static int gt_compreads_compress_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments, GtError *err)
{
GtCsrHcrEncodeArguments *arguments = tool_arguments;
int had_err = 0;
GtAlphabet *alpha = NULL;
GtHcrEncoder *hcre = NULL;
GtTimer *timer = NULL;
gt_error_check(err);
gt_assert(arguments);
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("start");
gt_timer_start(timer);
gt_assert(timer);
}
if (gt_str_length(arguments->smap) > 0) {
alpha = gt_alphabet_new_from_file_no_suffix(gt_str_get(arguments->smap),
err);
if (!alpha)
had_err = 1;
}
else {
alpha = gt_alphabet_new_dna();
if (!alpha)
had_err = 1;
}
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "encoding", stdout);
hcre = gt_hcr_encoder_new(arguments->files, alpha, arguments->descs,
arguments->qrng, timer, err);
if (!hcre)
had_err = 1;
else {
if (arguments->pagewise)
gt_hcr_encoder_set_sampling_page(hcre);
else if (arguments->regular)
gt_hcr_encoder_set_sampling_regular(hcre);
gt_hcr_encoder_set_sampling_rate(hcre, arguments->srate);
if (gt_hcr_encoder_encode(hcre, gt_str_get(arguments->name),
timer, err) != 0)
had_err = 1;
}
gt_hcr_encoder_delete(hcre);
}
gt_alphabet_delete(alpha);
if (timer != NULL) {
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
return had_err;
}
static int gt_compreads_decompress_benchmark(GtHcrDecoder *hcrd,
unsigned long amount,
GtTimer *timer,
GtError *err) {
char qual[BUFSIZ] = {0},
seq[BUFSIZ] = {0};
int had_err = 0;
unsigned long rand,
max_rand = gt_hcr_decoder_num_of_reads(hcrd) - 1,
count;
GtStr *timer_comment = gt_str_new_cstr("extracting ");
GtStr *desc = gt_str_new();
gt_str_append_ulong(timer_comment, amount);
gt_str_append_cstr(timer_comment, " reads of ");
gt_str_append_ulong(timer_comment, max_rand + 1);
gt_str_append_cstr(timer_comment, "!");
if (timer == NULL) {
timer = gt_timer_new_with_progress_description("extract random reads");
gt_timer_start(timer);
}
else {
gt_timer_show_progress(timer, "extract random reads", stdout);
}
gt_log_log("%s",gt_str_get(timer_comment));
for (count = 0; count < amount; count++) {
if (!had_err) {
rand = gt_rand_max(max_rand);
gt_log_log("get read: %lu", rand);
had_err = gt_hcr_decoder_decode(hcrd, rand, seq, qual, desc, err);
gt_log_log("%s",gt_str_get(desc));
gt_log_log("%s",seq);
gt_log_log("%s",qual);
}
}
gt_str_delete(timer_comment);
gt_str_delete(desc);
if (!gt_showtime_enabled())
gt_timer_delete(timer);
return had_err;
}
static int gt_seed_extend_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtSeedExtendArguments *arguments = tool_arguments;
GtEncseqLoader *encseq_loader = NULL;
GtEncseq *aencseq = NULL, *bencseq = NULL;
GtGreedyextendmatchinfo *grextinfo = NULL;
GtXdropmatchinfo *xdropinfo = NULL;
GtQuerymatchoutoptions *querymatchoutopt = NULL;
GtTimer *seedextendtimer = NULL;
GtExtendCharAccess cam = GT_EXTEND_CHAR_ACCESS_ANY;
GtUword errorpercentage = 0UL;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments != NULL);
gt_assert(arguments->se_minidentity >= GT_EXTEND_MIN_IDENTITY_PERCENTAGE &&
arguments->se_minidentity <= 100UL);
/* Calculate error percentage from minidentity */
errorpercentage = 100UL - arguments->se_minidentity;
