ThreeStateScore * reload_ThreeStateDB(ThreeStateScore * prev,ThreeStateDB * mdb,int * return_status)
{
ThreeStateModel * tsm;
ThreeStateScore * tss;
free_ThreeStateScore(prev);
if( mdb->dbtype == TSMDB_SINGLE ) {
*return_status = DB_RETURN_END;
return NULL;
}
tsm = read_TSM_ThreeStateDB(mdb,return_status);
if( *return_status != DB_RETURN_OK)
return NULL;
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
fold_RandomModel_into_ThreeStateModel(tsm,mdb->rm);
tss = ThreeStateScore_from_ThreeStateModel(tsm);
free_ThreeStateModel(tsm);
*return_status = DB_RETURN_OK;
return tss;
}
ThreeStateScore * init_ThreeStateDB(ThreeStateDB * mdb,int * return_status)
{
ThreeStateModel * tsm;
ThreeStateScore * tss;
if( open_ThreeStateDB(mdb) == FALSE) {
warn("Could not open ThreeStateDB, hence could not init it!");
*return_status = DB_RETURN_ERROR;
return NULL;
}
tsm = read_TSM_ThreeStateDB(mdb,return_status);
if( *return_status == DB_RETURN_ERROR)
return NULL;
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
fold_RandomModel_into_ThreeStateModel(tsm,mdb->rm);
tss = ThreeStateScore_from_ThreeStateModel(tsm);
free_ThreeStateModel(tsm);
*return_status = DB_RETURN_OK;
return tss;
}
boolean build_objects(void)
{
boolean ret = TRUE;
Protein * pro_temp;
Genomic * gen_temp;
FILE * ifp;
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( tstart_str != NULL ) {
if( is_integer_string(tstart_str,&tstart) == FALSE || tstart < 0) {
warn("Could not make %s out as target start",tstart);
ret = FALSE;
}
}
if( tend_str != NULL ) {
if( is_integer_string(tend_str,&tend) == FALSE || tend < 0) {
warn("Could not make %s out as target end",tend);
ret = FALSE;
}
}
if( is_integer_string(gap_str,&gap) == FALSE ) {
warn("Could not make %s out as gap penalty (must be integer at the moment)",gap_str);
ret = FALSE;
}
if( is_integer_string(ext_str,&ext) == FALSE ) {
warn("Could not make %s out as gap penalty (must be integer at the moment)",ext_str);
ret = FALSE;
}
if( is_embl == FALSE ) {
if( (gen = read_fasta_file_Genomic(dna_seq_file,length_of_N)) == NULL ) {
ret = FALSE;
warn("Could not read genomic sequence in %s",dna_seq_file);
gen = NULL;
}
} else {
embl = read_EMBL_GenomicRegion_file(dna_seq_file);
if( embl == NULL ) {
warn("Could not read genomic EMBL file in %s",dna_seq_file);
gen = NULL;
ret = FALSE;
} else {
gen = hard_link_Genomic(embl->genomic);
}
}
if( gen != NULL ) {
if( tstart != -1 || tend != -1 ) {
if( tstart == -1 )
tstart = 0;
if( tend == -1 )
tend = gen->baseseq->len;
gen_temp = truncate_Genomic(gen,tstart-1,tend);
if( gen_temp == NULL ){
ret = FALSE;
} else {
free_Genomic(gen);
gen = gen_temp;
}
} else {
/* no truncation required */
}
if( reverse == TRUE ) {
if( tstart > tend ) {
warn("You have already reversed the DNA by using %d - %d truncation. Re-reversing",tstart,tend);
}
gen_temp = reverse_complement_Genomic(gen);
free_Genomic(gen);
gen = gen_temp;
}
}
/*
* Can't truncate on GenomicRegion (for good reasons!).
