;
gt_match_delete(match);
curr_hits++;
if (curr_hits == max_hits) {
HitPosition *hit_extention;
max_hits += 100;
hits = gt_realloc(hits, sizeof (*hit_extention) * max_hits);
for (i=max_hits - 100; i < max_hits; i++) {
hits[i].range = gt_malloc(sizeof (*hits[i].range));
}
}
} else {
gt_error_set(err, "could not parse unique db header %s", dbseqid);
had_err = -1;
}
} else if (status == GT_MATCHER_STATUS_ERROR) {
had_err = -1;
}
}
gt_match_iterator_delete(mp);
}
/*extract sequences*/
if (!had_err) {
GtNREncseqDecompressor *decomp;
GtFile *coarse_hits;
if (timer != NULL)
gt_timer_show_progress(timer, "extract coarse search hits", stderr);
decomp = gt_n_r_encseq_decompressor_new(nrencseq);
coarse_hits = gt_file_new(gt_str_get(coarse_fname),"w", err);
/* TODO DW do NOT extract complete uniques! these could be complete
chromosomes!! just extract something around it? maybe +- max query
length*/
for (i = 0; i < curr_hits; i++) {
gt_n_r_encseq_decompressor_add_unique_idx_to_extract(decomp,
hits[i].idx);
}
had_err =
gt_n_r_encseq_decompressor_start_unique_extraction(coarse_hits,
decomp,
&coarse_db_len,
err);
gt_assert(coarse_db_len != 0);
gt_file_delete(coarse_hits);
gt_n_r_encseq_decompressor_delete(decomp);
}
gt_n_r_encseq_delete(nrencseq);
/* create BLAST database from decompressed database file */
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "create fine BLAST db", stderr);
if (arguments->blastn)
had_err =
gt_condenser_search_create_nucl_blastdb(gt_str_get(coarse_fname),
err);
else
had_err =
gt_condenser_search_create_prot_blastdb(gt_str_get(coarse_fname),
err);
}
/* perform fine BLAST search */
if (!had_err) {
GtBlastProcessCall *call;
if (timer != NULL)
gt_timer_show_progress(timer, "fine BLAST run", stderr);
if (arguments->feval == GT_UNDEF_DOUBLE) {
eval = raw_eval * coarse_db_len;
} else {
eval = arguments->feval;
}
if (arguments->blastp)
call = gt_blast_process_call_new_prot();
else
call = gt_blast_process_call_new_nucl();
gt_blast_process_call_set_db(call, gt_str_get(coarse_fname));
gt_blast_process_call_set_query(call, querypath);
gt_blast_process_call_set_evalue(call, eval);
gt_blast_process_call_set_num_threads(call, arguments->blthreads);
gt_logger_log(logger, "Fine E-value set to: %.4e (len)" GT_WU, eval,
coarse_db_len);
mp = gt_match_iterator_blast_process_new(call, err);
if (!mp)
had_err = -1;
gt_blast_process_call_delete(call);
if (!had_err) {
GtUword numofhits = 0;
while (!had_err &&
(status = gt_match_iterator_next(mp, &match, err)) !=
GT_MATCHER_STATUS_END) {
if (status == GT_MATCHER_STATUS_OK) {
GtMatchBlast *matchb = (GtMatchBlast*) match;
char *dbseqid = gt_malloc(sizeof (*dbseqid) * 50);
GtRange range_seq1;
GtRange range_seq2;
numofhits++;
gt_match_get_range_seq1(match, &range_seq1);
gt_match_get_range_seq2(match, &range_seq2);
gt_file_xprintf(
arguments->outfp,
"%s\t%s\t%.2f\t" GT_WU "\t" GT_WU "\t" GT_WU "\t" GT_WU "\t"
GT_WU "\t%g\t%.3f\n",
gt_match_get_seqid1(match),
gt_match_get_seqid2(match),
gt_match_blast_get_similarity(matchb),
gt_match_blast_get_align_length(matchb),
range_seq1.start,
range_seq1.end,
range_seq2.start,
range_seq2.end,
gt_match_blast_get_evalue(matchb),
(double) gt_match_blast_get_bitscore(matchb));
gt_match_delete(match);
gt_free(dbseqid);
} else if (status == GT_MATCHER_STATUS_ERROR) {
had_err = -1;
}
}
gt_log_log(GT_WU " hits found\n", numofhits);
}
gt_match_iterator_delete(mp);
}
if (!had_err)
if (timer != NULL)
gt_timer_show_progress_final(timer, stderr);
gt_timer_delete(timer);
/*cleanup*/
for (i=0; i < max_hits; i++) {
gt_free(hits[i].range);
}
gt_free(hits);
gt_str_delete(fastaname);
}
gt_str_delete(coarse_fname);
gt_logger_delete(logger);
return had_err;
}
static int gt_condenseq_compress_runner(GT_UNUSED int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
GtCondenseqCompressArguments *arguments = tool_arguments;
GtLogger *logger,
*kdb_logger;
FILE *kmer_fp = NULL;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX, stderr);
kdb_logger = gt_logger_new(arguments->kdb, GT_LOGGER_DEFLT_PREFIX, stderr);
if (arguments->kdb) {
kmer_fp = gt_fa_fopen("kmer_db.out", "w", err);
gt_logger_set_target(kdb_logger, kmer_fp);
}
if (gt_str_length(arguments->indexname) == 0UL) {
char *basenameptr;
basenameptr = gt_basename(argv[parsed_args]);
gt_str_set(arguments->indexname, basenameptr);
gt_free(basenameptr);
}
if (!had_err) {
GtEncseqLoader *es_l = gt_encseq_loader_new();
arguments->input_es = gt_encseq_loader_load(es_l, argv[parsed_args], err);
if (arguments->input_es == NULL)
had_err = -1;
gt_encseq_loader_delete(es_l);
}
if (!had_err) {
if (arguments->minalignlength == GT_UNDEF_UWORD)
arguments->minalignlength = arguments->initsize != GT_UNDEF_UWORD ?
arguments->initsize / (GtUword) 3UL :
GT_UNDEF_UWORD;
if (arguments->windowsize == GT_UNDEF_UINT)
arguments->windowsize = arguments->minalignlength != GT_UNDEF_UWORD ?