/* Measure whole running time */
if (arguments->benchmark || arguments->verbose) {
gt_showtime_enable();
}
if (gt_showtime_enabled())
{
seedextendtimer = gt_timer_new();
gt_timer_start(seedextendtimer);
}
/* Load encseq A */
encseq_loader = gt_encseq_loader_new();
gt_encseq_loader_enable_autosupport(encseq_loader);
aencseq = gt_encseq_loader_load(encseq_loader,
gt_str_get(arguments->dbs_indexname),
err);
if (aencseq == NULL)
had_err = -1;
/* If there is a 2nd read set: Load encseq B */
if (!had_err) {
if (strcmp(gt_str_get(arguments->dbs_queryname), "") != 0) {
bencseq = gt_encseq_loader_load(encseq_loader,
gt_str_get(arguments->dbs_queryname),
err);
} else {
bencseq = gt_encseq_ref(aencseq);
}
if (bencseq == NULL) {
had_err = -1;
gt_encseq_delete(aencseq);
}
}
gt_encseq_loader_delete(encseq_loader);
/* set character access method */
if (!had_err && (gt_option_is_set(arguments->se_option_greedy) ||
gt_option_is_set(arguments->se_option_xdrop) ||
arguments->se_alignmentwidth > 0))
{
cam = gt_greedy_extend_char_access(gt_str_get
(arguments->se_char_access_mode),
err);
if ((int) cam == -1) {
had_err = -1;
gt_encseq_delete(aencseq);
gt_encseq_delete(bencseq);
}
}
/* Use bias dependent parameters, adapted from E. Myers' DALIGNER */
if (!had_err && arguments->bias_parameters) {
const GtAlphabet *alpha = gt_encseq_alphabet(aencseq);
const double bias_factor[10] = {.690, .690, .690, .690, .780,
.850, .900, .933, .966, 1.000};
if (gt_alphabet_is_dna(alpha)) {
GtUword at, cg;
at = gt_encseq_charcount(aencseq, gt_alphabet_encode(alpha, 'a'));
at += gt_encseq_charcount(aencseq, gt_alphabet_encode(alpha, 't'));
cg = gt_encseq_charcount(aencseq, gt_alphabet_encode(alpha, 'c'));
cg += gt_encseq_charcount(aencseq, gt_alphabet_encode(alpha, 'g'));
if (at + cg > 0) {
const double ratio = (double)MIN(at, cg) / (at + cg);
int bias_index = (int)MAX(0.0, (ratio + 0.025) * 20.0 - 1.0);
gt_assert(bias_index < 10);
arguments->se_maxalilendiff = 30;
arguments->se_perc_match_hist = (GtUword)(100.0 - errorpercentage *
bias_factor[bias_index]);
if (arguments->verbose) {
printf("# Base ratio = %4.2lf -> percmathistory = "GT_WU"\n",
ratio, arguments->se_perc_match_hist);
}
} else {
had_err = -1;
}
} else {
had_err = -1;
}
if (had_err) {
gt_error_set(err, "option \"-bias-parameters\" can only be applied to "
"the DNA alphabet");
gt_encseq_delete(aencseq);
gt_encseq_delete(bencseq);
}
}
/* Prepare options for greedy extension */
if (!had_err && gt_option_is_set(arguments->se_option_greedy)) {
grextinfo = gt_greedy_extend_matchinfo_new(errorpercentage,
arguments->se_maxalilendiff,
arguments->se_historysize,
arguments->se_perc_match_hist,
arguments->se_alignlength,
cam,
arguments->se_extendgreedy);
if (arguments->benchmark) {
gt_greedy_extend_matchinfo_silent_set(grextinfo);
}
}
/* Prepare options for xdrop extension */
if (!had_err && gt_option_is_set(arguments->se_option_xdrop)) {
xdropinfo = gt_xdrop_matchinfo_new(arguments->se_alignlength,
errorpercentage,
arguments->se_xdropbelowscore,
arguments->se_extendxdrop);
if (arguments->benchmark) {
gt_xdrop_matchinfo_silent_set(xdropinfo);
}
}
/* Prepare output options */
