int main(int argc,char ** argv)
{
SequenceDB * db;
Sequence * seq;
SeqLookupInterface * sli;
SeqLookupPos * slp;
HSPScanInterface * hsi;
LinearHSPmanager * lm;
HitList * hl;
CompMat * mat;
int ret;
HSPScanInterfacePara p;
p.min_score= 30;
p.max_results = 200;
db = single_fasta_SequenceDB(argv[1]);
mat = read_Blast_file_CompMat("blosum62.bla");
sli = new_ghash_SeqLookupInterface();
for(seq = init_SequenceDB(db,&ret); seq != NULL;seq = get_next_SequenceDB(db) ) {
load_aa_flat_Sequence_SeqLookupInterface(sli,hard_link_Sequence(seq));
}
seq = read_fasta_file_Sequence(argv[2]);
assert(seq);
hsi = Wise2_new_one_off_HSPScanInterface(sli,mat,20,10);
/* hspm = simple_HSPScan_scan_query((void*)hsi->data,seq); */
lm = (*hsi->scan_query)(hsi->data,seq,&p);
hl = Wise2_HitList_from_LinearHSPmanager(lm);
Wise2_write_pseudoblast_HitList(hl,stdout);
}
HSPScanInterface * new_HSPScanInterface_from_ScanWiseHSPImpl(ScanWiseHSPImpl * i,ProteinIndexConstructor * pic,SeqLookupLoadPara * slp)
{
HSPScanInterface * out;
SeqLookupInterface * sli;
SequenceDB * db;
Sequence * seq;
CompMat * mat;
int ret;
int c;
mat = read_Blast_file_CompMat(i->matrix_file);
if( i->use_corba == FALSE && i->use_mysql == FALSE && i->use_wiseserver == FALSE && i->use_compress == FALSE && i->use_multiscan == FALSE) {
if( i->direct_sequence == NULL ) {
fatal("If no server based sequence, must have direct sequence");
} else {
db = single_fasta_SequenceDB(i->direct_sequence);
sli = new_SeqLookupInterface_from_ProteinIndexConstructor(pic);
load_SequenceDB_SeqLookupLoadPara(slp,db,sli);
free_SequenceDB(db);
out = new_one_off_HSPScanInterface(sli,mat,15,40);
}
} else if( i->use_corba == TRUE ) {
#ifdef SCAN_CORBA
if( i->ior_file == NULL ) {
fatal("Corba specified, but no ior file given");
}
out = new_corba_HSPScan(sorb,i->ior_file,mat);
#else
fatal("Asking for CORBA, but scanwisep was not compiled with SCAN_CORBA defined.");
#endif
} else if ( i->use_mysql == TRUE ) {
#ifdef SCAN_MYSQL
out = new_HSPScanInterface_MysqlProteinIndex(i->host,i->dbname,i->username,i->password,mat,i->step);
#else
fatal("Asking for mysql, but scanwisep was not compiled with SCAN_MYSQL defined");
#endif
} else if ( i->use_wiseserver == TRUE ) {
#ifdef SCAN_WISESERVER
out = new_wise_transfer_HSPScanInterface(i->host,i->port);
#else
fatal("Asking for wiseserver, but scanwisep was not compiled with SCAN_WISESERVER defined");
#endif
} else if ( i->use_compress == TRUE ) {
#ifdef SCAN_COMPRESS
sli = new_direct_CompressedProteinLookup();
db = single_fasta_SequenceDB(i->direct_sequence);
load_SequenceDB_SeqLookupLoadPara(slp,db,sli);
free_SequenceDB(db);
out = new_one_off_HSPScanInterface(sli,mat,15,40);
#else
fatal("Asking for compressed, but scanwisep was not compiled with SCAN_COMPRESS defined");
#endif
} else if( i->use_multiscan == TRUE ) {
if( i->multiscan_file == NULL ) {
fatal("Must provide a file for a multiple server scan");
}
out = new_multiclient_HSPScanInterface(i->multiscan_file);
}
assert(out != NULL);
free_CompMat(mat); /* hard linked internally */
return out;
}
boolean build_objects(void)
{
boolean ret = TRUE;
Protein * pro_temp;
SequenceDB * psdb;
startend = threestatemodel_mode_from_string(startend_string);
if( startend == TSM_unknown ) {
warn("String %s was unable to converted into a start/end policy\n",startend_string);
ret = FALSE;
}
if( use_single_dna == TRUE ) {
cdna = read_fasta_file_cDNA(dna_seq_file);
if( cdna == NULL ) {
warn("Could not open single dna sequence in %s",dna_seq_file);