(unsigned int) (arguments->minalignlength / 5U) :
GT_UNDEF_UINT;
if (arguments->windowsize < 4U)
arguments->windowsize = 4U;
if (arguments->kmersize == GT_UNDEF_UINT) {
unsigned int size =
gt_alphabet_num_of_chars(gt_encseq_alphabet(arguments->input_es));
/* size^k ~= 100000 */
gt_safe_assign(arguments->kmersize,
gt_round_to_long(gt_log_base(100000.0, (double) size)));
gt_logger_log(logger, "|A|: %u, k: %u",
size, arguments->kmersize);
}
if (arguments->windowsize == GT_UNDEF_UINT) {
arguments->windowsize = 5U * arguments->kmersize;
}
if (arguments->minalignlength == GT_UNDEF_UWORD) {
arguments->minalignlength = (GtUword) (3UL * arguments->windowsize);
}
if (arguments->initsize == GT_UNDEF_UWORD) {
arguments->initsize = (GtUword) (3UL * arguments->minalignlength);
}
}
if (!had_err &&
arguments->windowsize <= arguments->kmersize) {
gt_error_set(err, "-windowsize (%u) must be larger -kmersize (%u)!",
arguments->windowsize, arguments->kmersize);
had_err = -1;
}
if (!had_err &&
arguments->minalignlength < (GtUword) arguments->windowsize) {
gt_error_set(err, "-alignlength (" GT_WU ") must be at least "
"-windowsize (%u)!", arguments->minalignlength,
arguments->windowsize);
had_err = -1;
}
if (!had_err && (arguments->initsize < arguments->minalignlength)) {
gt_error_set(err, "-initsize (" GT_WU ") must be at least "
"-alignlength (" GT_WU ")!", arguments->initsize,
arguments->minalignlength);
had_err = -1;
}
if (!had_err) {
GtCondenseqCreator *ces_c;
if (!had_err) {
ces_c = gt_condenseq_creator_new(arguments->initsize,
arguments->minalignlength,
arguments->xdrop,
&(arguments->scores),
arguments->kmersize,
arguments->windowsize,
logger,
err);
if (ces_c == NULL)
had_err = -1;
}
if (!had_err) {
if (arguments->cutoff_value == GT_UNDEF_UWORD)
gt_condenseq_creator_use_mean_cutoff(ces_c);
else if (arguments->cutoff_value == 0)
gt_condenseq_creator_disable_cutoff(ces_c);
else
gt_condenseq_creator_set_cutoff(ces_c, arguments->cutoff_value);
gt_condenseq_creator_set_mean_fraction(ces_c, arguments->fraction);
if (arguments->prune)
gt_condenseq_creator_disable_prune(ces_c);
if (arguments->brute)
gt_condenseq_creator_enable_brute_force(ces_c);
if (!arguments->diags)
gt_condenseq_creator_disable_diagonals(ces_c);
if (arguments->full_diags)
gt_condenseq_creator_enable_full_diagonals(ces_c);
if (arguments->clean_percent != GT_UNDEF_UINT)
gt_condenseq_creator_set_diags_clean_limit(ces_c,
arguments->clean_percent);
had_err = gt_condenseq_creator_create(ces_c,
arguments->indexname,
arguments->input_es,
logger, kdb_logger, err);
gt_condenseq_creator_delete(ces_c);
}
}
gt_logger_delete(logger);
gt_logger_delete(kdb_logger);
if (arguments->kdb)
gt_fa_fclose(kmer_fp);
return had_err;
}
static int gt_condenseq_extract_runner(GT_UNUSED int argc,
const char **argv,
int parsed_args,
void *tool_arguments,
GtError *err)
{
int had_err = 0;
GtCondenserExtractArguments *arguments = tool_arguments;
GtCondenseq *condenseq = NULL;
GtLogger *logger = NULL;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX, stderr);
if (!had_err) {
condenseq = gt_condenseq_new_from_file(argv[parsed_args], logger, err);
if (condenseq == NULL) {
had_err = -1;
}
}
if (!had_err) {
const char *buffer = NULL;
const char *desc = NULL;
GtUword desclen,
seqlen,
rend = gt_condenseq_total_length(condenseq),
send = gt_condenseq_num_of_sequences(condenseq);
bool concat = strcmp(gt_str_get(arguments->mode), "concat") == 0;
/* single sequence to extract = range of length 1 */
if (arguments->seq != GT_UNDEF_UWORD) {
arguments->seqrange.start = arguments->seqrange.end = arguments->seq;
}
/* no range given at all: extract all seqs */
if (arguments->range.start == GT_UNDEF_UWORD &&
arguments->seqrange.start == GT_UNDEF_UWORD) {
arguments->seqrange.start = 0;
arguments->seqrange.end = send - 1;
}
/* if seqs are specified, and concat is given, switch to posrange */
if (concat && arguments->seqrange.start != GT_UNDEF_UWORD) {
if (arguments->seqrange.end >= send) {
had_err = -1;
gt_error_set(err, "range end " GT_WU " excedes number of sequences "
GT_WU " (ranges are zero based sequence ids)",
arguments->seqrange.end, send);
}
else {
arguments->range.start =
gt_condenseq_seqstartpos(condenseq, arguments->seqrange.start);
arguments->range.end =
gt_condenseq_seqstartpos(condenseq, arguments->seqrange.end) +
gt_condenseq_seqlength(condenseq, arguments->seqrange.end) - 1;
}
}
/* extract sequence region */
if (!had_err && arguments->range.start != GT_UNDEF_UWORD) {
const GtUword maxbuffsize = ((GtUword) 1) << 17; /* ~ 100000byte */
GtUword clen,
rstart,
current_length = 0, i;
const char sepchar = gt_str_get(arguments->sepchar)[0];
if (arguments->range.end >= rend) {
had_err = -1;
gt_error_set(err, "range end " GT_WU " excedes length of sequence "
GT_WU " (ranges are zero based positions)",
arguments->range.end, rend);
}
if (!had_err) {
rstart = arguments->range.start;
rend = arguments->range.end;
/* nextlength = gt_condenseq_seqlength(condenseq, seqnum); */
/* seqstart = gt_condenseq_seqstartpos(condenseq, seqnum); */
/* gt_assert(rstart >= seqstart); */
/* nextlength -= rstart - seqstart; [> handle first seq <] */
while (rstart <= rend) {
GtRange cur_range;
if (rend - rstart > maxbuffsize) {
GtUword seqnum = gt_condenseq_pos2seqnum(condenseq,
rstart + maxbuffsize),
closest_sep = gt_condenseq_seqstartpos(condenseq,
seqnum) - 1;
gt_assert(closest_sep > rstart);
clen = closest_sep - rstart + 1;
}
else
clen = rend - rstart + 1;
cur_range.start = rstart;
cur_range.end = rstart + clen - 1;
buffer = gt_condenseq_extract_decoded_range(condenseq, cur_range,
sepchar);
gt_assert(buffer != NULL);
for (i = 0; i < clen; i++, current_length++) {
if (arguments->width && current_length == arguments->width) {
gt_file_xfputc('\n', arguments->outfp);