if (!had_err && (arguments->se_alignmentwidth > 0 ||
gt_option_is_set(arguments->se_option_xdrop)))
{
querymatchoutopt
= gt_querymatchoutoptions_new(arguments->se_alignmentwidth);
if (gt_option_is_set(arguments->se_option_xdrop) ||
gt_option_is_set(arguments->se_option_greedy))
{
const GtUword sensitivity = gt_option_is_set(arguments->se_option_greedy)
? arguments->se_extendgreedy : 100;
gt_querymatchoutoptions_extend(querymatchoutopt,
errorpercentage,
arguments->se_maxalilendiff,
arguments->se_historysize,
arguments->se_perc_match_hist,
cam,
sensitivity);
}
}
/* Start algorithm */
if (!had_err) {
GtDiagbandseed dbsarguments;
dbsarguments.errorpercentage = errorpercentage;
dbsarguments.userdefinedleastlength = arguments->se_alignlength;
dbsarguments.seedlength = arguments->dbs_seedlength;
dbsarguments.logdiagbandwidth = arguments->dbs_logdiagbandwidth;
dbsarguments.mincoverage = arguments->dbs_mincoverage;
dbsarguments.maxfreq = arguments->dbs_maxfreq;
dbsarguments.memlimit = arguments->dbs_memlimit;
dbsarguments.mirror = arguments->mirror;
dbsarguments.overlappingseeds = arguments->overlappingseeds;
dbsarguments.verify = arguments->dbs_verify;
dbsarguments.verbose = arguments->verbose;
dbsarguments.debug_kmer = arguments->dbs_debug_kmer;
dbsarguments.debug_seedpair = arguments->dbs_debug_seedpair;
dbsarguments.seed_display = arguments->seed_display;
dbsarguments.extendgreedyinfo = grextinfo;
dbsarguments.extendxdropinfo = xdropinfo;
dbsarguments.querymatchoutopt = querymatchoutopt;
had_err = gt_diagbandseed_run(aencseq, bencseq, &dbsarguments, err);
/* clean up */
gt_encseq_delete(aencseq);
gt_encseq_delete(bencseq);
if (gt_option_is_set(arguments->se_option_greedy)) {
gt_greedy_extend_matchinfo_delete(grextinfo);
}
if (gt_option_is_set(arguments->se_option_xdrop)) {
gt_xdrop_matchinfo_delete(xdropinfo);
}
if (arguments->se_alignmentwidth > 0 ||
gt_option_is_set(arguments->se_option_xdrop)) {
gt_querymatchoutoptions_delete(querymatchoutopt);
}
}
if (gt_showtime_enabled()) {
if (!had_err) {
char *keystring
= gt_seed_extend_params_keystring(gt_option_is_set(arguments->
se_option_greedy),
gt_option_is_set(arguments->
se_option_xdrop),
arguments->dbs_seedlength,
arguments->se_alignlength,
arguments->se_minidentity,
arguments->se_maxalilendiff,
arguments->se_perc_match_hist,
arguments->se_extendgreedy,
arguments->se_extendxdrop,
arguments->se_xdropbelowscore);
printf("# TIME seedextend-%s", keystring);
gt_free(keystring);
gt_timer_show_formatted(seedextendtimer,
" overall " GT_WD ".%06ld\n",
stdout);
}
gt_timer_delete(seedextendtimer);
}
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;
}
static int gt_speck_runner(int argc, const char **argv, int parsed_args,
void *tool_arguments, GtError *err)
{
GtNodeStream *gff3_in_stream = NULL, *checker_stream = NULL,
*a_in_stream = NULL, *a_out_stream = NULL,
*feature_stream = NULL, *sort_stream = NULL,
*last_stream = NULL;
GtNodeVisitor *spec_visitor = NULL;
GtSpecResults *res = NULL;
GtFeatureIndex *fi = NULL;
GtTimer *t = NULL;
GtRegionMapping *rm = NULL;
GtArray *arr = gt_array_new(sizeof (GtFeatureNode*));
SpeccheckArguments *arguments = tool_arguments;
int had_err = 0;
