int main(int argc,char **argv)
{
int i,nseqs;
char infile[FILENAMELEN+1];
char outfile[FILENAMELEN+1];
ALN mult_aln;
OPT opt;
if(argc!=3) {
fprintf(stderr,"Usage: %s input_aln output_aln\n",argv[0]);
exit(1);
}
strcpy(infile,argv[1]);
strcpy(outfile,argv[2]);
init_options(&opt);
(*opt.alnout_opt).output_clustal=FALSE;
(*opt.alnout_opt).output_gcg=TRUE;
/* read in the sequences */
seq_input(infile,opt.explicit_type,FALSE,&mult_aln);
if(mult_aln.nseqs<=0) {
fprintf(stderr,"ERROR: No sequences in %s\n",infile);
exit(1);
}
nseqs=mult_aln.nseqs;
/* write out the sequences */
strcpy((*opt.alnout_opt).gcg_outname, outfile);
for (i=0;i<mult_aln.nseqs;i++) mult_aln.seqs[i].output_index = i;
if(!open_alignment_output(infile,opt.alnout_opt)) exit(1);
create_alignment_output(mult_aln,*opt.alnout_opt);
}
int main(int argc,char **argv)
{
sint i,j,k,n,s;
sint status;
sint result_type;
char c;
char infile[FILENAMELEN+1];
char treefile[FILENAMELEN+1];
float meanid;
FILE *tree;
sint maxres,*gapptr=NULL;
IN_TREEPTR itree;
double dscore;
double meanscore;
double **tmat;
sint window;
sint block_cutoff;
OPT opt;
if(argc!=2) {
fprintf(stdout,"Usage: %s input_aln\n",argv[0]);
exit(1);
}
strcpy(infile,argv[1]);
init_options(&opt);
/* read in the sequences */
seq_input(infile,opt.explicit_type,FALSE,&mult_aln);
if(mult_aln.nseqs<=0) {
error("No sequences in %s\n",infile);
exit(1);
}
window=8;
/* count pairwise residue percent identities */
tmat = (double **) ckalloc( (mult_aln.nseqs+1) * sizeof (double *) );
for(i=0;i<mult_aln.nseqs;i++)
tmat[i] = (double *)ckalloc( (mult_aln.nseqs+1) * sizeof (double) );
meanscore=0;
for (i=0,n=0;i<mult_aln.nseqs;i++) {
for (j=i+1;j<mult_aln.nseqs;j++) {
dscore = countid(mult_aln.seqs[i],mult_aln.seqs[j]);
tmat[j][i] = tmat[i][j] = (100.0 - dscore)/100.0;
n++;
meanscore+=dscore;
}
}
meanscore/=(float)n;
/*if(mult_aln.nseqs<100) block_cutoff=8;
else if(mult_aln.nseqs<250) block_cutoff=6;
else*/ block_cutoff=5;
if(meanscore>50) block_cutoff=50;
/* make a tree from the percent identities (used for sequence weighting) */
strcpy(treefile,infile);
strcat(treefile,".ph");
if((tree = open_explicit_file(treefile))==NULL) exit(1);
guide_tree(tree,mult_aln.seqs,mult_aln.nseqs, tmat, QUICKNJ);
itree=(IN_TREEPTR)ckalloc(sizeof(IN_TREE));
status = read_tree(treefile, mult_aln.seqs, 0, mult_aln.nseqs,itree);
for(i=0;i<mult_aln.nseqs;i++)
ckfree(tmat[i]);
ckfree(tmat);
if (status < 0) exit(1);
seq_weight = calc_seq_weights(0,mult_aln.nseqs,itree,FALSE);
free_tree(itree);
remove(treefile);
/* find the start and end positions of each sequence */
is = (sint *)ckalloc((mult_aln.nseqs+1) * sizeof(sint));
ie = (sint *)ckalloc((mult_aln.nseqs+1) * sizeof(sint));
for(s=0;s<mult_aln.nseqs;s++) {
is[s]=0;
ie[s] = mult_aln.seqs[s].len;
for (i=0; i<mult_aln.seqs[s].len; i++) {
c = mult_aln.seqs[s].data[i];
if (!isalpha(c))
is[s]++;
else
break;
}
for (i=mult_aln.seqs[s].len-1; i>=0; i--) {
c = mult_aln.seqs[s].data[i];
if (!isalpha(c))
ie[s]--;
else
break;
}
}
matrix.format=0;