* but we want only a section of the EMBL file to be used
*
* So... swap genomic now. Positions in EMBL are still valid,
* however - some genes will loose their sequence, which will be damaging. ;)
*/
if( is_embl ) {
free_Genomic(embl->genomic);
embl->genomic = hard_link_Genomic(gen); /* pointer could be dead anyway ;) */
}
if( target_abs == TRUE ) {
if( is_embl == TRUE ) {
warn("Sorry you can't both use absolute positioning and EMBL files as I can't cope with all the coordinate remapping. You'll have to convert to fasta.");
ret = FALSE;
}
gen->baseseq->offset = 1;
gen->baseseq->end = strlen(gen->baseseq->seq);
}
if( alg_str != NULL ) {
alg = gwrap_alg_type_from_string(alg_str);
} else {
if( use_tsm == TRUE ) {
alg_str = "623L";
} else {
alg_str = "623";
}
alg = gwrap_alg_type_from_string(alg_str);
}
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( use_tsm == FALSE ) {
if( startend != TSM_default && startend != TSM_global && startend != TSM_local && startend != TSM_endbiased) {
warn("Proteins can only have local/global/endbias 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;
}
}
}
} else {
/** using a HMM **/
/*tsm = read_HMMer_1_7_ascii_file(hmm_file);*/
/*tsm = Wise2_read_ThreeStateModel_from_hmmer1_file(hmm_file);*/
tsm = HMMer2_read_ThreeStateModel(hmm_file);
if( tsm == NULL ) {
warn("Could not read hmm from %s\n",hmm_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);
}
if( tsm == NULL ) {
warn("Could not read %s as a hmm",hmm_file);
}
/** have to set start/end **/
set_startend_policy_ThreeStateModel(tsm,startend,30,0.1);
}
} /* end of else tsm != NULL */
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( 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(cfreq_string,"model") == 0 ) {
model_codon = TRUE;
} else if ( strcmp(cfreq_string,"flat") == 0 ) {
model_codon = FALSE;
} else {
warn("Cannot interpret [%s] as a codon modelling parameter\n",cfreq_string);
ret = FALSE;
}
if( strcmp(splice_string,"model") == 0 ) {
model_splice = TRUE;
} else if ( strcmp(splice_string,"flat") == 0 ) {
model_splice = FALSE;
gmp->use_gtag_splice = TRUE;
} else {
warn("Cannot interpret [%s] as a splice modelling parameter\n",splice_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( strcmp(intron_string,"model") == 0 ) {
use_tied_model = FALSE;
} else if ( strcmp(intron_string,"tied") == 0 ) {
use_tied_model = TRUE;
} else {
warn("Cannot interpret [%s] as a intron tieing switch\n",intron_string);
ret = FALSE;
}
if( (rm = default_RandomModel()) == NULL) {
warn("Could not make default random model\n");
ret = FALSE;
}
if( use_new_stats == 0 ) {
if( (gf = read_GeneFrequency21_file(gene_file)) == NULL) {
ret = FALSE;
warn("Could not read a GeneFrequency file in %s",gene_file);
}
} else {
if( (gs = GeneStats_from_GeneModelParam(gmp)) == NULL ){
ret=FALSE;
warn("Could not read gene statistics in %s",new_gene_file);
}
} /* end of else using new gene stats */
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( (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();
return ret;
}
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;
}
ThreeStateModel * read_TSM_ThreeStateDB(ThreeStateDB * mdb,int * return_status)
{
ThreeStateModel * tsm;
Protein * pro;
Sequence * seq;
if( mdb->hmm_model_end != -1 && mdb->current_no == mdb->hmm_model_end ) {
*return_status = DB_RETURN_END;
return NULL;
}
mdb->current_no++;
switch( mdb->dbtype ) {
case TSMDB_SINGLE :
*return_status = DB_RETURN_END;
if( mdb->single->rm == NULL ) {
warn("Threestate model without an internal random model!");
mdb->single->rm = hard_link_RandomModel(mdb->rm);
}
return hard_link_ThreeStateModel(mdb->single);
case TSMDB_HMMER1PFAM :
tsm= read_next_TSM_PfamHmmer1DB(mdb->phdb,return_status);
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
return tsm;
case TSMDB_PROTEIN :
if( mdb->seq_cache != NULL ) {
/* just after an open. Should actually use this sequence, and flush the cache */
pro = Protein_from_Sequence(hard_link_Sequence(mdb->seq_cache));
mdb->seq_cache = free_Sequence(mdb->seq_cache);
*return_status = DB_RETURN_OK;
} else {
/* reload a sequence from a database */