gt_error_check(err);
res = gt_spec_results_new();
gt_assert(res);
t = gt_timer_new();
gt_assert(t);
spec_visitor = gt_spec_visitor_new(gt_str_get(arguments->specfile), res,
err);
if (!spec_visitor)
return -1;
/* add region mapping if given */
if (gt_seqid2file_option_used(arguments->s2fi)) {
rm = gt_seqid2file_region_mapping_new(arguments->s2fi, err);
if (!rm)
had_err = -1;
if (!had_err)
gt_spec_visitor_add_region_mapping((GtSpecVisitor*) spec_visitor, rm);
}
/* set runtime error behaviour */
if (arguments->fail_hard)
gt_spec_visitor_fail_on_runtime_error((GtSpecVisitor*) spec_visitor);
else
gt_spec_visitor_report_runtime_errors((GtSpecVisitor*) spec_visitor);
/* redirect warnings */
gt_warning_set_handler(gt_speck_record_warning, res);
last_stream = gff3_in_stream = gt_gff3_in_stream_new_unsorted(
argc - parsed_args,
argv + parsed_args);
gt_assert(gff3_in_stream);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream*) gff3_in_stream);
/* insert sort stream if requested */
if (arguments->sort) {
last_stream = sort_stream = gt_sort_stream_new(last_stream);
}
/* if -provideindex is given, collect input features and index them first */
if (arguments->provideindex) {
fi = gt_feature_index_memory_new();
gt_assert(fi);
last_stream = feature_stream = gt_feature_stream_new(last_stream, fi);
gt_assert(feature_stream);
last_stream = a_out_stream = gt_array_out_stream_all_new(last_stream, arr,
err);
if (!a_out_stream)
had_err = -1;
gt_timer_start(t);
if (!had_err)
had_err = gt_node_stream_pull(last_stream, err);
if (!had_err) {
gt_spec_visitor_add_feature_index((GtSpecVisitor*) spec_visitor,
gt_feature_index_ref(fi));
last_stream = a_in_stream = gt_array_in_stream_new(arr, NULL, err);
if (!a_in_stream)
had_err = -1;
}
} else {
gt_timer_start(t);
}
if (!had_err) {
checker_stream = gt_visitor_stream_new(last_stream, spec_visitor);
gt_assert(checker_stream);
}
/* perform checking */
if (!had_err)
had_err = gt_node_stream_pull(checker_stream, err);
gt_timer_stop(t);
/* reset warnings output */
gt_warning_set_handler(gt_warning_default_handler, NULL);
/* output results */
if (!had_err)
gt_spec_results_report(res, arguments->outfp,
gt_str_get(arguments->specfile),
arguments->verbose, arguments->colored,
!arguments->allexpects);
if (!had_err)
gt_timer_show_formatted(t, "Finished in " GT_WD ".%06ld s.\n", stderr);
/* free */
gt_node_stream_delete(gff3_in_stream);
gt_node_stream_delete(a_in_stream);
gt_node_stream_delete(a_out_stream);
gt_node_stream_delete(checker_stream);
gt_node_stream_delete(feature_stream);
gt_node_stream_delete(sort_stream);
gt_spec_results_delete(res);
gt_feature_index_delete(fi);
gt_timer_delete(t);
gt_array_delete(arr);
return had_err;
}
static int gt_compreads_decompress_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments, GtError *err)
{
GtCsrHcrDecodeArguments *arguments = tool_arguments;
int had_err = 0;
GtAlphabet *alpha = NULL;
GtHcrDecoder *hcrd = NULL;
GtTimer *timer = NULL;
unsigned long start,
end;
gt_error_check(err);
gt_assert(arguments);
if (gt_showtime_enabled()) {
timer = gt_timer_new_with_progress_description("start");
gt_timer_start(timer);
gt_assert(timer);
}