ret = FALSE;
}
} else {
sdb = single_fasta_SequenceDB(dna_seq_file);
if( sdb == NULL ) {
warn("Could not build a sequence database on %s",dna_seq_file);
ret = FALSE;
}
}
rm = default_RandomModel();
if( (mat = read_Blast_file_CompMat(matrix_file)) == NULL) {
if( use_tsm == TRUE ) {
info("I could not read the Comparison matrix file in %s; however, you are using a HMM so it is not needed. Please set the WISECONFIGDIR or WISEPERSONALDIR variable correctly to prevent this message.",matrix_file);
} else {
warn("Could not read Comparison matrix file in %s",matrix_file);
ret = FALSE;
}
}
if( is_integer_string(gap_str,&gap) == FALSE ) {
warn("Could not get gap string number %s",gap_str);
ret = FALSE;
}
if( is_integer_string(ext_str,&ext) == FALSE ) {
warn("Could not get ext string number %s",ext_str);
ret = FALSE;
}
if( qstart_str != NULL ) {
if( is_integer_string(qstart_str,&qstart) == FALSE || qstart < 0) {
warn("Could not make %s out as query start",qstart);
ret = FALSE;
}
}
if( qend_str != NULL ) {
if( is_integer_string(qend_str,&qend) == FALSE || qend < 0) {
warn("Could not make %s out as query end",qend);
ret = FALSE;
}
}
if( aln_number_str != NULL ) {
if( is_integer_string(aln_number_str,&aln_number) == FALSE || aln_number < 0) {
warn("Weird aln number string %s...\n",aln_number_str);
ret = FALSE;
}
}
if( report_str != NULL ) {
if( is_integer_string(report_str,&report_stagger) == FALSE ) {
warn("Weird report stagger asked for %s",report_str);
ret = FALSE;
}
}
if( use_pfam1 == TRUE ) {
tsmdb = new_PfamHmmer1DB_ThreeStateDB(protein_file);
if( set_search_type_ThreeStateDB(tsmdb,startend_string) == FALSE) {
warn("Unable to set global/local switch on threestatedb");
ret = FALSE;
}
} else if ( use_pfam2 == TRUE ) {
tsmdb = HMMer2_ThreeStateDB(protein_file);
if( set_search_type_ThreeStateDB(tsmdb,startend_string) == FALSE) {
warn("Unable to set global/local switch on threestatedb");
ret = FALSE;
}
} else if ( use_tsm == TRUE) {
/** using a HMM **/
tsm = HMMer2_read_ThreeStateModel(protein_file);
if( tsm == NULL ) {
warn("Could not read hmm from %s\n",protein_file);
ret = FALSE;
} else {
display_char_in_ThreeStateModel(tsm);
if( hmm_name != NULL ) {
if( tsm->name != NULL )
ckfree(tsm->name);
tsm->name = stringalloc(hmm_name);
} else {
if( tsm->name == NULL ) {
tsm->name = stringalloc(protein_file);
}
}
/** have to set start/end **/
set_startend_policy_ThreeStateModel(tsm,startend,15,0.2);
tsmdb = new_single_ThreeStateDB(tsm,rm);
if( tsmdb == NULL ) {
warn("Could not build a threestatemodel database from a single tsm. Weird!");
ret = FALSE;
}
} /* end of else tsm != NULL */
} /* end of else is tsm */
else if( use_single_pro ) {
if( startend != TSM_default && startend != TSM_global && startend != TSM_local ) {
warn("Proteins can only have local/global startend policies set, not %s",startend_string);
ret = FALSE;
}
if( (pro = read_fasta_file_Protein(protein_file)) == NULL ) {
ret = FALSE;
warn("Could not read Protein sequence in %s",protein_file);
} else {
if( qstart != -1 || qend != -1 ) {
if( qstart == -1 )
qstart = 0;
if( qend == -1 )
qend = pro->baseseq->len;
pro_temp = truncate_Protein(pro,qstart-1,qend);
if( pro_temp == NULL ){
ret = FALSE;
} else {
free_Protein(pro);
pro = pro_temp;
}
}
if( startend == TSM_global)
tsm = global_ThreeStateModel_from_half_bit_Sequence(pro,mat,rm,-gap,-ext);
else
tsm = ThreeStateModel_from_half_bit_Sequence(pro,mat,rm,-gap,-ext);