current_length = 0;
}
gt_file_xfputc(buffer[i], arguments->outfp);
}
rstart += clen;
}
gt_file_xfputc('\n', arguments->outfp);
}
}
else if (!had_err) { /* extract seqwise and always fasta */
GtUword seqnum,
sstart = arguments->seqrange.start;
if (arguments->seqrange.end >= send) {
had_err = -1;
gt_error_set(err, "range end " GT_WU " excedes number of sequences "
GT_WU " (ranges are zero based sequence ids)",
arguments->seqrange.end, send);
}
send = arguments->seqrange.end;
for (seqnum = sstart;
!had_err && seqnum <= send;
++seqnum) {
buffer = gt_condenseq_extract_decoded(condenseq, &seqlen, seqnum);
desc = gt_condenseq_description(condenseq, &desclen, seqnum);
gt_fasta_show_entry_nt(desc, desclen,
buffer, seqlen,
arguments->width,
arguments->outfp);
}
}
}
gt_condenseq_delete(condenseq);
gt_logger_delete(logger);
return had_err;
}
static int gt_genomediff_runner(int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
bool mirrored = false;
int had_err = 0,
i;
GtEncseq *encseq = NULL;
GtGenomediffArguments *arguments = tool_arguments;
GtLogger *logger;
GtShuUnitFileInfo *unit_info = NULL;
GtTimer *timer = NULL;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose,
GT_LOGGER_DEFLT_PREFIX,
stdout);
gt_assert(logger);
for (i = parsed_args; i < argc; i++) {
gt_str_array_add_cstr(arguments->filenames, argv[i]);
}
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("start");
gt_timer_start(timer);
gt_assert(timer);
}
if (arguments->with_units) {
gt_logger_log(logger, "unitfile option set, filename is %s\n",
gt_str_get(arguments->unitfile));
}
if (timer != NULL)
gt_timer_show_progress(timer, "start shu search", stdout);
if (gt_str_array_size(arguments->filenames) > 1UL) {
GtEncseqEncoder *ee = gt_encseq_encoder_new();
gt_encseq_encoder_set_timer(ee, timer);
gt_encseq_encoder_set_logger(ee, logger);
/* kr only makes sense for dna, so we can check this already with ee */
gt_encseq_encoder_set_input_dna(ee);
had_err = gt_encseq_encoder_encode(ee, arguments->filenames,
gt_str_get(arguments->indexname), err);
gt_encseq_encoder_delete(ee);
}
else {
gt_str_append_str(arguments->indexname,
gt_str_array_get_str(arguments->filenames, 0));
if (arguments->with_esa || arguments->with_pck) {
GtStr *current_line = gt_str_new();
FILE *prj_fp;
const char *buffer;
char **elements = NULL;
prj_fp = gt_fa_fopen_with_suffix(gt_str_get(arguments->indexname),
GT_PROJECTFILESUFFIX,"rb",err);
if (prj_fp == NULL)
had_err = -1;
while (!had_err && gt_str_read_next_line(current_line, prj_fp) != EOF) {
buffer = gt_str_get(current_line);
if (elements != NULL) {
gt_free(elements[0]);
gt_free(elements[1]);
}
gt_free(elements);
elements = gt_cstr_split(buffer, '=');
gt_log_log("%s", elements[0]);
if (strcmp("mirrored", elements[0]) == 0) {
gt_log_log("%s", elements[1]);
if (strcmp("1", elements[1]) == 0) {
mirrored = true;
gt_log_log("sequences are treated as mirrored");
}
}
gt_str_reset(current_line);
}
gt_str_delete(current_line);
if (elements != NULL) {
gt_free(elements[0]);
gt_free(elements[1]);
}
gt_free(elements);
gt_fa_xfclose(prj_fp);
}
}
if (!had_err) {
GtEncseqLoader *el = gt_encseq_loader_new_from_options(arguments->loadopts,
err);
if (mirrored)
gt_encseq_loader_mirror(el);
encseq =
gt_encseq_loader_load(el, gt_str_get(arguments->indexname), err);
gt_encseq_loader_delete(el);
}
if (encseq == NULL)
had_err = -1;
if (!had_err) {
unit_info = gt_shu_unit_info_new(encseq);
if (arguments->with_units)
had_err = gt_shu_unit_file_info_read(arguments->unitfile, unit_info,
logger, err);
}
if (!had_err) {
uint64_t **shusums = NULL;
if (arguments->with_esa || arguments->with_pck) {
shusums = gt_genomediff_shulen_sum(arguments, unit_info,
logger, timer, err);
if (shusums == NULL)
had_err = -1;
}
else {
const bool doesa = true;
GenomediffInfo gd_info;
Suffixeratoroptions sopts;
sopts.beverbose = arguments->verbose;
sopts.indexname = arguments->indexname;
sopts.db = NULL;
sopts.encopts = NULL;
sopts.genomediff = true;
sopts.inputindex = arguments->indexname;
sopts.loadopts = arguments->loadopts;
sopts.showprogress = false;
sopts.idxopts = arguments->idxopts;
gt_assert(unit_info != NULL);
gt_array2dim_calloc(shusums, unit_info->num_of_genomes,
unit_info->num_of_genomes);
gd_info.shulensums = shusums;
gd_info.unit_info = unit_info;
had_err = gt_runsuffixerator(doesa, &sopts, &gd_info, logger, err);
}
if (!had_err && shusums != NULL) {
had_err = gt_genomediff_kr_calc(shusums, arguments, unit_info,
arguments->with_pck, logger, timer, err);
gt_array2dim_delete(shusums);
}
}
if (timer != NULL) {
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
gt_logger_delete(logger);
gt_encseq_delete(encseq);
gt_shu_unit_info_delete(unit_info);
return had_err;
}
// Main method
int main(int argc, char * const *argv)
{
GtError *error;
GtLogger *logger;
GtQueue *streams;
GtNodeStream *stream, *last_stream;
CanonGFF3Options options = { NULL, NULL, false };
gt_lib_init();
error = gt_error_new();
canon_gff3_parse_options(argc, argv + 0, &options, error);
streams = gt_queue_new();
logger = gt_logger_new(true, "", stderr);
stream = gt_gff3_in_stream_new_unsorted(argc - optind, (const char **)
argv+optind);
gt_gff3_in_stream_check_id_attributes((GtGFF3InStream *)stream);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream *)stream);
gt_queue_add(streams, stream);
last_stream = stream;
if(options.infer)
{
GtHashmap *type_parents = gt_hashmap_new(GT_HASH_STRING, gt_free_func,
gt_free_func);
gt_hashmap_add(type_parents, gt_cstr_dup("mRNA"), gt_cstr_dup("gene"));
gt_hashmap_add(type_parents, gt_cstr_dup("tRNA"), gt_cstr_dup("gene"));
stream = agn_infer_parent_stream_new(last_stream,
type_parents);
gt_hashmap_delete(type_parents);
gt_queue_add(streams, stream);
last_stream = stream;
}
stream = agn_gene_stream_new(last_stream, logger);