if (gt_str_length(arguments->smap) > 0) {
alpha = gt_alphabet_new_from_file_no_suffix(gt_str_get(arguments->smap),
err);
if (!alpha)
had_err = -1;
}
else {
alpha = gt_alphabet_new_dna();
if (!alpha)
had_err = -1;
}
if (!had_err) {
if (timer != NULL)
gt_timer_show_progress(timer, "decoding", stdout);
if (gt_str_length(arguments->name) == 0) {
char *basenameptr;
basenameptr = gt_basename(gt_str_get(arguments->file));
gt_str_set(arguments->name, basenameptr);
gt_free(basenameptr);
}
hcrd = gt_hcr_decoder_new(gt_str_get(arguments->file), alpha,
arguments->descs, timer, err);
if (hcrd == NULL)
had_err = -1;
else {
if (arguments->bench != 0) {
had_err = gt_compreads_decompress_benchmark(hcrd,
arguments->bench,
timer, err);
}
else {
if (arguments->rng.start != GT_UNDEF_ULONG
&& arguments->rng.end != GT_UNDEF_ULONG) {
if (arguments->rng.start >= gt_hcr_decoder_num_of_reads(hcrd)
|| arguments->rng.end >= gt_hcr_decoder_num_of_reads(hcrd)) {
gt_error_set(err, "range %lu-%lu includes a read number exceeding "
"the total number of reads (%lu)",
arguments->rng.start,
arguments->rng.end,
gt_hcr_decoder_num_of_reads(hcrd));
had_err = -1;
}
start = arguments->rng.start;
end = arguments->rng.end;
}
else {
start = 0;
end = gt_hcr_decoder_num_of_reads(hcrd) - 1;
}
if (!had_err) {
gt_log_log("filebasename: %s", gt_str_get(arguments->name));
if (gt_hcr_decoder_decode_range(hcrd, gt_str_get(arguments->name),
start, end, timer, err)
!= 0)
had_err = -1;
}
}
}
gt_hcr_decoder_delete(hcrd);
}
gt_alphabet_delete(alpha);
if (timer != NULL) {
gt_timer_show_progress_final(timer, stdout);
gt_timer_delete(timer);
}
if (had_err)
gt_assert(gt_error_is_set(err));
return had_err;
}
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_linspace_align_runner(GT_UNUSED int argc,
GT_UNUSED const char **argv,
GT_UNUSED int parsed_args,
void *tool_arguments,
GtError *err)
{
GtLinspaceArguments *arguments = tool_arguments;
int had_err = 0;
GtAlignment *align;
GtWord left_dist = 0, right_dist = 0;
GtSequenceTable *sequence_table1, *sequence_table2;
GtLinspaceManagement *spacemanager;
GtScoreHandler *scorehandler = NULL;
GtTimer *linspacetimer = NULL;
GtAlphabet *alphabet = NULL;
gt_error_check(err);
gt_assert(arguments);
sequence_table1 = gt_sequence_table_new();
sequence_table2 = gt_sequence_table_new();
align = gt_alignment_new();
spacemanager = gt_linspace_management_new();
gt_linspace_management_set_TSfactor(spacemanager,arguments->timesquarefactor);
/* get sequences */
if (gt_str_array_size(arguments->strings) > 0)
{
get_onesequence(sequence_table1, arguments->strings, 0);
sequence_table1->size++;
get_onesequence(sequence_table2, arguments->strings, 1);
sequence_table2->size++;
}
else if (gt_str_array_size(arguments->files) > 0)
{
had_err = get_fastasequences(sequence_table1,
gt_str_array_get_str(arguments->files,0),err);
if (!had_err)
{
had_err = get_fastasequences(sequence_table2,
gt_str_array_get_str(arguments->files,1),err);
}
}
if (arguments->dna)
{
alphabet = gt_alphabet_new_dna();
} else
{
gt_assert(arguments->protein);
alphabet = gt_alphabet_new_protein();
}
gt_encode_sequence_table(alphabet,sequence_table1);
gt_encode_sequence_table(alphabet,sequence_table2);