if( tsm == NULL ) {
warn("Could not build ThreeStateModel from a single protein sequence...");
ret = FALSE;
} else {
tsmdb = new_single_ThreeStateDB(tsm,rm);
if( tsmdb == NULL ) {
warn("Could not build a threestatemodel database from a single tsm. Weird!");
ret = FALSE;
}
} /* end of could build a TSM */
} /* else is a real protein */
} /* end of else is single protein */
else if (use_db_pro == TRUE ) {
psdb = single_fasta_SequenceDB(protein_file);
tsmdb = new_proteindb_ThreeStateDB(psdb,mat,-gap,-ext);
free_SequenceDB(psdb);
}
else {
warn("No protein input file! Yikes!");
}
/***
if( use_tsm == FALSE ) {
} else {
****/
if( main_block_str != NULL ) {
if( is_integer_string(main_block_str,&main_block) == FALSE ) {
warn("Could not get maximum main_block number %s",main_block_str);
ret = FALSE;
}
}
if( evalue_search_str != NULL && is_double_string(evalue_search_str,&evalue_search_cutoff) == FALSE ) {
warn("Could not convert %s to a double",evalue_search_str);
ret = FALSE;
}
if( is_double_string(search_cutoff_str,&search_cutoff) == FALSE ) {
warn("Could not convert %s to a double",search_cutoff_str);
ret = FALSE;
}
if( is_double_string(subs_string,&subs_error) == FALSE ) {
warn("Could not convert %s to a double",subs_error);
ret = FALSE;
}
if( is_double_string(indel_string,&indel_error) == FALSE ) {
warn("Could not convert %s to a double",indel_error);
ret = FALSE;
}
if( is_double_string(allN_string,&allN) == FALSE ) {
warn("Could not convert %s to a double",allN_string);
ret = FALSE;
}
if( strcmp(null_string,"syn") == 0 ) {
use_syn = TRUE;
} else if ( strcmp(null_string,"flat") == 0 ) {
use_syn = FALSE;
} else {
warn("Cannot interpret [%s] as a null model string\n",null_string);
ret = FALSE;
}
if( alg_str != NULL ) {
alg = alg_estwrap_from_string(alg_str);
} else {
alg_str = "312";
alg = alg_estwrap_from_string(alg_str);
}
if( aln_alg_str != NULL ) {
aln_alg = alg_estwrap_from_string(aln_alg_str);
} else {
/* if it is a protein, don't loop */
if( use_single_pro == TRUE || use_db_pro == TRUE )
aln_alg_str = "333";
else
aln_alg_str = "333L";
aln_alg = alg_estwrap_from_string(aln_alg_str);
}
if( (rm = default_RandomModel()) == NULL) {
warn("Could not make default random model\n");
ret = FALSE;
}
if( (ct = read_CodonTable_file(codon_file)) == NULL) {
ret = FALSE;
warn("Could not read codon table file in %s",codon_file);
}
if( (ofp = openfile(output_file,"W")) == NULL) {
warn("Could not open %s as an output file",output_file);
ret = FALSE;
}
rmd = RandomModelDNA_std();
cps = flat_cDNAParser(indel_error);
cm = flat_CodonMapper(ct);
sprinkle_errors_over_CodonMapper(cm,subs_error);
return ret;
}
int main(int argc,char ** argv)
{
DPRunImpl * dpri = NULL;
ScanWiseHSPImpl * scani = NULL;
HSP2HitListImpl * hsp2hiti = NULL;
HitListOutputImpl * hloi = NULL;
ProteinIndexConstructor * pic = NULL;
HSPScanInterface * hsi;
HSPScanInterfacePara * para;
SearchStatInterface * ssi;
SearchStatInterface * ssl;
SeqLookupLoadPara * slp;
HSPset2HitPairPara * hsp2hit;
CompMat * mat;
SequenceDB * db;
Sequence * seq;
int ret;
int i;
int effective_db_size = 300000;
int kk;
int count = 0;
LinearHSPmanager * lm;
HitList * hl;
boolean use_mott = 1;
boolean trunc_best_hsp = 0;
boolean verbose = 0;
static struct rusage use;
struct timeval t0, t1;
gettimeofday(&t0, NULL);
dpri = new_DPRunImpl_from_argv(&argc,argv);
dpri->memory = DPIM_Explicit;
scani = new_ScanWiseHSPImpl_from_argv(&argc,argv);
hsp2hiti = new_HSP2HitListImpl_from_argv(&argc,argv);