gt_queue_add(streams, stream);
last_stream = stream;
if(options.source != NULL)
{
GtNodeVisitor *ssv = gt_set_source_visitor_new(options.source);
stream = gt_visitor_stream_new(last_stream, ssv);
gt_queue_add(streams, stream);
last_stream = stream;
}
stream = gt_gff3_out_stream_new(last_stream, options.outstream);
if(!options.infer)
gt_gff3_out_stream_retain_id_attributes((GtGFF3OutStream *)stream);
gt_queue_add(streams, stream);
last_stream = stream;
if(gt_node_stream_pull(last_stream, error) == -1)
{
fprintf(stderr, "[CanonGFF3] error processing node stream: %s",
gt_error_get(error));
}
while(gt_queue_size(streams) > 0)
{
stream = gt_queue_get(streams);
gt_node_stream_delete(stream);
}
gt_queue_delete(streams);
if(options.source != NULL)
gt_str_delete(options.source);
if(options.outstream != NULL)
gt_file_delete(options.outstream);
gt_error_delete(error);
gt_logger_delete(logger);
gt_lib_clean();
return 0;
}
static int gt_encseq2spm_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtEncseq2spmArguments *arguments = tool_arguments;
GtEncseqLoader *el = NULL;
GtEncseq *encseq = NULL;
bool haserr = false;
gt_error_check(err);
gt_assert(arguments);
el = gt_encseq_loader_new();
gt_encseq_loader_drop_description_support(el);
gt_encseq_loader_disable_autosupport(el);
encseq = gt_encseq_loader_load(el, gt_str_get(arguments->encseqinput),
err);
if (encseq == NULL)
{
haserr = true;
}
if (!haserr)
{
if (arguments->singlestrand)
{
gt_error_set(err,"option -singlestand is not implemented");
haserr = true;
} else
{
if (gt_encseq_mirror(encseq, err) != 0)
{
haserr = true;
}
}
}
if (!haserr && arguments->singlescan > 0)
{
GtTimer *timer = NULL;
if (gt_showtime_enabled())
{
char *outmsg;
switch (arguments->singlescan)
{
case 1:
outmsg = "to run fast scanning";
break;
case 2:
outmsg = "to run fast scanning with check";
break;
case 3:
outmsg = "to run fast scanning with output";
break;
case 4:
outmsg = "to run old scanning code";
break;
default:
gt_error_set(err,"argument %u to option -singlescan not allowed",
arguments->singlescan);
haserr = true;
}
if (!haserr)
{
timer = gt_timer_new_with_progress_description(outmsg);
gt_timer_start(timer);
}
}
if (!haserr)
{
unsigned int kmersize = 0;
haserr = gt_encseq2spm_kmersize(arguments, &kmersize, err);
if (!haserr)
{
if (arguments->singlescan == 4U)
{
gt_rungetencseqkmers(encseq,kmersize);
} else
{
if (arguments->singlescan > 0)
{
gt_firstcode_runkmerscan(encseq,arguments->singlescan - 1,kmersize,
arguments->minmatchlength);
}
}
}
}
if (timer != NULL)
{
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
}
if (!haserr && arguments->singlescan == 0)
{
GtLogger *logger;
const GtReadmode readmode = GT_READMODE_FORWARD;
GtBUstate_spmsk **spmsk_states = NULL;
unsigned int kmersize, threadcount;
#ifdef GT_THREADS_ENABLED
const unsigned int threads = gt_jobs;
#else
const unsigned int threads = 1U;
#endif
if (arguments->countspms || arguments->outputspms)
{
spmsk_states = gt_malloc(sizeof (*spmsk_states) * threads);
for (threadcount = 0; threadcount < threads; threadcount++)
{
spmsk_states[threadcount]
= gt_spmsk_inl_new(encseq,
readmode,
(unsigned long) arguments->minmatchlength,
arguments->countspms,
arguments->outputspms,
gt_str_get(arguments->encseqinput));
}
}
logger = gt_logger_new(arguments->verbose,GT_LOGGER_DEFLT_PREFIX, stdout);
haserr = gt_encseq2spm_kmersize(arguments, &kmersize, err);
if (!haserr)
{
if (storefirstcodes_getencseqkmers_twobitencoding(encseq,
kmersize,
arguments->numofparts,
arguments->maximumspace,
arguments->minmatchlength,
/* use false */ arguments->checksuftab,
/* use false */ arguments->onlyaccum,
/* use false */ arguments->
onlyallfirstcodes,
/* use 5U */ arguments->
addbscache_depth,
/* specify the extra space needed for
the function processing the interval */
arguments->phase2extra,
/* use true */ arguments->radixlarge ?
false : true,
/* use 2 without threads and
use 1 with threads */
arguments->radixparts,
spmsk_states != NULL
? gt_spmsk_inl_process
: NULL,
gt_spmsk_inl_process_end,
spmsk_states,
logger,
err) != 0)
{
haserr = true;
}
}
if (spmsk_states != NULL)
{
unsigned long countmatches = 0;
for (threadcount = 0; threadcount < threads; threadcount++)
{
countmatches += gt_spmsk_inl_delete(spmsk_states[threadcount]);
}
if (arguments->countspms)
{
printf("number of suffix-prefix matches=%lu\n",countmatches);
}
gt_free(spmsk_states);
}
gt_logger_delete(logger);
}
gt_encseq_delete(encseq);
gt_encseq_loader_delete(el);
return haserr ? -1 : 0;
}
static int gt_kmer_database_runner(GT_UNUSED int argc, const char **argv,
int parsed_args, void *tool_arguments,
GtError *err)
{
GtKmerDatabaseArguments *arguments = tool_arguments;
int had_err = 0;
GtEncseq *es;
GtUword es_length,
nu_kmer_codes = 0;
GtKmerDatabase *compare_db = NULL,
*db = NULL;
GtLogger *logger;
FILE *fp = NULL;
GtHashmap *kmer_hash = NULL;
GtTimer *timer = NULL;
gt_error_check(err);
gt_assert(arguments);
if (arguments->use_hash)
kmer_hash = gt_hashmap_new(GT_HASH_DIRECT, NULL,
(GtFree) gt_kmer_database_delete_hash_value);
if (arguments->bench)
timer = gt_timer_new_with_progress_description("loading encoded sequence");
logger = gt_logger_new(arguments->verbose, GT_LOGGER_DEFLT_PREFIX, stderr);
if (arguments->verbose && gt_str_length(arguments->print_filename) > 0UL) {
fp = gt_fa_fopen(gt_str_get(arguments->print_filename), "w", err);
gt_logger_set_target(logger, fp);
}
if (!had_err) {
GtEncseqLoader *es_l;
if (arguments->bench)
gt_timer_start(timer);
es_l = gt_encseq_loader_new();
es = gt_encseq_loader_load(es_l, argv[parsed_args], err);
if (arguments->bench)
gt_timer_show_progress(timer, "saving kmers (+iterating over file)",
stdout);
if (es == NULL) {
had_err = -1;
}
gt_encseq_loader_delete(es_l);
}
if (!had_err) {
