AlnBlock * Align_Sequences_ProteinBlockAligner(Sequence * one,Sequence * two,CompMat * comp,int bentry,int bexit,int b1exit,int b_self,int b_on,DPRunImpl * dpri)
{
ComplexSequence * cone=NULL;
ComplexSequence * ctwo=NULL;
AlnBlock * alb= NULL;
PackAln * pal=NULL;
ComplexSequenceEvalSet * evalfunc = NULL;
if( one == NULL || two == NULL || comp == NULL ) {
warn("Passed in NULL objects into Align_Sequences_ProteinSmithWaterman!");
return NULL;
}
if( one->type != SEQUENCE_PROTEIN ) {
warn("Sequence %s is not typed as protein... ignoring!\n",one->name);
}
if( two->type != SEQUENCE_PROTEIN ) {
warn("Sequence %s is not typed as protein... ignoring!\n",two->name);
}
evalfunc = default_aminoacid_ComplexSequenceEvalSet();
cone = new_ComplexSequence(one,evalfunc);
if( cone == NULL )
goto cleanup;
ctwo = new_ComplexSequence(two,evalfunc);
if( ctwo == NULL )
goto cleanup;
pal = PackAln_bestmemory_ProteinBlockAligner(cone,ctwo,comp,bentry,b1exit,b_on,b_self,bexit,NULL,dpri);
if( pal == NULL )
goto cleanup;
alb = convert_PackAln_to_AlnBlock_ProteinSW(pal);
goto cleanup;
cleanup :
if( cone != NULL )
free_ComplexSequence(cone);
if( ctwo != NULL )
free_ComplexSequence(ctwo);
if( pal != NULL )
free_PackAln(pal);
if( evalfunc != NULL )
free_ComplexSequenceEvalSet(evalfunc);
return alb;
}
AlnBlock * Align_Sequences_ProteinSmithWaterman(Sequence * one,Sequence * two,CompMat * comp,int gap,int ext,DPRunImpl * dpri)
{
AlnBlock * out = NULL;
ComplexSequenceEvalSet * evalfunc = NULL;
ComplexSequence * query_cs = NULL;
ComplexSequence * target_cs = NULL;
PackAln * pal = NULL;
if( one == NULL || two == NULL || comp == NULL ) {
warn("Passed in NULL objects into Align_Sequences_ProteinSmithWaterman!");
return NULL;
}
if( one->type != SEQUENCE_PROTEIN ) {
warn("Sequence %s is not typed as protein... ignoring!\n",one->name);
}
if( two->type != SEQUENCE_PROTEIN ) {
warn("Sequence %s is not typed as protein... ignoring!\n",two->name);
}
if( gap > 0 || ext > 0 ) {
warn("Gap penalties %d,%d only make sense if they are negative",gap,ext);
return NULL;
}
evalfunc = default_aminoacid_ComplexSequenceEvalSet();
query_cs = new_ComplexSequence(one,evalfunc);
if( query_cs == NULL )
goto cleanup;
target_cs = new_ComplexSequence(two,evalfunc);
if( target_cs == NULL )
goto cleanup;
pal = PackAln_bestmemory_ProteinSW(query_cs,target_cs,comp,gap,ext,NULL,dpri);
if( pal == NULL )
goto cleanup;
out = convert_PackAln_to_AlnBlock_ProteinSW(pal);
goto cleanup;
cleanup :
if( query_cs != NULL )
free_ComplexSequence(query_cs);
if( target_cs != NULL )
free_ComplexSequence(target_cs);
if( pal != NULL )
free_PackAln(pal);
if( evalfunc != NULL )
free_ComplexSequenceEvalSet(evalfunc);
return out;
}
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);
}
}
}
int main(int argc,char ** argv)
{
Sequence * query;
Sequence * target;
CompMat * comp;
char * comp_file;
int gap = (12);
int ext = (2);
int a = 120;
int b = 10;
int c = 3;
ComplexSequence * query_cs;
ComplexSequence * target_cs;
ComplexSequenceEvalSet * evalfunc;
boolean show_label_output = FALSE;
