boolean show_pretty_aln(void)
{
Protein * ps;
fprintf(ofp,"\n%s output\nScore %4.2f bits over entire alignment\n",program_name,Score2Bits(pal->score));
if( alg == GWWRAP_2193L || alg == GWWRAP_2193) {
fprintf(ofp,"Entrie alignment score contains unseen 'random' score segments\nYou should only use the per-alignments score printed below\nfor the bits score of the alignment\n\n");
}
if( use_syn == FALSE ) {
fprintf(ofp,"Scores as bits over a flat simple random model\n\n");
} else {
fprintf(ofp,"Scores as bits over a synchronous coding model\n\n");
}
if( use_tsm == FALSE ) {
fprintf(ofp,"Warning: The bits scores is not probablistically correct for single seqs\nSee WWW help for more info\n\n");
protgene_ascii_display(alb,pro->baseseq->seq,pro->baseseq->name,pro->baseseq->offset,gen,ct,15,main_block,(alg == GWWRAP_623L || alg == GWWRAP_2193L || alg == GWWRAP_2193) ? TRUE : FALSE, ofp);
} else {
ps = pseudo_Protein_from_ThreeStateModel(tsm);
protgene_ascii_display(alb,ps->baseseq->seq,ps->baseseq->name,ps->baseseq->offset,gen,ct,15,main_block,(alg == GWWRAP_623L || alg == GWWRAP_2193L || alg == GWWRAP_2193) ? TRUE : FALSE,ofp);
free_Protein(ps);
}
fprintf(ofp,"%s\n",divide_str);
return TRUE;
}
void free_objects(void)
{
if( gwdb != NULL )
gwdb = free_GeneWiseDB(gwdb);
if( cdb != NULL )
cdb = free_cDNADB(cdb);
if( sdb != NULL )
sdb = free_SequenceDB(sdb);
if( cdna != NULL )
cdna = free_cDNA(cdna);
if( pro != NULL )
pro = free_Protein(pro);
if( tsm != NULL )
tsm = free_ThreeStateModel(tsm);
if( tsmdb != NULL )
tsmdb = free_ThreeStateDB(tsmdb);
if( gws != NULL )
gws = free_GeneWiseScore(gws);
if( hs != NULL )
hs = free_Hscore(hs);
if( cm != NULL )
cm = free_CodonMapper(cm);
if( cps != NULL )
cps = free_cDNAParser(cps);
}
/* Function: free_Translation(obj)
*
* Descrip: Free Function: removes the memory held by obj
* Will chain up to owned members and clear all lists
*
*
* Arg: obj [UNKN ] Object that is free'd [Translation *]
*
* Return [UNKN ] Undocumented return value [Translation *]
*
*/
Translation * free_Translation(Translation * obj)
{
int return_early = 0;
if( obj == NULL) {
warn("Attempting to free a NULL pointer to a Translation obj. Should be trappable");
return NULL;
}
#ifdef PTHREAD
assert(pthread_mutex_lock(&(obj->dynamite_mutex)) == 0);
#endif
if( obj->dynamite_hard_link > 1) {
return_early = 1;
obj->dynamite_hard_link--;
}
#ifdef PTHREAD
assert(pthread_mutex_unlock(&(obj->dynamite_mutex)) == 0);
#endif
if( return_early == 1)
return NULL;
/* obj->parent is linked in */
if( obj->protein != NULL)
free_Protein(obj->protein);
ckfree(obj);
return NULL;
}
boolean free_io_objects(void)
{
if( use_tsm == TRUE) {
free_ThreeStateModel(tsm);
} else {
free_Protein(pro);
}
free_CodonTable(ct);
if( gf != NULL ) {
free_GeneFrequency21(gf);
}
free_RandomModelDNA(rmd);
if( is_embl ) {
free_GenomicRegion(embl);
}
free_Genomic(gen);
return TRUE;
}
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 show_output(void)
{
int i;
cDNA * cdna;
Protein * trans;
GenomicOverlapResults * gor;