es_length = gt_encseq_total_length(es);
if (es_length < (GtUword) arguments->kmersize) {
gt_error_set(err, "Input is too short for used kmersize. File length: "
GT_WU " kmersize: %u", es_length, arguments->kmersize);
had_err = -1;
}
}
if (!had_err) {
GtAlphabet *alphabet;
alphabet = gt_encseq_alphabet(es);
if (arguments->bench)
nu_kmer_codes = gt_power_for_small_exponents(
gt_alphabet_num_of_chars(alphabet),
arguments->kmersize);
if (!arguments->merge_only && !arguments->use_hash && !arguments->bench) {
compare_db = gt_kmer_database_new(gt_alphabet_num_of_chars(alphabet),
arguments->kmersize, arguments->sb_size, es);
}
if (!arguments->use_hash) {
db = gt_kmer_database_new(gt_alphabet_num_of_chars(alphabet),
arguments->kmersize,
arguments->sb_size, es);
if (arguments->cutoff) {
if (arguments->mean_cutoff)
gt_kmer_database_use_mean_cutoff(db, (GtUword) 2,
arguments->cutoff_value);
else
gt_kmer_database_set_cutoff(db, arguments->cutoff_value);
if (!arguments->prune)
gt_kmer_database_set_prune(db);
}
}
}
if (!had_err) {
GtUword startpos = 0,
endpos;
GtKmercodeiterator *iter;
const GtKmercode *kmercode = NULL;
iter = gt_kmercodeiterator_encseq_new(es, GT_READMODE_FORWARD,
arguments->kmersize, 0);
while (!had_err && startpos < es_length - (arguments->kmersize - 1)) {
GtUword startpos_add_kmer = startpos;
if (arguments->merge_only) {
endpos = startpos + (arguments->kmersize - 1) +
(gt_rand_max((arguments->sb_size - 1) * 2));
if (endpos > es_length)
endpos = es_length;
}
else {
endpos = startpos + (arguments->kmersize - 1) +
(gt_rand_max(arguments->sb_size - 1));
}
gt_kmercodeiterator_reset(iter, GT_READMODE_FORWARD, startpos);
while ((kmercode = gt_kmercodeiterator_encseq_next(iter)) != NULL &&
startpos_add_kmer <= endpos - (arguments->kmersize - 1)) {
if (!arguments->merge_only && !arguments->use_hash &&
!kmercode->definedspecialposition && !arguments->bench) {
gt_kmer_database_add_kmer(compare_db, kmercode->code,
startpos_add_kmer);
}
if (arguments->use_hash && !kmercode->definedspecialposition) {
gt_kmer_database_add_to_hash(kmer_hash, kmercode->code,
startpos_add_kmer);
}
startpos_add_kmer++;
}
if (!arguments->use_hash) {
gt_kmer_database_add_interval(db, startpos, endpos);
gt_kmer_database_print_buffer(db, logger);
if (!arguments->bench)
had_err = gt_kmer_database_check_consistency(db, err);
}
startpos = endpos + 1;
}
if (!arguments->use_hash) {
gt_kmer_database_flush(db);
gt_kmer_database_print_buffer(db, logger);
if (!had_err && !arguments->bench)
had_err = gt_kmer_database_check_consistency(db, err);
if (!arguments->merge_only && !had_err && !arguments->bench)
had_err = gt_kmer_database_check_consistency(compare_db, err);
if (!arguments->merge_only && !arguments->bench)
gt_kmer_database_print(compare_db, logger, true);
if (!arguments->merge_only && !had_err && !arguments->bench)
had_err = gt_kmer_database_compare(compare_db, db, err);
gt_kmer_database_print(db, logger, true);
}
gt_kmercodeiterator_delete(iter);
}
if (arguments->bench) {
GtKmerStartpos pos;
GtArrayGtUword *pos_hash;
GtUword rand_access = (GtUword) 50000000,
rand_code,
i,
sum = 0;
gt_timer_show_progress(timer, "random access", stdout);
for (i = 0; i < rand_access; i++) {
rand_code = gt_rand_max(nu_kmer_codes - 1);
if (arguments->use_hash) {
pos_hash = gt_hashmap_get(kmer_hash, (const void *) rand_code);
if (pos_hash != NULL)
sum += pos_hash->spaceGtUword[pos_hash->nextfreeGtUword - 1];
}
else {
pos = gt_kmer_database_get_startpos(db, rand_code);
if (pos.no_positions > 0)
sum += pos.startpos[pos.no_positions - 1];
}
}
printf("sum: " GT_WU "\n", sum);
gt_timer_show_progress(timer, "", stdout);
gt_timer_stop(timer);
gt_timer_delete(timer);
}
if (arguments->use_hash)
gt_hashmap_delete(kmer_hash);
gt_encseq_delete(es);
if (!arguments->use_hash)
gt_kmer_database_delete(db);
if (!arguments->merge_only && !arguments->bench)
gt_kmer_database_delete(compare_db);
gt_logger_delete(logger);
gt_fa_fclose(fp);
return had_err;
}
static int gt_gdiffcalc_runner(int argc, const char **argv, int parsed_args,
void *tool_arguments, GT_UNUSED GtError *err)
{
GtGenomediffArguments *arguments = tool_arguments;
int had_err = 0, i;
GtUword lcounter = 0, zcounter = 0;
double **shusums = NULL;
GtEncseq *encseq = NULL;
GtLogger *logger;
GtShuUnitFileInfo *unit_info = NULL;
GtTimer *timer = NULL;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(arguments->verbose,
GT_LOGGER_DEFLT_PREFIX,
stdout);
gt_assert(logger);
for (i = parsed_args; i < argc; i++) {
gt_str_array_add_cstr(arguments->filenames, argv[i]);
}
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("load encseq");
gt_timer_start(timer);
gt_assert(timer);
}
if (arguments->with_units) {
gt_logger_log(logger, "unitfile option set, filename is %s\n",
gt_str_get(arguments->unitfile));
}
if (!had_err) {
GtEncseqLoader *el = gt_encseq_loader_new_from_options(arguments->loadopts,
err);
encseq =
gt_encseq_loader_load(el, gt_str_get(arguments->indexname), err);
gt_encseq_loader_delete(el);
}
if (encseq == NULL)
had_err = -1;
if (timer != NULL)
gt_timer_show_progress(timer, "load units", stdout);
if (!had_err) {
unit_info = gt_shu_unit_info_new(encseq);
if (arguments->with_units)
had_err = gt_shu_unit_file_info_read(arguments->unitfile, unit_info,
logger, err);
}
if (timer != NULL)
gt_timer_show_progress(timer, "read table", stdout);
if (!had_err) {
GtIO *table_file = NULL;
GtTokenizer *tokenizer = NULL;
GtStr *line = NULL;
gt_assert(unit_info != NULL);
gt_array2dim_calloc(shusums, unit_info->num_of_genomes,
unit_info->num_of_genomes);
table_file = gt_io_new(gt_str_array_get(arguments->filenames, 0), "r");
tokenizer = gt_tokenizer_new(table_file);
line = gt_tokenizer_get_token(tokenizer);
while (line != NULL && !had_err) {
char *cline = gt_str_get(line);