maxres = get_cl_matrix(FALSE, gon250mt, gapptr, TRUE, 100, &matrix);
all_blocks(infile,window,block_cutoff);
}
void main_menu(void)
{
int catchint;
catchint = signal(SIGINT, SIG_IGN) != SIG_IGN;
if (catchint) {
if (setjmp(jmpbuf) != 0)
fprintf(stdout,"\n.. Interrupt\n");
#ifdef UNIX
if (signal(SIGINT,jumper) == BADSIG)
fprintf(stdout,"Error: signal\n");
#else
if (signal(SIGINT,SIG_DFL) == (void*)BADSIG)
fprintf(stdout,"Error: signal\n");
#endif
}
while(TRUE) {
fprintf(stdout,"\n\n\n");
fprintf(stdout," **************************************************************\n");
fprintf(stdout," ******** CLUSTAL %s Multiple Sequence Alignments ********\n",revision_level);
fprintf(stdout," **************************************************************\n");
fprintf(stdout,"\n\n");
fprintf(stdout," 1. Sequence Input From Disc\n");
fprintf(stdout," 2. Multiple Alignments\n");
fprintf(stdout," 3. Profile / Structure Alignments\n");
fprintf(stdout," 4. Phylogenetic trees\n");
fprintf(stdout,"\n");
fprintf(stdout," S. Execute a system command\n");
fprintf(stdout," H. HELP\n");
fprintf(stdout," X. EXIT (leave program)\n\n\n");
getstr("Your choice",lin1);
switch(toupper(*lin1)) {
case '1': seq_input(FALSE);
phylip_name[0]=EOS;
clustal_name[0]=EOS;
dist_name[0]=EOS;
nexus_name[0]=EOS;
break;
case '2': multiple_align_menu();
break;
case '3': profile_align_menu();
break;
case '4': phylogenetic_tree_menu();
break;
case 'S': do_system();
break;
case '?':
case 'H': get_help('1');
break;
case 'Q':
case 'X': exit(0);
break;
default: fprintf(stdout,"\n\nUnrecognised Command\n\n");
break;
}
}
}
static void phylogenetic_tree_menu(void)
{
int catchint;
catchint = signal(SIGINT, SIG_IGN) != SIG_IGN;
if (catchint) {
if (setjmp(jmpbuf) != 0)
fprintf(stdout,"\n.. Interrupt\n");
#ifdef UNIX
if (signal(SIGINT,jumper) == BADSIG)
fprintf(stdout,"Error: signal\n");
#else
if (signal(SIGINT,SIG_DFL) == (void*)BADSIG)
fprintf(stdout,"Error: signal\n");
#endif
}
while(TRUE)
{
fprintf(stdout,"\n\n\n");
fprintf(stdout,"****** PHYLOGENETIC TREE MENU ******\n");
fprintf(stdout,"\n\n");
fprintf(stdout," 1. Input an alignment\n");
fprintf(stdout," 2. Exclude positions with gaps? ");
if(tossgaps)
fprintf(stdout,"= ON\n");
else
fprintf(stdout,"= OFF\n");
fprintf(stdout," 3. Correct for multiple substitutions? ");
if(kimura)
fprintf(stdout,"= ON\n");
else
fprintf(stdout,"= OFF\n");
fprintf(stdout," 4. Draw tree now\n");
fprintf(stdout," 5. Bootstrap tree\n");
fprintf(stdout," 6. Output format options\n");
fprintf(stdout,"\n");
fprintf(stdout," S. Execute a system command\n");
fprintf(stdout," H. HELP\n");
fprintf(stdout," or press [RETURN] to go back to main menu\n\n\n");
getstr("Your choice",lin1);
if(*lin1 == EOS) return;
switch(toupper(*lin1))
{
case '1': seq_input(FALSE);
phylip_name[0]=EOS;
clustal_name[0]=EOS;
dist_name[0]=EOS;
nexus_name[0]=EOS;
break;
case '2': tossgaps ^= TRUE;
break;
case '3': kimura ^= TRUE;;
break;
case '4': phylogenetic_tree(phylip_name,clustal_name,dist_name,nexus_name,"amenu.pim");
break;
