static int gt_inlineseq_split_runner(int argc, const char **argv,
int parsed_args,
void *tool_arguments, GtError *err)
{
GtInlineseqSplitArguments *arguments = tool_arguments;
GtNodeStream *gff3_in_stream = NULL,
*gff3_out_stream = NULL,
*split_stream = NULL,
*last_stream = NULL;
GtFile *seq_out_file = NULL,
*gff3_out_file = NULL;
int had_err = 0;
gt_error_check(err);
gt_assert(arguments);
if (gt_str_length(arguments->seqoutfile) > 0) {
seq_out_file = gt_file_new(gt_str_get(arguments->seqoutfile), "w+", err);
if (!seq_out_file)
had_err = -1;
}
if (!had_err && gt_str_length(arguments->gffoutfile) > 0) {
gff3_out_file = gt_file_new(gt_str_get(arguments->gffoutfile), "w+", err);
if (!gff3_out_file)
had_err = -1;
}
if (!had_err) {
last_stream = gff3_in_stream = gt_gff3_in_stream_new_unsorted(
argc - parsed_args,
argv + parsed_args);
gt_assert(gff3_in_stream);
}
if (!had_err) {
last_stream = split_stream = gt_sequence_node_out_stream_new(last_stream,
seq_out_file,
err);
gt_assert(split_stream);
}
if (!had_err) {
last_stream = gff3_out_stream = gt_gff3_out_stream_new(last_stream,
gff3_out_file);
had_err = gt_node_stream_pull(last_stream, err);
}
gt_node_stream_delete(gff3_in_stream);
gt_node_stream_delete(gff3_out_stream);
gt_node_stream_delete(split_stream);
gt_file_delete(seq_out_file);
gt_file_delete(gff3_out_file);
return had_err;
}
static int create_manpage(const char *outdir, const char *toolname,
GtOptionParser *option_parser, GtError *err)
{
GtFile *outfile = NULL;
GtStr *man, *pathbuf;
char *utoolname;
int had_err = 0;
gt_error_check(err);
gt_assert(outdir && toolname && option_parser);
man = gt_str_new();
pathbuf = gt_str_new_cstr(outdir);
utoolname = gt_cstr_dup(toolname);
gt_cstr_rep(utoolname, ' ', '_');
if (!gt_file_exists(gt_str_get(pathbuf)))
gt_xmkdir(gt_str_get(pathbuf));
gt_str_append_char(pathbuf, GT_PATH_SEPARATOR);
gt_str_append_cstr(pathbuf, utoolname);
gt_str_append_cstr(pathbuf, ".mansrc");
gt_free(utoolname);
if (!(outfile = gt_file_new(gt_str_get(pathbuf), "w+", err)))
had_err = -1;
if (!had_err)
had_err = gt_option_parser_manpage(option_parser, toolname, man, err);
if (!had_err)
gt_file_xprintf(outfile, "%s", gt_str_get(man));
gt_file_delete(outfile);
gt_str_delete(pathbuf);
gt_str_delete(man);
return had_err;
}
static void canon_gff3_parse_options(int argc, char * const *argv,
CanonGFF3Options *options, GtError *error)
{
int opt = 0;
int optindex = 0;
const char *optstr = "hio:s:v";
const struct option init_options[] =
{
{ "help", no_argument, NULL, 'h' },
{ "infer", no_argument, NULL, 'i' },
{ "outfile", required_argument, NULL, 'o' },
{ "source", required_argument, NULL, 's' },
{ "version", no_argument, NULL, 'v' },
{ NULL, no_argument, NULL, 0 },
};
for(opt = getopt_long(argc, argv, optstr, init_options, &optindex);
opt != -1;
opt = getopt_long(argc, argv, optstr, init_options, &optindex))
{
if(opt == 'h')
{
print_usage(stdout);
exit(0);
}
else if(opt == 'i')
options->infer = true;
else if(opt == 'o')
{
if(options->outstream != NULL)
gt_file_delete(options->outstream);
options->outstream = gt_file_new(optarg, "w", error);
}
else if(opt == 's')
{
if(options->source != NULL)
gt_str_delete(options->source);
options->source = gt_str_new_cstr(optarg);
}
else if(opt == 'v')
{
agn_print_version("CanonGFF3", stdout);
exit(0);
}
}
}
static int gtf_in_stream_process_file(GtGTFInStream *gtf_in_stream,
GtError *err)
{
GtGTFParser *gtf_parser;
GtStr *filenamestr;
GtFile *fpin;
int had_err = 0;
gt_error_check(err);
gt_assert(gtf_in_stream);
gtf_parser = gt_gtf_parser_new(gtf_in_stream->type_checker);
/* open input file */
if (gtf_in_stream->filename) {
if (!(fpin = gt_file_new(gtf_in_stream->filename, "r", err)))
had_err = -1;
}
else
fpin = NULL;
/* parse input file */
if (!had_err) {
filenamestr = gt_str_new_cstr(gtf_in_stream->filename