AlnColumn * alt;
if( show_pretty == TRUE ) {
show_pretty_aln();
}
if( show_match_sum == TRUE ) {
show_MatchSummary_genewise_header(ofp);
show_MatchSummarySet_genewise(mss,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_pretty_gene == TRUE ) {
show_pretty_GenomicRegion(gr,0,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_supp_gene == TRUE ) {
show_pretty_GenomicRegion(gr,1,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_embl == TRUE ) {
write_Embl_FT_GenomicRegion(gr,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_diana == TRUE ) {
write_Diana_FT_GenomicRegion(gr,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_overlap == TRUE ) {
gor = Genomic_overlap(gr,embl);
show_GenomicOverlapResults(gor,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_trans == TRUE ) {
for(i=0;i<gr->len;i++) {
if( gr->gene[i]->ispseudo == TRUE ) {
fprintf(ofp,"#Gene %d is a pseudo gene - no translation possible\n",i);
} else {
trans = get_Protein_from_Translation(gr->gene[i]->transcript[0]->translation[0],ct);
write_fasta_Sequence(trans->baseseq,ofp);
}
}
fprintf(ofp,"%s\n",divide_str);
}
if( show_pep == TRUE ) {
alt = alb->start;
for(;alt != NULL;) {
trans = Protein_from_GeneWise_AlnColumn(gen->baseseq,alt,1,&alt,ct,is_random_AlnColumn_genewise);
if ( trans == NULL )
break;
write_fasta_Sequence(trans->baseseq,ofp);
free_Protein(trans);
}
fprintf(ofp,"%s\n",divide_str);
}
if( show_cdna == TRUE ) {
for(i=0;i<gr->len;i++) {
cdna = get_cDNA_from_Transcript(gr->gene[i]->transcript[0]);
write_fasta_Sequence(cdna->baseseq,ofp);
}
fprintf(ofp,"%s\n",divide_str);
}
if( show_ace == TRUE ) {
show_ace_GenomicRegion(gr,gen->baseseq->name,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_gff == TRUE ) {
show_GFF_GenomicRegion(gr,gen->baseseq->name,"GeneWise",ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_gene_plain == TRUE ) {
show_GenomicRegion(gr,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_AlnBlock == TRUE ) {
mapped_ascii_AlnBlock(alb,Score2Bits,0,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_cumlative_PackAln == TRUE ) {
show_bits_and_cumlative_PackAln(pal,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_PackAln == TRUE ) {
show_simple_PackAln(pal,ofp);
fprintf(ofp,"%s\n",divide_str);
}
return TRUE;
}
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;
}
boolean show_output(void)
{
int i,k;
ThreeStateModel * temptsm;
AlnBlock * alb;
PackAln * pal;
MatchSummarySet * mss;
Protein * ps;
cDNA * cdna;
double bits;
boolean fitted_res = FALSE;
AlnBlockList * alist;
AlnBlock * anchored;
SequenceSet * set;
AlnColumn * alt;
Protein * trans;
/* sort by bit score first */
sort_Hscore_by_score(hs);
if( search_mode == PC_SEARCH_S2DB ) {
if( hs->his == NULL || hs->his->total < 1000 ) {
info("Cannot fit histogram to a db smaller than 1,000");
fprintf(ofp,"[Warning: Can't fit histogram to a db smaller than 1,000]\n\n");
show_histogram = FALSE;
} else {
fitted_res = TRUE;
fit_Hscore_to_EVD(hs,20);
}
}
/* deal with initialising anchored alignment.