if (!had_err)
{
scorehandler = gt_arguments2scorehandler(arguments,err);
if (scorehandler == NULL)
{
had_err = -1;
} else
{
if (arguments->global && arguments->protein && !arguments->has_costmatrix)
{
GtScoreHandler *costhandler = gt_scorehandler2costhandler(scorehandler);
gt_scorehandler_delete(scorehandler);
scorehandler = costhandler;
}
}
}
/* get diagonal band */
if (!had_err && arguments->diagonal)
{
if (gt_str_array_size(arguments->diagonalbonds) > 0)
{
had_err = gt_parse_score_value(__LINE__,&left_dist,
gt_str_array_get(arguments->diagonalbonds,0),
false, err);
if (!had_err)
{
had_err = gt_parse_score_value(__LINE__,&right_dist,
gt_str_array_get(arguments->diagonalbonds,1),
false, err);
}
}
}
if (!had_err && arguments->spacetime)
{
linspacetimer = gt_timer_new();
}
/* alignment functions with linear gap costs */
if (!had_err)
{
bool affine;
if (gt_str_array_size(arguments->linearcosts) > 0)
{
affine = false;
} else
{
gt_assert(gt_str_array_size(arguments->affinecosts) > 0);
affine = true;
}
had_err = gt_all_against_all_alignment_check (
affine, align, arguments,
spacemanager,
scorehandler,
gt_alphabet_characters(alphabet),
gt_alphabet_wildcard_show(alphabet),
sequence_table1,
sequence_table2,
left_dist,
right_dist,
linspacetimer,err);
}
/*spacetime option*/
if (!had_err && arguments->spacetime)
{
printf("# combined space peak in kilobytes: %f\n",
GT_KILOBYTES(gt_linspace_management_get_spacepeak(spacemanager)));
gt_timer_show_formatted(linspacetimer,"# TIME overall " GT_WD ".%02ld\n",
stdout);
}
gt_timer_delete(linspacetimer);
gt_linspace_management_delete(spacemanager);
gt_sequence_table_delete(sequence_table1);
gt_sequence_table_delete(sequence_table2);
gt_alignment_delete(align);
gt_alphabet_delete(alphabet);
gt_scorehandler_delete(scorehandler);
return had_err;
}
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_speck_runner(int argc, const char **argv, int parsed_args,
void *tool_arguments, GtError *err)
{
GtNodeStream *gff3_in_stream = NULL, *checker_stream = NULL,
*a_in_stream = NULL, *a_out_stream = NULL,
*feature_stream = NULL, *sort_stream = NULL,
*last_stream = NULL;
GtNodeVisitor *spec_visitor = NULL;
GtSpecResults *res = NULL;
GtFeatureIndex *fi = NULL;
GtTypeChecker *type_checker = NULL;
GtTimer *t = NULL;
GtRegionMapping *rm = NULL;
GtArray *arr = gt_array_new(sizeof (GtFeatureNode*));
GtStr *prog, *speclib;
SpeccheckArguments *arguments = tool_arguments;
int had_err = 0;
gt_error_check(err);
res = gt_spec_results_new();
gt_assert(res);
if (gt_file_exists(gt_str_get(arguments->format))) {
speclib = gt_str_ref(arguments->format);
} else {
prog = gt_str_new();
gt_str_append_cstr_nt(prog, gt_error_get_progname(err),
gt_cstr_length_up_to_char(gt_error_get_progname(err), ' '));
speclib = gt_get_gtdata_path(gt_str_get(prog), NULL);
gt_str_delete(prog);
gt_str_append_cstr(speclib, "/spec/output_drivers/");
gt_str_append_str(speclib, arguments->format);
if (!gt_file_exists(gt_str_get(speclib))) {
gt_error_set(err, "output driver file \"%s\" does not exist",
gt_str_get(speclib));
had_err = -1;
}
}
if (!had_err) {
spec_visitor = gt_spec_visitor_new(gt_str_get(arguments->specfile), res,