case '5': bootstrap_tree(phylip_name,clustal_name,nexus_name);
break;
case '6': tree_format_options_menu();
break;
case 'S': do_system();
break;
case '?':
case 'H': get_help('7');
break;
case 'Q':
case 'X': return;
default: fprintf(stdout,"\n\nUnrecognised Command\n\n");
break;
}
}
}
int main(int argc, char **argv)
{
FILE *ofd,*ifd;
ALN mult_aln;
OPT opt;
char infile[FILENAMELEN+1];
char outfile[FILENAMELEN+1];
int nseqs;
int i,j,l,n,ires;
int err,ix,ntot;
float min_nn,nn;
float tmp;
Boolean eof,found;
if(argc!=3 && argc!=7 && argc!=8) {
usage(argv[0]);
return 0;
}
strcpy(infile,argv[1]);
strcpy(outfile,argv[2]);
/* open the matrix file */
verbose=FALSE;
if(argc==3) {
get_default_matrix();
go=0.0;
ge=0.1;
egap=0.0;
}
else {
if(argc==8) verbose=TRUE;
if((ifd=fopen(argv[3],"r"))==NULL) {
fprintf(stderr,"Cannot open matrix file [%s]",argv[3]);
return 0;
}
err=readmatrix(ifd);
if(err<=0) {
fprintf(stderr,"Error: bad matrix in %s\n",argv[3]);
return 0;
}
go=atof(argv[4]);
ge=atof(argv[5]);
egap=atof(argv[6]);
}
init_options(&opt);
(*opt.alnout_opt).output_clustal=FALSE;
(*opt.alnout_opt).output_relacs=TRUE;
/* read in the sequences */
seq_input(infile,opt.explicit_type,FALSE,&mult_aln);
if(mult_aln.nseqs<=0) {
error("No sequences in %s\n",infile);
exit(1);
}
nseqs=mult_aln.nseqs;
/* remove the gaps */
seqlength=0;
useqlen_array=(int *)ckalloc((nseqs+1)*sizeof(int));
for(i=0;i<nseqs;i++) {
if(mult_aln.seqs[i].len>seqlength) seqlength=mult_aln.seqs[i].len;
l=0;
for(j=0;j<mult_aln.seqs[i].len;j++)
if(isalpha(mult_aln.seqs[i].data[j])) {
l++;
}
useqlen_array[i]=l;
}
maxlen=0;
for(i=0;i<nseqs;i++)
if(useqlen_array[i]>maxlen) maxlen=useqlen_array[i];
minlen=10000;
for(i=0;i<nseqs;i++)
if(useqlen_array[i]<minlen) minlen=useqlen_array[i];
/* remove any column score data that already exists in the input file */
/*if (mult_aln.ncol_scores==1)
ckfree(mult_aln.col_score[0].data);*/
ix=mult_aln.ncol_scores;
mult_aln.col_score[ix].data=(sint *)ckalloc((seqlength+1)*sizeof(sint));
mult_aln.ncol_scores=ix+1;
/* calculate some simple statistics */
pcid=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
pcid[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
for(i=0;i<nseqs;i++) {
for(j=i+1;j<nseqs;j++) {
pcid[j][i]=pcid[i][j]=pcidentity(mult_aln,i,j);
}
}
/* find the nearest neighbor for each sequence */
min_nn=1.0;
for(i=0;i<nseqs;i++) {
nn=0.0;
for(j=0;j<nseqs;j++) {
if(i!=j && pcid[i][j]>nn) nn=pcid[i][j];
}
if(nn<min_nn) min_nn=nn;
}
seqweight=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
seqweight[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
for(i=0;i<nseqs;i++)
for(j=i;j<nseqs;j++) {
seqweight[j][i]=seqweight[i][j]=1.0-pcid[i][j];
}
fragment=(Boolean *)ckalloc((nseqs+1)*sizeof(Boolean));
/* calculate pairwise alignment scores using k-tuple scores */
qpw_id=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
qpw_id[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
for(i=0,n=0;i<nseqs;i++)
for(j=i+1;j<nseqs;j++) {
qpw_id[i][j]=100.0*pcid[i][j];
if(qpw_id[i][j]<60) {