char *elem = strtok(cline, ";");
zcounter = 0;
while (elem != NULL && !had_err) {
if (*elem != '#') {
if (1 != sscanf(elem, "%lf",
&shusums[lcounter][zcounter])) {
had_err = 1;
gt_error_set(err, "couldn't scan");
break;
}
gt_logger_log(logger,"wert: %lf", shusums[lcounter][zcounter]);
zcounter++;
}
else {
gt_logger_log(logger, "name: %s", elem++);
}
elem = strtok(NULL, ";");
}
gt_tokenizer_next_token(tokenizer);
gt_str_delete(line);
line = gt_tokenizer_get_token(tokenizer);
lcounter++;
gt_logger_log(logger, "line "GT_WD"", lcounter);
}
}
if (!had_err) {
GtUword num_of_seq, file_idx, seq_idx, startpos;
GT_UNUSED GtUword oldpos = 0;
gt_assert(unit_info != NULL);
gt_assert(lcounter == zcounter);
gt_assert(lcounter == unit_info->num_of_genomes);
num_of_seq = gt_encseq_num_of_sequences(unit_info->encseq);
for (seq_idx = 0; seq_idx < num_of_seq; seq_idx++) {
startpos = gt_encseq_seqstartpos(unit_info->encseq, seq_idx);
file_idx = gt_encseq_filenum(unit_info->encseq, startpos);
gt_log_log("seq: "GT_WU" starts at: "GT_WU"\n"
"belonges to file: "GT_WU" which is part of genome: %s",
seq_idx, startpos, file_idx,
gt_str_array_get(unit_info->genome_names,
unit_info->map_files[file_idx]));
gt_assert(oldpos <= startpos);
oldpos = startpos;
}
}
if (!had_err && shusums != NULL) {
had_err = gt_genomediff_calculate_div_from_avg(shusums, arguments,
unit_info,
logger, timer, err);
gt_array2dim_delete(shusums);
}
if (timer != NULL) {
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
gt_logger_delete(logger);
gt_encseq_delete(encseq);
gt_shu_unit_info_delete(unit_info);
return had_err;
}
static void showoptions(const Suffixeratoroptions *so)
{
GtUword i;
Sfxstrategy sfxtrategy;
GtLogger *logger = gt_logger_new(true, GT_LOGGER_DEFLT_PREFIX, stdout);
if (gt_str_length(gt_encseq_options_smap_value(so->encopts)) > 0)
{
gt_logger_log_force(logger, "smap=\"%s\"",
gt_str_get(gt_encseq_options_smap_value(so->encopts)));
}
if (gt_encseq_options_dna_value(so->encopts))
{
gt_logger_log_force(logger, "dna=yes");
}
if (gt_encseq_options_protein_value(so->encopts))
{
gt_logger_log_force(logger, "protein=yes");
}
if (gt_encseq_options_plain_value(so->encopts))
{
gt_logger_log_force(logger, "plain=yes");
}
gt_logger_log_force(logger, "indexname=\"%s\"",
gt_str_get(so->indexname));
if (gt_index_options_prefixlength_value(so->idxopts)
== GT_PREFIXLENGTH_AUTOMATIC)
{
gt_logger_log_force(logger, "prefixlength=automatic");
} else
{
gt_logger_log_force(logger, "prefixlength=%u",
gt_index_options_prefixlength_value(so->idxopts));
}
sfxtrategy = gt_index_options_sfxstrategy_value(so->idxopts);
gt_logger_log_force(logger, "storespecialcodes=%s",
sfxtrategy.storespecialcodes ? "true" : "false");
for (i=0; i<gt_str_array_size(so->db); i++)
{
gt_logger_log_force(logger, "inputfile["GT_WU"]=%s", i,
gt_str_array_get(so->db, i));
}
if (gt_str_length(so->inputindex) > 0)
{
gt_logger_log_force(logger, "inputindex=%s",
gt_str_get(so->inputindex));
}
gt_assert(gt_str_length(so->indexname) > 0);
gt_logger_log_force(logger, "indexname=%s",
gt_str_get(so->indexname));
gt_logger_log_force(logger, "outtistab=%s,outsuftab=%s,outlcptab=%s,"
"outbwttab=%s,outbcktab=%s,outdestab=%s,"
"outsdstab=%s,outssptab=%s,outkystab=%s",
gt_encseq_options_tis_value(so->encopts) ? "true" : "false",
gt_index_options_outsuftab_value(so->idxopts) ? "true" : "false",
gt_index_options_outlcptab_value(so->idxopts) ? "true" : "false",
gt_index_options_outbwttab_value(so->idxopts) ? "true" : "false",
gt_index_options_outbcktab_value(so->idxopts) ? "true" : "false",
gt_encseq_options_des_value(so->encopts) ? "true" : "false",
gt_encseq_options_sds_value(so->encopts) ? "true" : "false",
gt_encseq_options_ssp_value(so->encopts) ? "true" : "false",
gt_index_options_outkystab_value(so->idxopts) ?
(gt_index_options_outkyssort_value(so->idxopts) ?
"true with sort" : "true") :
"false");
if (gt_index_options_maximumspace_value(so->idxopts) > 0)
{
gt_assert(gt_index_options_numofparts_value(so->idxopts) == 1U);
gt_logger_log_force(logger, "maximumspace=%.0f MB",
GT_MEGABYTES(gt_index_options_maximumspace_value(so->idxopts)));
} else
{
gt_logger_log_force(logger, "parts=%u",
gt_index_options_numofparts_value(so->idxopts));
}
gt_logger_log_force(logger, "maxinsertionsort="GT_WU"",
sfxtrategy.maxinsertionsort);
gt_logger_log_force(logger, "maxbltriesort="GT_WU"",
sfxtrategy.maxbltriesort);
gt_logger_log_force(logger, "maxcountingsort="GT_WU"",
sfxtrategy.maxcountingsort);
gt_logger_log_force(logger, "lcpdist=%s",
gt_index_options_lcpdist_value(so->idxopts)
? "true"
: "false");
gt_logger_delete(logger);
}
static int gt_matstat_runner(GT_UNUSED int argc, GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments, GtError *err)
{
Gfmsubcallinfo *arguments = tool_arguments;
Fmindex fmindex;
Suffixarray suffixarray;
void *packedindex = NULL;
GtLogger *logger = NULL;
bool haserr = false;
const GtAlphabet *alphabet = NULL;
#ifdef WITHBCKTAB
unsigned int prefixlength = 0;
#endif
GtUword totallength;
bool gt_mapfmindexfail = false;
gt_error_check(err);
gt_assert(arguments);
logger = gt_logger_new(false, GT_LOGGER_DEFLT_PREFIX, stdout);
if (arguments->indextype == Fmindextype)
{
if (gt_mapfmindex(&fmindex,gt_str_get(arguments->indexname),
logger, err) != 0)
{
haserr = true;
gt_mapfmindexfail = true;
} else
{
alphabet = fmindex.alphabet;
}
totallength = fmindex.bwtlength-1;
} else
{
unsigned int mappedbits;
if (arguments->indextype == Esaindextype)
{
mappedbits = SARR_ESQTAB | SARR_SUFTAB
#undef WITHBCKTAB
#ifdef WITHBCKTAB
| SARR_BCKTAB
#endif
;
} else
{
if (dotestsequence(arguments))
{
mappedbits = SARR_ESQTAB;
} else
{
mappedbits = 0;
}
}
if (gt_mapsuffixarray(&suffixarray,
mappedbits,
gt_str_get(arguments->indexname),
logger,
err) != 0)
{
haserr = true;