boolean show_fancy_output = FALSE;
boolean use_abc = FALSE;
PackAln * pal;
AlnBlock * alb;
DPRunImpl * dpri = NULL;
/*
* Process command line options
* -h or -help gives us help
* -g for gap value (an int) - rely on commandline error processing
* -e for ext value (an int) - rely on commandline error processing
* -m for matrix (a char)
* -l - label output
* -f - fancy output
*
*
* Use calls to commandline.h functions
*
*/
if( strip_out_boolean_argument(&argc,argv,"h") == TRUE || strip_out_boolean_argument(&argc,argv,"-help") == TRUE) {
show_help(stdout);
exit(1);
}
dpri = new_DPRunImpl_from_argv(&argc,argv);
if( dpri == NULL ) {
fatal("Unable to build DPRun implementation. Bad arguments");
}
show_label_output = strip_out_boolean_argument(&argc,argv,"l");
show_fancy_output = strip_out_boolean_argument(&argc,argv,"f");
/** if all FALSE, set fancy to TRUE **/
if( show_label_output == FALSE )
show_fancy_output = TRUE;
(void) strip_out_integer_argument(&argc,argv,"g",&gap);
(void) strip_out_integer_argument(&argc,argv,"e",&ext);
(void) strip_out_integer_argument(&argc,argv,"a",&a);
(void) strip_out_integer_argument(&argc,argv,"b",&b);
(void) strip_out_integer_argument(&argc,argv,"c",&c);
use_abc = strip_out_boolean_argument(&argc,argv,"abc");
comp_file = strip_out_assigned_argument(&argc,argv,"m");
if( comp_file == NULL)
comp_file = "blosum62.bla";
if( argc != 3 ) {
warn("Must have two arguments for sequence 1 and sequence 2 %d",argc);
show_help(stdout);
exit(1);
}
/*
* Read in two sequences
*/
if( (query=read_fasta_file_Sequence(argv[1])) == NULL ) {
fatal("Unable to read the sequence in file %s",argv[1]);
}
if( (target=read_fasta_file_Sequence(argv[2])) == NULL ) {
fatal("Unable to read the sequence in file %s",argv[2]);
}
/*
* Open a blosum matrix. This will be opened from WISECONFIGDIR
* or WISEPERSONALDIR if it is not present in the current directory.
*/
comp = read_Blast_file_CompMat(comp_file);
if( comp == NULL ) {
fatal("unable to read file %s",comp_file);
}
/* if abc - factor up matrix! */
if( use_abc == TRUE ) {
factor_CompMat(comp,10);
}
/*
* Make an alignment. I don't care about the implementation:
* hand it over to sw_wrap function to do it
*
*/
if( use_abc ) {
evalfunc = default_aminoacid_ComplexSequenceEvalSet();
query_cs = new_ComplexSequence(query,evalfunc);
if( query_cs == NULL )
fatal("Cannot build cs objects!");
target_cs = new_ComplexSequence(target,evalfunc);
if( target_cs == NULL )
fatal("Cannot build cs objects!");
pal = PackAln_bestmemory_abc(query_cs,target_cs,comp,-a,-b,-c,NULL,dpri);
alb = convert_PackAln_to_AlnBlock_abc(pal);
free_ComplexSequence(query_cs);
free_ComplexSequence(target_cs);
} else {
alb = Align_Sequences_ProteinSmithWaterman(query,target,comp,-gap,-ext,dpri);
}
/*
* show output. If multiple outputs, divide using //
*/
if( show_label_output == TRUE ) {
show_flat_AlnBlock(alb,stdout);
puts("//\n");
}
if( show_fancy_output == TRUE ) {
write_pretty_seq_align(alb,query,target,15,50,stdout);
puts("//\n");
}
/*
* Destroy the memory.