? gtf_in_stream->filename : "stdin");
had_err = gt_gtf_parser_parse(gtf_parser, gtf_in_stream->genome_node_buffer,
filenamestr, fpin, gtf_in_stream->tidy, err);
gt_str_delete(filenamestr);
}
/* close input file, if necessary */
gt_file_delete(fpin);
/* free */
gt_gtf_parser_delete(gtf_parser);
return had_err;
}
int gt_condenseq_output_to_gff3(const GtCondenseq *condenseq,
GtError *err)
{
int had_err = 0;
GtUword idx,
name_len,
seqnum = 0, seqstart = 0, seqend = 0,
desclen;
GtStr *filename = NULL,
*id = gt_str_new_cstr("U"),
*name = gt_str_new_cstr("unique"),
*parent_unique = gt_str_new_cstr("U"),
*seqid = gt_str_new(),
*source = gt_str_new_cstr("Condenseq");
GtFile *outfile = NULL;
GtGFF3Visitor *gffv = NULL;
GtNodeVisitor *nodev = NULL;
GtFeatureNode *fnode = NULL;
GtGenomeNode *node = NULL;
GtRange range;
gt_assert(condenseq != NULL);
filename = gt_str_new_cstr(gt_condenseq_basefilename(condenseq));
name_len = gt_str_length(name);
gt_str_append_cstr(filename, ".gff3");
outfile = gt_file_new(gt_str_get(filename), "w", err);
nodev = gt_gff3_visitor_new(outfile);
gffv = (GtGFF3Visitor *) nodev;
gt_gff3_visitor_retain_id_attributes(gffv);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
for (idx = 0; !had_err && idx < condenseq->udb_nelems; ++idx) {
GtCondenseqUnique uq = condenseq->uniques[idx];
if (seqend <= uq.orig_startpos) {
const char *desc;
gt_genome_node_delete(node);
seqnum = gt_condenseq_pos2seqnum(condenseq, uq.orig_startpos);
seqstart = gt_condenseq_seqstartpos(condenseq, seqnum);
seqend = seqstart + condenseq_seqlength_help(condenseq, seqnum, seqstart);
desc = gt_condenseq_description(condenseq, &desclen, seqnum);
gt_str_reset(seqid);
gt_str_append_cstr_nt(seqid, desc, desclen);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
}
gt_str_set_length(name, name_len);
gt_str_append_uword(name, idx);
gt_str_set_length(id, (GtUword) 1);
gt_str_append_uword(id, idx);
gt_feature_node_set_attribute(fnode, "Name", gt_str_get(name));
gt_feature_node_set_attribute(fnode, "ID", gt_str_get(id));
/* 1 Based coordinates! */
range.start = uq.orig_startpos + 1 - seqstart;
range.end = uq.orig_startpos + uq.len - seqstart;
gt_genome_node_set_range(node, &range);
had_err = gt_genome_node_accept(node, nodev, err);
}
gt_str_reset(name);
gt_str_append_cstr(name, "link");
gt_str_reset(id);
gt_str_append_cstr(id, "L");
name_len = gt_str_length(name);
seqend = 0;
for (idx = 0; !had_err && idx < condenseq->ldb_nelems; ++idx) {
GtCondenseqLink link = condenseq->links[idx];
if (seqend <= link.orig_startpos) {
const char *desc;
gt_genome_node_delete(node);
seqnum = gt_condenseq_pos2seqnum(condenseq, link.orig_startpos);
seqstart = gt_condenseq_seqstartpos(condenseq, seqnum);
seqend = seqstart + condenseq_seqlength_help(condenseq, seqnum, seqstart);
desc = gt_condenseq_description(condenseq, &desclen, seqnum);
gt_str_reset(seqid);
gt_str_append_cstr_nt(seqid, desc, desclen);
node = gt_feature_node_new(seqid, "experimental_feature", (GtUword) 1,
(GtUword) 1, GT_STRAND_BOTH);
fnode = (GtFeatureNode*) node;
gt_feature_node_set_source(fnode, source);
}
gt_str_set_length(name, name_len);
gt_str_append_uword(name, idx);
gt_str_set_length(id, (GtUword) 1);
gt_str_append_uword(id, idx);
gt_feature_node_set_attribute(fnode, "Name", gt_str_get(name));
gt_feature_node_set_attribute(fnode, "ID", gt_str_get(id));
gt_str_set_length(parent_unique, (GtUword) 1);
gt_str_append_uword(parent_unique, link.unique_id);
gt_feature_node_set_attribute(fnode, "Derives_from",
gt_str_get(parent_unique));
/* 1 Based coordinates! */
range.start = link.orig_startpos + 1 - seqstart;
range.end = link.orig_startpos + link.len - seqstart;
gt_genome_node_set_range(node, &range);
had_err = gt_genome_node_accept(node, nodev, err);
}
gt_file_delete(outfile);
gt_genome_node_delete(node);
