AlnBlock * Align_strings_ProteinSmithWaterman(char * one,char * two,CompMat * comp,int gap,int ext,DPRunImpl * dpri)
{
Sequence * one_s;
Sequence * two_s;
AlnBlock * out;
/* error check the strings? */
one_s = new_Sequence_from_strings(NULL,one);
if( one_s == NULL ) {
warn("Cannot make new sequence...\n");
return NULL;
}
two_s = new_Sequence_from_strings(NULL,two);
if( two_s == NULL ) {
warn("Cannot make new sequence...\n");
return NULL;
}
out = Align_Sequences_ProteinSmithWaterman(one_s,two_s,comp,gap,ext,dpri);
free_Sequence(one_s);
free_Sequence(two_s);
return out;
}
AlnBlock * Align_Proteins_SmithWaterman(Protein * one,Protein * two,CompMat * comp,int gap,int ext,DPRunImpl * dpri)
{
if( one == NULL || two == NULL || comp == NULL ) {
warn("Passed in NULL objects into Align_Proteins_SmithWaterman!");
return NULL;
}
return Align_Sequences_ProteinSmithWaterman(one->baseseq,two->baseseq,comp,gap,ext,dpri);
}
HitPair * HitPair_from_HSPset_sw(HSPset * set,DPRunImpl * dpri,CompMat * mat)
{
HitPair * out;
HitAln * aln;
int i;
out = HitPair_alloc_std();
out->query = hard_link_Sequence(set->hsp[0]->query);
out->target = hard_link_Sequence(set->hsp[0]->target);
aln = HitAln_alloc();
aln->alb = Align_Sequences_ProteinSmithWaterman(out->query,out->target,mat,-12,-2,NULL,dpri);
aln->raw_score = aln->alb->score;
add_HitPair(out,aln);
out->raw_score = aln->alb->score;
return out;
}
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;
}