totallength = 0;
} else
{
alphabet = gt_encseq_alphabet(suffixarray.encseq);
#ifdef WITHBCKTAB
prefixlength = suffixarray.prefixlength;
#endif
totallength = gt_encseq_total_length(suffixarray.encseq);
}
if (!haserr)
{
if (arguments->indextype == Packedindextype)
{
packedindex =
gt_loadvoidBWTSeqForSA(gt_str_get(arguments->indexname),
false,
err);
if (packedindex == NULL)
{
haserr = true;
}
}
}
}
if (!haserr)
{
const void *theindex;
Greedygmatchforwardfunction gmatchforwardfunction;
if (arguments->indextype == Fmindextype)
{
theindex = (const void *) &fmindex;
if (arguments->doms)
{
gmatchforwardfunction = gt_skfmmstats;
} else
{
gmatchforwardfunction = gt_skfmuniqueforward;
}
} else
{
if (arguments->indextype == Esaindextype)
{
theindex = (const void *) &suffixarray;
if (arguments->doms)
{
gmatchforwardfunction = gt_suffixarraymstats;
} else
{
gmatchforwardfunction = gt_suffixarrayuniqueforward;
}
} else
{
gt_assert(arguments->indextype == Packedindextype);
theindex = (const void *) packedindex;
if (arguments->doms)
{
gmatchforwardfunction = gt_voidpackedindexmstatsforward;
} else
{
gmatchforwardfunction = gt_voidpackedindexuniqueforward;
}
}
}
if (!haserr)
{
#ifdef WITHBCKTAB
if (prefixlength > 0 &&
arguments->indextype == Esaindextype &&
runsubstringiteration(gmatchforwardfunction,
theindex,
totallength,
suffixarray.bcktab,
suffixarray.countspecialcodes,
alphabet,
prefixlength,
arguments->queryfilenames,
err) != 0)
{
haserr = true;
}
#endif
if (!haserr &&
gt_findsubquerygmatchforward(dotestsequence(arguments)
? suffixarray.encseq
: NULL,
theindex,
totallength,
gmatchforwardfunction,
alphabet,
arguments->queryfilenames,
arguments->minlength,
arguments->maxlength,
(arguments->showmode & SHOWSEQUENCE)
? true : false,
(arguments->showmode & SHOWQUERYPOS)
? true : false,
(arguments->showmode & SHOWSUBJECTPOS)
? true : false,
err) != 0)
{
haserr = true;
}
}
}
if (arguments->indextype == Fmindextype)
{
if (!gt_mapfmindexfail)
{
gt_freefmindex(&fmindex);
}
} else
{
if (arguments->indextype == Packedindextype && packedindex != NULL)
{
gt_deletevoidBWTSeq(packedindex);
}
gt_freesuffixarray(&suffixarray);
}
gt_logger_delete(logger);
return haserr ? -1 : 0;;
}
int gt_runidxlocali(const IdxlocaliOptions *idxlocalioptions,GtError *err)
{
Genericindex *genericindex = NULL;
bool haserr = false;
GtLogger *logger;
const GtEncseq *encseq = NULL;
logger = gt_logger_new(idxlocalioptions->verbose,
GT_LOGGER_DEFLT_PREFIX, stdout);
if (idxlocalioptions->doonline)
{
GtEncseqLoader *el;
el = gt_encseq_loader_new();
gt_encseq_loader_require_multiseq_support(el);
gt_encseq_loader_drop_description_support(el);
gt_encseq_loader_set_logger(el, logger);
encseq = gt_encseq_loader_load(el, gt_str_get(idxlocalioptions->indexname),
err);
gt_encseq_loader_delete(el);
if (encseq == NULL)
{
haserr = true;
}
} else
{
genericindex = genericindex_new(gt_str_get(idxlocalioptions->indexname),
idxlocalioptions->withesa,
idxlocalioptions->withesa ||
idxlocalioptions->docompare,
false,
true,
0,
logger,
err);
if (genericindex == NULL)
{
haserr = true;
} else
{
encseq = genericindex_getencseq(genericindex);
}
}
if (!haserr)
{
GtSeqIterator *seqit;
const GtUchar *query;
unsigned long querylen;
char *desc = NULL;
int retval;
Limdfsresources *limdfsresources = NULL;
const AbstractDfstransformer *dfst;
SWdpresource *swdpresource = NULL;
Showmatchinfo showmatchinfo;
ProcessIdxMatch processmatch;
GtAlphabet *a;
void *processmatchinfoonline, *processmatchinfooffline;
Storematchinfo storeonline, storeoffline;
a = gt_encseq_alphabet(encseq);
if (idxlocalioptions->docompare)
{
processmatch = storematch;
gt_initstorematch(&storeonline,encseq);
gt_initstorematch(&storeoffline,encseq);
processmatchinfoonline = &storeonline;
processmatchinfooffline = &storeoffline;
} else
{
processmatch = showmatch;
showmatchinfo.encseq = encseq;
showmatchinfo.characters = gt_alphabet_characters(a);
showmatchinfo.wildcardshow = gt_alphabet_wildcard_show(a);
showmatchinfo.showalignment = idxlocalioptions->showalignment;
processmatchinfoonline = processmatchinfooffline = &showmatchinfo;
}
if (idxlocalioptions->doonline || idxlocalioptions->docompare)
{
swdpresource = gt_newSWdpresource(idxlocalioptions->matchscore,
idxlocalioptions->mismatchscore,
idxlocalioptions->gapextend,
idxlocalioptions->threshold,
idxlocalioptions->showalignment,
processmatch,
processmatchinfoonline);
}
dfst = gt_locali_AbstractDfstransformer();
if (!idxlocalioptions->doonline || idxlocalioptions->docompare)
{
gt_assert(genericindex != NULL);
limdfsresources = gt_newLimdfsresources(genericindex,
true,
0,
0, /* maxpathlength */
true, /* keepexpandedonstack */
processmatch,
processmatchinfooffline,
NULL, /* processresult */
NULL, /* processresult info */
dfst);
}
seqit = gt_seq_iterator_sequence_buffer_new(idxlocalioptions->queryfiles,
err);
if (!seqit)
haserr = true;
if (!haserr)
{
gt_seq_iterator_set_symbolmap(seqit, gt_alphabet_symbolmap(a));
for (showmatchinfo.queryunit = 0; /* Nothing */;
showmatchinfo.queryunit++)
{
retval = gt_seq_iterator_next(seqit,
&query,
&querylen,
&desc,
err);
if (retval < 0)
{
haserr = true;
break;
}
if (retval == 0)
{
break;
}
printf("process sequence " Formatuint64_t " of length %lu\n",
PRINTuint64_tcast(showmatchinfo.queryunit),querylen);
if (idxlocalioptions->doonline || idxlocalioptions->docompare)
{
gt_multiapplysmithwaterman(swdpresource,encseq,query,querylen);
}
if (!idxlocalioptions->doonline || idxlocalioptions->docompare)
{
gt_indexbasedlocali(limdfsresources,
idxlocalioptions->matchscore,
idxlocalioptions->mismatchscore,
idxlocalioptions->gapstart,
idxlocalioptions->gapextend,
idxlocalioptions->threshold,
query,
querylen,
dfst);
}
if (idxlocalioptions->docompare)
{
gt_checkandresetstorematch(showmatchinfo.queryunit,