err);
if (!spec_visitor) {
gt_spec_results_delete(res);
return -1;
}
}
t = gt_timer_new();
gt_assert(t);
/* add region mapping if given */
if (!had_err && gt_seqid2file_option_used(arguments->s2fi)) {
rm = gt_seqid2file_region_mapping_new(arguments->s2fi, err);
if (!rm)
had_err = -1;
if (!had_err)
gt_spec_visitor_add_region_mapping((GtSpecVisitor*) spec_visitor, rm);
}
/* set type checker if necessary */
if (!had_err && gt_typecheck_info_option_used(arguments->tci)) {
type_checker = gt_typecheck_info_create_type_checker(arguments->tci, err);
if (!type_checker)
had_err = -1;
if (!had_err)
gt_spec_visitor_add_type_checker((GtSpecVisitor*) spec_visitor,
type_checker);
}
if (!had_err) {
/* set runtime error behaviour */
if (arguments->fail_hard)
gt_spec_visitor_fail_on_runtime_error((GtSpecVisitor*) spec_visitor);
else
gt_spec_visitor_report_runtime_errors((GtSpecVisitor*) spec_visitor);
/* redirect warnings */
gt_warning_set_handler(gt_speck_record_warning, res);
last_stream = gff3_in_stream = gt_gff3_in_stream_new_unsorted(
argc - parsed_args,
argv + parsed_args);
gt_assert(gff3_in_stream);
gt_gff3_in_stream_enable_tidy_mode((GtGFF3InStream*) gff3_in_stream);
/* insert sort stream if requested */
if (arguments->sort) {
last_stream = sort_stream = gt_sort_stream_new(last_stream);
}
/* if -provideindex is given, collect input features and index them first */
if (arguments->provideindex) {
fi = gt_feature_index_memory_new();
gt_assert(fi);
last_stream = feature_stream = gt_feature_stream_new(last_stream, fi);
gt_assert(feature_stream);
last_stream = a_out_stream = gt_array_out_stream_all_new(last_stream, arr,
err);
if (!a_out_stream)
had_err = -1;
gt_timer_start(t);
if (!had_err)
had_err = gt_node_stream_pull(last_stream, err);
if (!had_err) {
gt_spec_visitor_add_feature_index((GtSpecVisitor*) spec_visitor,
gt_feature_index_ref(fi));
last_stream = a_in_stream = gt_array_in_stream_new(arr, NULL, err);
if (!a_in_stream)
had_err = -1;
}
} else {
gt_timer_start(t);
}
if (!had_err) {
checker_stream = gt_visitor_stream_new(last_stream, spec_visitor);
gt_assert(checker_stream);
}
/* perform checking */
if (!had_err)
had_err = gt_node_stream_pull(checker_stream, err);
gt_timer_stop(t);
/* reset warnings output */
gt_warning_set_handler(gt_warning_default_handler, NULL);
/* output results */
if (!had_err) {
GtStr *runtime = gt_str_new();
gt_timer_get_formatted(t, GT_WD ".%06ld", runtime);
had_err = gt_spec_results_render_template(res, gt_str_get(speclib),
arguments->outfp,
gt_str_get(arguments->specfile),
arguments->verbose,
arguments->colored,
gt_str_get(runtime), err);
gt_str_delete(runtime);
}
}
/* free */
gt_node_stream_delete(gff3_in_stream);
gt_node_stream_delete(a_in_stream);
gt_node_stream_delete(a_out_stream);
gt_node_stream_delete(checker_stream);
gt_node_stream_delete(feature_stream);
gt_node_stream_delete(sort_stream);
gt_spec_results_delete(res);
gt_feature_index_delete(fi);
gt_type_checker_delete(type_checker);
gt_timer_delete(t);
gt_array_delete(arr);
gt_str_delete(speclib);
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 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;
}