seq = reload_SequenceDB(NULL,mdb->sdb,return_status);
/* exit now if error */
if( *return_status == DB_RETURN_ERROR ) {
return NULL; /* might have leaked memory. Ugh! */
}
/* if we get NULL... for the moment, silent flag end */
if( seq == NULL ) {
*return_status = DB_RETURN_END;
return NULL;
}
pro = Protein_from_Sequence(seq);
}
if( pro == NULL ) {
warn("Could not convert sequence to a protein. Exiting!");
*return_status = DB_RETURN_ERROR;
return NULL;
}
/* convert protein to threestatemodel */
tsm = ThreeStateModel_from_half_bit_Sequence(pro,mdb->comp,mdb->rm,mdb->gap,mdb->ext);
if( tsm == NULL ) {
warn("Could not convert protein to threestatemode. Exiting!");
free_Protein(pro);
*return_status = DB_RETURN_ERROR;
return NULL;
}
/* DB status already set by seqdb */
return tsm;
case TSMDB_GENERIC :
tsm = ((*mdb->reload_generic)(mdb,return_status));
if( tsm == NULL ) {
return NULL; /* means end of database */
}
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
return tsm;
default :
warn("Got an unrecognisable tsm db type in read-load");
return NULL;
}
}
ThreeStateModel * indexed_ThreeStateModel_ThreeStateDB(ThreeStateDB * mdb,DataEntry * en)
{
Sequence * seq;
Protein * pro;
ThreeStateModel * tsm;
switch(mdb->dbtype) {
case TSMDB_SINGLE :
return hard_link_ThreeStateModel(mdb->single);
case TSMDB_HMMER1PFAM :
tsm = ThreeStateModel_from_name_PfamHmmer1DB(mdb->phdb,en->name);
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
return tsm;
case TSMDB_PROTEIN :
seq = get_Sequence_from_SequenceDB(mdb->sdb,en);
if( seq == NULL ) {
warn("could not retrieve %s as a sequence from database",en->name);
return NULL;
}
pro = Protein_from_Sequence(seq);
if( pro == NULL ) {
warn("Could not convert sequence to a protein. Exiting!");
return NULL;
}
/* convert protein to threestatemodel */
tsm = ThreeStateModel_from_half_bit_Sequence(pro,mdb->comp,mdb->rm,mdb->gap,mdb->ext);
if( tsm == NULL ) {
warn("Could not convert protein to threestatemode. Exiting!");
free_Protein(pro);
return NULL;
}
free_Protein(pro);
/* DB status already set by seqdb */
return tsm;
case TSMDB_GENERIC :
tsm = ((*mdb->index_generic)(mdb,en));
if( tsm == NULL ) {
return NULL;
}
/* fprintf(stdout,"Setting %d as policy\n",mdb->type); */
set_startend_policy_ThreeStateModel(tsm,mdb->type,30,0.2);
return tsm;
default :
warn("Unknown threestatedb type");
return NULL;
}
warn("Should never get here - in threestatedb reload!");
return NULL;
}
int main(int argc,char ** argv)
{
int i;
SequenceSet * in;
Sequence * trans;
ThreeStateDB * tsd;
DPRunImpl * dpri;
CodonTable * ct;
int return_status;
ThreeStateModel * tsm;
ThreeStateScore * tss;
Protein * hmmp;
ComplexSequence * cs;
ComplexSequenceEvalSet * cses;
PackAln * pal;
AlnBlock * alb;
int show_align = 0;
int show_alb = 0;
int show_verbose = 1;
int show_trans = 0;
ct = read_CodonTable_file("codon.table");
cses = default_aminoacid_ComplexSequenceEvalSet();
dpri = new_DPRunImpl_from_argv(&argc,argv);
strip_out_boolean_def_argument(&argc,argv,"pretty",&show_align);
strip_out_boolean_def_argument(&argc,argv,"alb",&show_alb);
strip_out_boolean_def_argument(&argc,argv,"trans",&show_trans);
if( argc != 3 ) {
show_help(stdout);
exit(63);
}
in = read_fasta_SequenceSet_file(argv[1]);
tsd = HMMer2_ThreeStateDB(argv[2]);
assert(in);
assert(tsd);
assert(in->len == 2);
trans = translate_Sequence(in->set[0],ct);
if( show_trans ) {
write_fasta_Sequence(trans,stdout);
}
cs = new_ComplexSequence(trans,cses);
open_ThreeStateDB(tsd);
while( (tsm = read_TSM_ThreeStateDB(tsd,&return_status)) != NULL ) {
fold_RandomModel_into_ThreeStateModel(tsm,tsm->rm);
set_startend_policy_ThreeStateModel(tsm,TSM_local,10,1.0);
tss = ThreeStateScore_from_ThreeStateModel(tsm);
hmmp = pseudo_Protein_from_ThreeStateModel(tsm);
pal = PackAln_bestmemory_ThreeStateLoop(tss,cs,NULL,dpri);
alb = convert_PackAln_to_AlnBlock_ThreeStateLoop(pal);
if( show_alb ) {
show_flat_AlnBlock(alb,stdout);
}
if( show_align ) {
write_pretty_seq_align(alb,hmmp->baseseq,trans,15,50,stdout);
}
if( show_verbose ) {
show_verbose_evo(alb,tsm,in->set[0],in->set[1],ct,stdout);
}
}
}