* Could be done for either single HMMs or single proteins,
* but we will only do it for HMMs at the moment
*/
if( make_anchored_aln == TRUE ) {
if( tsm == NULL ) {
warn("Attempting to make an achored alignment without a HMM. impossible!");
make_anchored_aln = FALSE;
} else {
anchored = single_unit_AlnBlock(tsm->len,"MATCH_STATE");
set = SequenceSet_alloc_std();
}
}
/* dofus catcher */
if( aln_alg != alg ) {
fprintf(ofp,"\n#\n#WARNING!\n#\n# Your alignment algorithm is different from your search algorithm.\n# This is probably quite sensible but will lead to differing scores.\n# Use the search score as an indicator of the significance of the match\n# Read the docs for more information\n#\n");
}
fprintf(ofp,"\n\n#High Score list\n");
fprintf(ofp,"#Protein ID DNA Str ID Bits Evalue\n");
fprintf(ofp,"--------------------------------------------------------------------------\n");
for(i=0;i<hs->len;i++) {
bits = Score2Bits(hs->ds[i]->score);
if( bits < search_cutoff ) {
break;
}
if( fitted_res == TRUE && evalue_search_str != NULL ) {
if( hs->ds[i]->evalue > evalue_search_cutoff )
break;
}
if( fitted_res == TRUE)
fprintf(ofp,"Protein %-20sDNA [%c] %-24s %.2f %.2g\n",hs->ds[i]->query->name,hs->ds[i]->target->is_reversed == TRUE ? '-' : '+',hs->ds[i]->target->name,bits,hs->ds[i]->evalue);
else
fprintf(ofp,"Protein %-20sDNA [%c] %-24s %.2f\n",hs->ds[i]->query->name,hs->ds[i]->target->is_reversed == TRUE ? '-' : '+',hs->ds[i]->target->name,bits);
}
if( search_mode == PC_SEARCH_S2DB && show_histogram == TRUE ) {
fprintf(ofp,"\n\n#Histogram\n");
fprintf(ofp,"-----------------------------------------------------------------------\n");
PrintASCIIHistogram(hs->his,ofp);
}
fprintf(ofp,"\n\n#Alignments\n");
fprintf(ofp,"-----------------------------------------------------------------------\n");
for(i=0;i<hs->len;i++) {
bits = Score2Bits(hs->ds[i]->score);
if( bits < search_cutoff ) {
break;
}
if( i >= aln_number ) {
break;
}
if( fitted_res == TRUE && evalue_search_str != NULL ) {
if( hs->ds[i]->evalue > evalue_search_cutoff )
break;
}
fprintf(ofp,"\n\n>Results for %s vs %s (%s) [%d]\n",hs->ds[i]->query->name,hs->ds[i]->target->name,hs->ds[i]->target->is_reversed == TRUE ? "reverse" : "forward",i+1 );
cdna = get_cDNA_from_cDNADB(cdb,hs->ds[i]->target);
temptsm = indexed_ThreeStateModel_ThreeStateDB(tsmdb,hs->ds[i]->query);
alb = AlnBlock_from_TSM_estwise_wrap(temptsm,cdna,cps,cm,ct,rmd,aln_alg,use_syn,allN,flat_insert,dpri,&pal);
if( alb == NULL ) {
warn("Got a NULL alignment. Exiting now due to presumed problems");
fprintf(ofp,"\n\n*Got a NULL alignment. Exiting now due to presumed problems*\n\n");
return FALSE;
}
if( use_single_pro == FALSE)
mss = MatchSummarySet_from_AlnBlock_genewise(alb,temptsm->name,1,cdna->baseseq);
else
mss = MatchSummarySet_from_AlnBlock_genewise(alb,pro->baseseq->name,pro->baseseq->offset,cdna->baseseq);
if( show_pretty == TRUE ) {
fprintf(ofp,"\n%s output\nScore %4.2f bits over entire alignment.\nThis will be different from per-alignment scores. See manual for details\nFor computer parsable output, try %s -help or read the manual\n",program_name,Score2Bits(pal->score),program_name);