gt_node_visitor_delete(nodev);
gt_str_delete(filename);
gt_str_delete(id);
gt_str_delete(name);
gt_str_delete(parent_unique);
gt_str_delete(seqid);
gt_str_delete(source);
return had_err;
}
void vcfoutput_init(VcfOutput *v, const char *of) {
gt_assert(v);
gt_assert(of);
GtError *err = gt_error_new();
v->file = gt_file_new(of,"w", err);
}
int main(int argc, char ** argv)
{
GtNodeStream * in, * score, * out;
GtFile * out_file;
GtError * err;
if (argc != 4)
{
usage(argv[0]);
exit(1);
}
// initilaize genometools
gt_lib_init();
err = gt_error_new();
if (!(in = gt_gff3_in_stream_new_sorted(argv[1])))
{
fprintf(stderr, "Failed to open input stream with arg %s\n", argv[1]);
exit(1);
}
if (!(out_file = gt_file_new(argv[2], "w+", err)))
{
gt_node_stream_delete(in);
fprintf(stderr, "Failed to create output file %s\n", argv[2]);
exit(1);
}
if (!(score = CpGI_score_stream_new(in, argv[3])))
{
gt_file_delete(out_file);
gt_node_stream_delete(in);
fprintf(stderr, "Failed to create CpGI score stream\n");
exit(1);
}
out = gt_gff3_out_stream_new(in, out_file);
if (!(out = gt_gff3_out_stream_new(score, out_file)))
{
gt_node_stream_delete(score);
gt_file_delete(out_file);
gt_node_stream_delete(in);
fprintf(stderr, "Failed to create output stream\n");
exit(1);
}
if (gt_node_stream_pull(out, err))
{
fprintf(stderr, "Failed to pull through out stream\n");
}
// close genome tools
gt_node_stream_delete(out);
gt_node_stream_delete(score);
gt_file_delete(out_file);
gt_node_stream_delete(in);
gt_error_delete(err);
gt_lib_clean();
return 0;
}
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_select_runner(int argc, const char **argv, int parsed_args,
void *tool_arguments, GtError *err)
{
SelectArguments *arguments = tool_arguments;
GtNodeStream *gff3_in_stream, *select_stream,
*targetbest_select_stream = NULL, *gff3_out_stream;
int had_err;
GtFile *drop_file = NULL;
GtNodeVisitor *gff3outvis = NULL;
gt_error_check(err);
gt_assert(arguments);
/* create a gff3 input stream */
gff3_in_stream = gt_gff3_in_stream_new_unsorted(argc - parsed_args,
argv + parsed_args);
if (arguments->verbose && arguments->outfp)
gt_gff3_in_stream_show_progress_bar((GtGFF3InStream*) gff3_in_stream);
/* create a filter stream */
select_stream = gt_select_stream_new(gff3_in_stream, arguments->seqid,
arguments->source,
&arguments->contain_range,
&arguments->overlap_range,
arguments->strand,
arguments->targetstrand,
arguments->has_CDS,
arguments->max_gene_length,
arguments->max_gene_num,
arguments->min_gene_score,
arguments->max_gene_score,
arguments->min_average_splice_site_prob,
arguments->feature_num,
arguments->filter_files,
arguments->filter_logic,
err);
if (select_stream) {
GtSelectStream *fs = (GtSelectStream*) select_stream;
if (gt_str_length(arguments->dropped_file) > 0) {
drop_file = gt_file_new(gt_str_get(arguments->dropped_file), "w", err);
gff3outvis = gt_gff3_visitor_new(drop_file);
gt_select_stream_set_drophandler(fs, print_to_file_drophandler,
(void*) gff3outvis);
} else {
gt_select_stream_set_drophandler(fs, default_drophandler, NULL);
}
gt_select_stream_set_single_intron_factor(select_stream,
arguments->single_intron_factor);
if (arguments->targetbest)
targetbest_select_stream = gt_targetbest_select_stream_new(select_stream);
/* create a gff3 output stream */
gff3_out_stream = gt_gff3_out_stream_new(arguments->targetbest
? targetbest_select_stream
: select_stream,
arguments->outfp);
/* pull the features through the stream and free them afterwards */
had_err = gt_node_stream_pull(gff3_out_stream, err);
/* free */
gt_node_stream_delete(gff3_out_stream);
gt_node_stream_delete(select_stream);
gt_node_stream_delete(targetbest_select_stream);
} else {
had_err = -1;
}
gt_file_delete(drop_file);
gt_node_visitor_delete(gff3outvis);
gt_node_stream_delete(gff3_in_stream);
return had_err;
}