hloi = new_HitListOutputImpl_from_argv(&argc,argv);
slp = new_SeqLookupLoadPara_from_argv(&argc,argv);
pic = new_ProteinIndexConstructor_from_argv(&argc,argv);
hsp2hit = new_HSPset2HitPairPara_from_argv(&argc,argv);
para = new_HSPScanInterfacePara_from_argv(&argc,argv);
verbose = strip_out_boolean_argument(&argc,argv,"verbose") ;
strip_out_boolean_def_argument(&argc,argv,"mott",&use_mott);
strip_out_boolean_def_argument(&argc,argv,"besthsp",&trunc_best_hsp);
strip_out_integer_argument(&argc,argv,"dbsize",&effective_db_size);
#ifdef SCAN_CORBA
sorb = get_Wise2Corba_Singleton(&argc,argv,"orbit-local-orb");
#endif
if( dpri == NULL ) {
fatal("Unable to build DPRun implementation. Bad arguments");
}
strip_out_standard_options(&argc,argv,show_help,show_version);
if( argc != 2 ) {
show_help(stdout);
exit(12);
}
/* ugly, but we don't want to bounce matrices around the network... */
mat = read_Blast_file_CompMat("BLOSUM62.bla");
erroroff(REPORT);
hsi = new_HSPScanInterface_from_ScanWiseHSPImpl(scani,pic,slp);
ssi = new_Mott_SearchStatInterface();
ssl = new_lookup_SearchStatInterface(40,2.3);
if( verbose ) {
info("contacted database");
}
db = single_fasta_SequenceDB(argv[1]);
if( db == NULL ) {
fatal("Could not open sequence db...\n");
}
for(seq = init_SequenceDB(db,&ret); seq != NULL;seq = get_next_SequenceDB(db) ) {
count++;
for(i=0;i<seq->len;i++) {
if( !isalpha(seq->seq[i]) ) {
fatal("Sequence position %d [%c] is not valid",i,seq->seq[i]);
}
seq->seq[i] = toupper(seq->seq[i]);
}
info("Processing %s",seq->name);
getrusage(RUSAGE_SELF,&use);
/* info("Before query %s %.3fu %.3fs\n", seq->name,
use.ru_utime.tv_sec + use.ru_utime.tv_usec*MICROSECOND,
use.ru_stime.tv_sec + use.ru_stime.tv_usec*MICROSECOND
);
*/
lm = (*hsi->scan_query)(hsi->data,seq,para);
fprintf(stderr,"Got linear manager is %d entries\n",lm->len);
if( lm->mat == NULL ) {
lm->mat = hard_link_CompMat(mat);
}
getrusage(RUSAGE_SELF,&use);
/*
info("After query %s %.3fu %.3fs\n", seq->name,
use.ru_utime.tv_sec + use.ru_utime.tv_usec*MICROSECOND,
use.ru_stime.tv_sec + use.ru_stime.tv_usec*MICROSECOND
);
*/
sort_LinearHSPmanager(lm,compare_HSPset_score);
if( trunc_best_hsp == 1 ) {
for(kk=1;kk<lm->len;kk++) {
free_HSPset(lm->set[kk]);
lm->set[kk] = NULL;
}
lm->len = 1;
}
getrusage(RUSAGE_SELF,&use);
/*
info("After sort %s %.3fu %.3fs\n", seq->name,
use.ru_utime.tv_sec + use.ru_utime.tv_usec*MICROSECOND,
use.ru_stime.tv_sec + use.ru_stime.tv_usec*MICROSECOND
);
*/
hl = HitList_from_HSP_HSP2HitListImpl(hsp2hiti,lm,dpri,hsp2hit);
getrusage(RUSAGE_SELF,&use);
/*
info("After conversion %s %.3fu %.3fs\n", seq->name,
use.ru_utime.tv_sec + use.ru_utime.tv_usec*MICROSECOND,
use.ru_stime.tv_sec + use.ru_stime.tv_usec*MICROSECOND
);
*/
free_LinearHSPmanager(lm);
if( use_mott == 1 ) {
apply_SearchStat_to_HitList(hl,ssi,effective_db_size);
} else {
for(kk=0;kk<hl->len;kk++) {
hl->pair[kk]->bit_score = hl->pair[kk]->raw_score / 2.0;
}
}
sort_HitList_by_score(hl);
show_HitList_HitListOutputImpl(hloi,hl,stdout);
getrusage(RUSAGE_SELF,&use);
/*
info("After output %s %.3fu %.3fs\n", seq->name,
use.ru_utime.tv_sec + use.ru_utime.tv_usec*MICROSECOND,
use.ru_stime.tv_sec + use.ru_stime.tv_usec*MICROSECOND
);
*/
free_HitList(hl);
free_Sequence(seq);
}
free_DPRunImpl(dpri);
free_HSPScanInterface(hsi);
gettimeofday(&t1, NULL);
fprintf(stderr, "[client stats] queries, time (s): %d %f\n",
count,
(t1.tv_sec - t0.tv_sec) +
(t1.tv_usec - t0.tv_usec) * 1e-6);
return 0;
}