if( use_syn == FALSE ) {
fprintf(ofp,"Scores as bits over a flat simple random model\n\n");
} else {
fprintf(ofp,"Scores as bits over a synchronous coding model\n\n");
}
ps = pseudo_Protein_from_ThreeStateModel(temptsm);
protcdna_ascii_display(alb,ps->baseseq->seq,ps->baseseq->name,ps->baseseq->offset,cdna,ct,15,main_block,TRUE,ofp);
free_Protein(ps);
fprintf(ofp,"%s\n",divide_str);
}
if( show_match_sum == TRUE ) {
show_MatchSummary_genewise_header(ofp);
show_MatchSummarySet_genewise(mss,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_pep == TRUE ) {
alt = alb->start;
for(;alt != NULL;) {
trans = Protein_from_GeneWise_AlnColumn(cdna->baseseq,alt,1,&alt,ct,is_random_AlnColumn_genewise);
if ( trans == NULL )
break;
write_fasta_Sequence(trans->baseseq,ofp);
free_Protein(trans);
}
fprintf(ofp,"%s\n",divide_str);
}
if( show_AlnBlock == TRUE ) {
mapped_ascii_AlnBlock(alb,Score2Bits,0,ofp);
fprintf(ofp,"%s\n",divide_str);
}
if( show_PackAln == TRUE ) {
show_simple_PackAln(pal,ofp);
fprintf(ofp,"%s\n",divide_str);
}
/*
* This goes at the end because it destroys the alb structure
*/
if( make_anchored_aln == TRUE ) {
/* attach sequence to als in alb, so we have it for later use */
alb->seq[1]->data = (void *) cdna->baseseq;
/* add to SequenceSet so we can destroy the memory */
add_SequenceSet(set,hard_link_Sequence(cdna->baseseq));
alist = split_AlnBlock(alb,is_random_AlnColumn_genewise);
for(k=0;k<alist->len;k++) {
/* actually produce the anchored alignment */
/*mapped_ascii_AlnBlock(alist->alb[k],Score2Bits,stderr);*/
add_to_anchored_AlnBlock(anchored,alist->alb[k]);
/* dump_ascii_AlnBlock(anchored,stderr);*/
}
}
alb = free_AlnBlock(alb);
pal = free_PackAln(pal);
mss = free_MatchSummarySet(mss);
cdna = free_cDNA(cdna);
temptsm = free_ThreeStateModel(temptsm);
}
if( do_complete_analysis == TRUE ) {
fprintf(ofp,"\n\n#Complete Analysis\n");
fprintf(ofp,"-------------------------------------------------------------\n\n");
/* ok - end of loop over relevant hits. If we have an
* anchored alignment, print it out!
*/
if( make_anchored_aln == TRUE ) {
/*dump_ascii_AlnBlock(anchored,stderr);*/
write_mul_estwise_AlnBlock(anchored,ct,ofp);
fprintf(ofp,"%s\n",divide_str);
}
}
return TRUE;
}
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;
}
void show_verbose_evo(AlnBlock * alb,ThreeStateModel * tsm,Sequence * ref,Sequence * diff,CodonTable * ct,FILE * ofp)
{
AlnColumn * alc;
Protein * hmmp;
Sequence * ref_trans;
Sequence * diff_trans;
DnaProbMatrix * negative_dm;
DnaProbMatrix * pseudo_dm;
int i;
int count = 0;
double est_mutation = 0.0;
int dna_offset;
Score total_pseudo = 0;
Score total_neg = 0;
Score pseudo = 0;
Score neg = 0;
int count_ref_positive = 0;
int count_ref_negative = 0;
int count_ref_negative_0_5 = 0;
int count_ref_negative_5_10 = 0;
int count_ref_negative_10_15 = 0;
int syn_sites = 0;
int nonsyn_sites = 0;
int syn_changes = 0;
int nonsyn_changes = 0;
int diff_score;
char diff_aa;
char ref_aa;
int score_ratio = 0;