*/
free_Sequence(query);
free_Sequence(target);
free_CompMat(comp);
free_AlnBlock(alb);
return 0;
}
int main(int argc,char ** argv)
{
Sequence * query;
Sequence * target;
ComplexSequence * query_cs;
ComplexSequence * target_cs;
ComplexSequenceEvalSet * evalfunc;
CompMat * comp;
char * comp_file;
int gap = (12);
int ext = (2);
boolean show_raw_output = FALSE;
boolean show_label_output = FALSE;
boolean show_fancy_output = FALSE;
boolean has_outputted = FALSE;
PackAln * pal;
AlnBlock * alb;
/*
* Process command line options
* -h or -help gives us help
* -g for gap value (an int) - rely on commandline error processing
* -e for ext value (an int) - rely on commandline error processing
* -m for matrix (a char)
* -r - raw matrix output
* -l - label output
* -f - fancy output
*
*
* Use calls to commandline.h functions
*
*/
if( strip_out_boolean_argument(&argc,argv,"h") == TRUE || strip_out_boolean_argument(&argc,argv,"-help") == TRUE) {
show_help(stdout);
exit(1);
}
show_raw_output = strip_out_boolean_argument(&argc,argv,"r");
show_label_output = strip_out_boolean_argument(&argc,argv,"l");
show_fancy_output = strip_out_boolean_argument(&argc,argv,"f");
/** if all FALSE, set fancy to TRUE **/
if( show_raw_output == FALSE && show_label_output == FALSE )
show_fancy_output = TRUE;
(void) strip_out_integer_argument(&argc,argv,"g",&gap);
(void) strip_out_integer_argument(&argc,argv,"e",&ext);
comp_file = strip_out_assigned_argument(&argc,argv,"m");
if( comp_file == NULL)
comp_file = "blosum62.bla";
if( argc != 3 ) {
warn("Must have two arguments for sequence 1 and sequence 2 %d",argc);
show_help(stdout);
exit(1);
}
/*
* Read in two sequences
*/
if( (query=read_fasta_file_Sequence(argv[1])) == NULL ) {
fatal("Unable to read the sequence in file %s",argv[1]);
}
if( (target=read_fasta_file_Sequence(argv[2])) == NULL ) {
fatal("Unable to read the sequence in file %s",argv[2]);
}
/*
* Open a blosum matrix. This will be opened from WISECONFIGDIR
* or WISEPERSONALDIR if it is not present in the current directory.
*/
comp = read_Blast_file_CompMat(comp_file);
if( comp == NULL ) {
fatal("unable to read file %s",comp_file);
}
/*
* Convert sequences to ComplexSequences:
* To do this we need an protein ComplexSequenceEvalSet
*
*/
evalfunc = default_aminoacid_ComplexSequenceEvalSet();
query_cs = new_ComplexSequence(query,evalfunc);
if( query_cs == NULL ) {
fatal("Unable to make a protein complex sequence from %s",query->name);
}
target_cs = new_ComplexSequence(target,evalfunc);
if( target_cs == NULL ) {
fatal("Unable to make a protein complex sequence from %s",target->name);
}
/*
* Make an alignment. I don't care about the implementation:
* If the sequences are small enough then it should use explicit memory.
* Long sequences should use divide and conquor methods.
*
* Calling PackAln_bestmemory_ProteinSW is the answer
* This function decides on the best method considering the
* memory and changes accordingly. It frees the matrix memory
* at the end as well.
*
*/
pal = PackAln_bestmemory_ProteinSW(query_cs,target_cs,comp,-gap,-ext,NULL);
if( pal == NULL ) {
fatal("Unable to make an alignment from %s and %s",query->name,target->name);
}
/*
* ok, make other alignment forms, and be ready to show
*/
alb = convert_PackAln_to_AlnBlock_ProteinSW(pal);
/*
* show output. If multiple outputs, divide using //
*/
if( show_raw_output == TRUE ) {
show_simple_PackAln(pal,stdout);
puts("//\n");
}
if( show_label_output == TRUE ) {
show_flat_AlnBlock(alb,stdout);
}
if( show_fancy_output == TRUE ) {
write_pretty_seq_align(alb,query,target,15,50,stdout);
puts("//\n");
}
/*
* Destroy the memory.
*/
free_Sequence(query);
free_Sequence(target);
free_CompMat(comp);
free_ComplexSequence(query_cs);
free_ComplexSequence(target_cs);
free_PackAln(pal);
free_AlnBlock(alb);
return 0;
}