&storeonline,&storeoffline);
}
}
if (limdfsresources != NULL)
{
gt_freeLimdfsresources(&limdfsresources,dfst);
}
if (swdpresource != NULL)
{
gt_freeSWdpresource(swdpresource);
swdpresource = NULL;
}
gt_seq_iterator_delete(seqit);
}
if (idxlocalioptions->docompare)
{
gt_freestorematch(&storeonline);
gt_freestorematch(&storeoffline);
}
}
if (genericindex == NULL)
{
gt_encseq_delete((GtEncseq *) encseq);
encseq = NULL;
} else
{
genericindex_delete(genericindex);
}
gt_logger_delete(logger);
logger = NULL;
return haserr ? -1 : 0;
}
static int gt_repfind_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments, GtError *err)
{
bool haserr = false;
Maxpairsoptions *arguments = tool_arguments;
GtLogger *logger = NULL;
GtQuerymatch *querymatchspaceptr = gt_querymatch_new();
GtXdropmatchinfo xdropmatchinfo;
gt_error_check(err);
xdropmatchinfo.querymatchspaceptr = querymatchspaceptr;
xdropmatchinfo.useq = gt_seqabstract_new_empty();
xdropmatchinfo.vseq = gt_seqabstract_new_empty();
xdropmatchinfo.arbitscores.mat = 2;
xdropmatchinfo.arbitscores.mis = -2;
xdropmatchinfo.arbitscores.ins = -3;
xdropmatchinfo.arbitscores.del = -3;
xdropmatchinfo.frontresource = gt_frontresource_new(100UL);
xdropmatchinfo.res = gt_xdrop_resources_new(&xdropmatchinfo.arbitscores);
xdropmatchinfo.belowscore = 5L;
logger = gt_logger_new(arguments->beverbose, GT_LOGGER_DEFLT_PREFIX, stdout);
if (parsed_args < argc)
{
gt_error_set(err,"superfluous arguments: \"%s\"",argv[argc-1]);
haserr = true;
}
if (!haserr)
{
if (gt_str_array_size(arguments->queryfiles) == 0)
{
if (arguments->samples == 0)
{
if (arguments->forward)
{
GtProcessmaxpairs processmaxpairs;
void *processmaxpairsdata;
if (arguments->searchspm)
{
processmaxpairs = gt_simplesuffixprefixmatchoutput;
processmaxpairsdata = NULL;
} else
{
if (arguments->extendseed)
{
processmaxpairs = gt_simplexdropselfmatchoutput;
processmaxpairsdata = (void *) &xdropmatchinfo;
} else
{
processmaxpairs = gt_simpleexactselfmatchoutput;
processmaxpairsdata = (void *) querymatchspaceptr;
}
}
if (gt_callenummaxpairs(gt_str_get(arguments->indexname),
arguments->userdefinedleastlength,
arguments->scanfile,
processmaxpairs,
processmaxpairsdata,
logger,
err) != 0)
{
haserr = true;
}
}
if (!haserr && arguments->reverse)
{
if (gt_callenumselfmatches(gt_str_get(arguments->indexname),
GT_READMODE_REVERSE,
arguments->userdefinedleastlength,
/*arguments->extendseed
? gt_processxdropquerymatches
:*/ gt_querymatch_output,
/*arguments->extendseed
? (void *) &xdropmatchinfo
:*/ NULL,
logger,
err) != 0)
{
haserr = true;
}
}
} else
{
if (gt_testmaxpairs(gt_str_get(arguments->indexname),
arguments->samples,
arguments->userdefinedleastlength,
(GtUword)
(100 * arguments->userdefinedleastlength),
logger,
err) != 0)
{
haserr = true;
}
}
} else
{
if (gt_callenumquerymatches(gt_str_get(arguments->indexname),
arguments->queryfiles,
false,
true,
false,
arguments->userdefinedleastlength,
NULL,
arguments->extendseed
? gt_processxdropquerymatches
: gt_querymatch_output,
arguments->extendseed
? (void *) &xdropmatchinfo
: NULL,
logger,
err) != 0)
{
haserr = true;
}
}
}
gt_querymatch_delete(querymatchspaceptr);
gt_seqabstract_delete(xdropmatchinfo.useq);
gt_seqabstract_delete(xdropmatchinfo.vseq);
gt_xdrop_resources_delete(xdropmatchinfo.res);
gt_frontresource_delete(xdropmatchinfo.frontresource);
gt_logger_delete(logger);
return haserr ? -1 : 0;
}
static int gt_readjoiner_cnttest_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv, GT_UNUSED int parsed_args,
void *tool_arguments, GT_UNUSED GtError *err)
{
GtReadjoinerCnttestArguments *arguments = tool_arguments;
GtEncseqLoader *el = NULL;
GtEncseq *reads = NULL;
GtBitsequence *bits = NULL;
GtUword nofreads;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if (arguments->test == GT_READJOINER_CNTTEST_SHOWLIST)
{
GtStr *fn = NULL;
fn = gt_str_clone(arguments->readset);
gt_str_append_cstr(fn, GT_READJOINER_SUFFIX_CNTLIST);
had_err = gt_cntlist_parse(gt_str_get(fn), true, &bits, &nofreads, err);
gt_str_delete(fn);
}
else if (arguments->test == GT_READJOINER_CNTTEST_BRUTEFORCE ||
arguments->test == GT_READJOINER_CNTTEST_KMP)
{
el = gt_encseq_loader_new();
gt_encseq_loader_drop_description_support(el);
gt_encseq_loader_disable_autosupport(el);
if (!arguments->singlestrand)
gt_encseq_loader_mirror(el);
reads = gt_encseq_loader_load(el, gt_str_get(arguments->readset), err);
if (reads == NULL)
had_err = -1;
else
{
gt_rdj_pairwise_exact(GT_OVLFIND_CNT, reads, !arguments->singlestrand,
false, arguments->test == GT_READJOINER_CNTTEST_KMP, 1UL, true,
NULL, NULL, false, NULL, &bits, &nofreads);
}
gt_encseq_delete(reads);
gt_encseq_loader_delete(el);
}
else if (arguments->test == GT_READJOINER_CNTTEST_ESA)
{
Sequentialsuffixarrayreader *ssar = NULL;
GtUword readlength = 0, firstrevcompl = 0;
GtLogger *verbose_logger = gt_logger_new(arguments->verbose,
GT_LOGGER_DEFLT_PREFIX, stderr);
ssar = gt_newSequentialsuffixarrayreaderfromfile(gt_str_get(
arguments->readset), SARR_LCPTAB | SARR_SUFTAB | SARR_SSPTAB,
true, verbose_logger, err);
if (gt_error_is_set(err))
had_err = -1;
else
{
nofreads = gt_encseq_num_of_sequences(ssar->encseq);
if (!arguments->singlestrand)
{
nofreads = GT_DIV2(nofreads);
firstrevcompl = nofreads;
}
GT_INITBITTAB(bits, nofreads);
if (!arguments->singlestrand)
if (gt_encseq_accesstype_get(ssar->encseq) == GT_ACCESS_TYPE_EQUALLENGTH)
readlength = gt_encseq_seqlength(ssar->encseq, 0);
(void)gt_contfind_bottomup(ssar, false, bits, arguments->singlestrand ? 0
: firstrevcompl, readlength);
}
if (ssar != NULL)
gt_freeSequentialsuffixarrayreader(&ssar);
gt_logger_delete(verbose_logger);
}
else
{
gt_assert(false);
}
if (!had_err)
had_err = gt_cntlist_show(bits, nofreads, NULL, false, err);
gt_free(bits);
return had_err;
}