qpw_id[i][j]=show_pair(mult_aln,i,j);
}
if(qpw_id[i][j]>40) {
tmp=(float)useqlen_array[i]/(float)useqlen_array[j];
if(tmp<0.8) fragment[i]=TRUE;
else if(tmp>1.25) fragment[j]=TRUE;
}
n++;
}
/*if(verbose)
for(i=0;i<nseqs;i++)
if(fragment[i]) fprintf(stdout,"%s fragment %s\n",argv[1],names[i]);*/
/* calculate sequence groups and keep first sequence in each group for processing */
use_seq=(int *)ckalloc((nseqs+1)*sizeof(int));
for(i=0;i<nseqs;i++)
use_seq[i]=2;
query=0;
seqgroup=(int *)ckalloc((nseqs+1)*sizeof(int));
groupseed=(int *)ckalloc((nseqs+1)*sizeof(int));
calc_groups(query,0.7,nseqs,pcid,seqgroup,groupseed);
if(ngroups<=0) {
fprintf(stderr,"Error: problem with sequence grouping\n");
exit(1);
}
for(j=0;j<nseqs;j++) use_seq[j]=(-1);
for(i=0;i<ngroups;i++) {
j=groupseed[i];
use_seq[j]=2;
}
for(i=0;i<nseqs;i++)
ckfree(pcid[i]);
ckfree(pcid);
ckfree(groupseed);
ckfree(seqgroup);
qpw=(float *)ckalloc((ngroups*ngroups+1)*sizeof(float));
for(i=0,n=0;i<nseqs;i++)
if(use_seq[i]>1)
for(j=i+1;j<nseqs;j++)
if(use_seq[j]>1)
qpw[n++]=qpw_id[i][j];
for(i=0;i<nseqs;i++)
ckfree(qpw_id[i]);
ckfree(qpw_id);
/* sort the pairwise k-tuple scores into ascending order */
sort_scores(qpw,0,n-1);
/* calculate the scores for the gaps */
gop=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
gop[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
gep=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
gep[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
egp=(float **)ckalloc((nseqs+1)*sizeof(float *));
for(i=0;i<nseqs;i++)
egp[i]=(float *)ckalloc((nseqs+1)*sizeof(float));
for(i=0;i<nseqs;i++)
if(use_seq[i]>1)
for(j=i+1;j<nseqs;j++)
if(use_seq[j]>1)
score_gaps(mult_aln,i,j);
calc_md(&mult_aln,ix);
for(i=0,ntot=0;i<nseqs;i++)
if(use_seq[i]>1) {
ntot++;
}
tmp=set_mdcutoff(ntot,q1);
normd_rs/=tmp;
tmp=1.0;
norm_md/=tmp;
mult_aln.alnscore=norm_md;
mult_aln.validalnscore=TRUE;
if(!verbose) {
fprintf(stdout,"%s\t%.3f\n",argv[1],norm_md);
/*fprintf(stdout,"%.3f\n",norm_md);*/
} else {
/*fprintf(stdout,"%s %.3f %.3f %.3f %.3f %.3f %.3f %d %d %d\n",
argv[1],norm_md,normd_rs,col,max_colscore,gap_extscore*0.1,q1,nseqs,minlen,maxlen);*/
fprintf(stdout,"FILE %s\n",argv[1]);
fprintf(stdout,"norMD %.3f\n",norm_md);
/*fprintf(stdout,"norMD_of %.3f\n",norm_md);
fprintf(stdout,"norMD_rs %.3f\n",normd_rs);*/
fprintf(stdout,"NSEQS %d\n",nseqs);
fprintf(stdout,"MD %.3f\n",col);
fprintf(stdout,"maxMD %.3f\n",max_colscore);
fprintf(stdout,"GOP %.3f\n",gap_openscore);
fprintf(stdout,"GEP %.3f\n",gap_extscore);
fprintf(stdout,"LQR %.3f\n",q1);
}
for(i=0;i<nseqs;i++)
ckfree(gop[i]);
ckfree(gop);
for(i=0;i<nseqs;i++)
ckfree(gep[i]);
ckfree(gep);
for(i=0;i<nseqs;i++)
ckfree(egp[i]);
ckfree(egp);
for(i=0;i<nseqs;i++)
ckfree(seqweight[i]);
ckfree(seqweight);
ckfree(qpw);
ckfree(use_seq);
ckfree(fragment);
ckfree(useqlen_array);
/* write out the sequences */
strcpy(opt.alnout_opt->relacs_outname,outfile);
if(!open_alignment_output(outfile,opt.alnout_opt)) exit(1);
create_alignment_output(mult_aln,*opt.alnout_opt);
return 0;
}