Score score_neg_5 = Probability2Score(Bits2Probability(-5.0));
Score score_neg_10 = Probability2Score(Bits2Probability(-10.0));
int k;
for(i=0;i<ref->len;i+=3) {
/* if this has changed, then it is definitely non syn */
if( aminoacid_from_seq(ct,ref->seq+i) != aminoacid_from_seq(ct,diff->seq+i)) {
for(k=0;k<3;k++) {
if( ref->seq[i+k] != diff->seq[i+k] ) {
nonsyn_changes++;
}
}
} else {
/* could still be syn change */
for(k=0;k<3;k++) {
if( ref->seq[i+k] != diff->seq[i+k] ) {
syn_changes++;
}
}
}
/* calculate the sites. There is always 2 non syn sites */
nonsyn_sites += 2;
if( four_fold_sites_CodonTable(ct,ref->seq+i) > 0 ) {
syn_sites++;
} else {
nonsyn_sites += 1;
}
}
for(i=0;i<ref->len;i++) {
if( ref->seq[i] != diff->seq[i] ) {
count++;
}
}
est_mutation = (double)count / (double)ref->len;
pseudo_dm = DnaProbMatrix_from_match(1.0 - est_mutation,NMaskType_BANNED);
negative_dm = DnaProbMatrix_from_match(1.0 - (est_mutation*2),NMaskType_BANNED);
ref_trans = translate_Sequence(ref,ct);
diff_trans = translate_Sequence(diff,ct);
hmmp = pseudo_Protein_from_ThreeStateModel(tsm);
for(alc=alb->start;alc != NULL;alc = alc->next) {
/* fprintf(stdout,"In position %s\n",alc->alu[0]->text_label); */
if( strcmp(alc->alu[0]->text_label,"SEQUENCE") == 0 &&
strcmp(alc->alu[1]->text_label,"SEQUENCE") == 0 ) {
dna_offset = alc->alu[1]->end*3;
pseudo = logl_pseudogene(ref->seq+dna_offset,diff->seq+dna_offset,pseudo_dm);
neg = logl_negative_selection(ref->seq+dna_offset,diff->seq+dna_offset,tsm->unit[alc->alu[0]->end],ct,
pseudo_dm);
/*
fprintf(ofp,"Position %d [%c], vs %d [%c,%c] Scores Negative %d, Pseudo %d\n",
alc->alu[0]->end,hmmp->baseseq->seq[alc->alu[0]->end],
alc->alu[1]->end,ref_trans->seq[alc->alu[1]->end],diff_trans->seq[alc->alu[1]->end],
neg,
pseudo
);
*/
ref_aa = ref_trans->seq[alc->alu[1]->end];
diff_aa = diff_trans->seq[alc->alu[1]->end];
if( ref_aa != diff_aa ) {
score_ratio += Probability2Score(tsm->unit[alc->alu[0]->end]->match_emission[ref_aa-'A']) - Probability2Score(tsm->unit[alc->alu[0]->end]->match_emission[diff_aa-'A']);
diff_score = Probability2Score(tsm->unit[alc->alu[0]->end]->match_emission[ref_aa-'A']) - Probability2Score(tsm->unit[alc->alu[0]->end]->match_emission[diff_aa-'A']);
if( diff_score < 0) {
count_ref_negative++;
if( diff_score > score_neg_5 ) {
count_ref_negative_0_5++;
} else if ( diff_score > score_neg_10 ) {
count_ref_negative_5_10++;
} else {
count_ref_negative_10_15++;
}
} else {
count_ref_positive++;
}
}
total_pseudo += pseudo;
total_neg += neg;
}
}
fprintf(ofp,"%s\t%s\t%.2f\t%d\t%d\t%d\t%d\t%d\n",ref->name,hmmp->baseseq->name,Score2Bits(score_ratio),
count_ref_positive,count_ref_negative,
count_ref_negative_0_5,
count_ref_negative_5_10,
count_ref_negative_10_15);
/*
fprintf(ofp,"%s,%s Total Pseudo %d vs Negative %d, Ratio %.4f Positive %d Negative %d Score %.2f Syn %d Changes %d NonSyn %d Changes %d\n",ref->name,hmmp->baseseq->name,total_pseudo,total_neg,Score2Bits(total_neg-total_pseudo),count_ref_positive,count_ref_negative,Score2Bits(score_ratio),syn_sites,syn_changes,nonsyn_sites,nonsyn_changes);
*/
free_Protein(hmmp);
}
