1 /* -*- mode: c; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4; c-file-style: "stroustrup"; -*-
3 * This source code is NOT REALLY part of
7 * GROningen MAchine for Chemical Simulations
10 * Written by David van der Spoel, Erik Lindahl, Berk Hess, and others.
11 * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
12 * Copyright (c) 2001-2004, The GROMACS development team,
13 * check out http://www.gromacs.org for more information.
15 * This program is free software; you can redistribute it and/or
16 * modify it under the terms of the GNU General Public License
17 * as published by the Free Software Foundation; either version 2
18 * of the License, or (at your option) any later version.
20 * If you want to redistribute modifications, please consider that
21 * scientific software is very special. Version control is crucial -
22 * bugs must be traceable. We will be happy to consider code for
23 * inclusion in the official distribution, but derived work must not
24 * be called official GROMACS. Details are found in the README & COPYING
25 * files - if they are missing, get the official version at www.gromacs.org.
27 * To help us fund GROMACS development, we humbly ask that you cite
28 * the papers on the package - you can find them in the top README file.
30 * For more info, check our website at http://www.gromacs.org
32 * Author: do_multiprot was written by Ran Friedman <r.friedman@bioc.uzh.ch>
35 * Green Red Orange Magenta Azure Cyan Skyblue
44 #include "gromacs/utility/smalloc.h"
45 #include "gromacs/commandline/pargs.h"
47 #include "gromacs/fileio/pdbio.h"
48 #include "gromacs/utility/fatalerror.h"
49 #include "gromacs/fileio/xvgr.h"
50 #include "gromacs/topology/index.h"
52 #include "gromacs/fileio/tpxio.h"
55 #include "gromacs/math/vec.h"
56 #include "gromacs/fileio/confio.h"
63 static void process_multiprot_output(const char *fn, real *rmsd, int *nres, rvec rotangles,
64 rvec translation, bool bCountres, t_countres *countres)
73 mpoutput=gmx_ffopen (fn,"r");
77 fgets(line, 256, mpoutput);
78 } while (strstr(line,"Match List") == NULL);
79 fgets(line, 256, mpoutput);
81 string = strtok (line,".");
82 while (i<2 && string != NULL) {
83 string = strtok (NULL,". ");
89 while (countres[j].resnr!=res)
93 fgets(line, 256, mpoutput);
94 } while (strstr(line,"End of Match List") == NULL);
98 fgets(line, 256, mpoutput);
99 } while (strstr(line,"Trans : ") == NULL);
101 string = strtok (line," :");
102 string = strtok (NULL," ");
103 while (i<3 && string != NULL) {
104 string = strtok (NULL," ");
105 rotangles[i]=atof(string);
109 while (i<3 && string != NULL) {
110 string = strtok (NULL," ");
111 translation[i]=atof(string)/10;
115 gmx_warning("Not enough values for rotation and translation vectors in the output of multiprot");
118 rotangles[YY]=rotangles[YY]*(-1);
121 fgets(line, 256, mpoutput);
122 if (strstr(line,"RMSD : ") != NULL) {
123 string = strtok (line,":");
124 string = strtok (NULL,":");
125 (*rmsd)=atof(string)/10;
129 fgets(line,256, mpoutput);
130 if (strstr(line,"Match List") != NULL) {
131 string = strtok (line,":");
132 string = strtok (NULL,":");
133 (*nres) = atoi(string);
136 gmx_ffclose(mpoutput);
139 int main(int argc,char *argv[])
141 const char *desc[] = {
142 "[TT]do_multiprot[tt] ",
143 "reads a trajectory file and aligns it to a reference structure ",
145 "calling the multiprot program. This allows you to use a reference",
146 "structure whose sequence is different than that of the protein in the ",
147 "trajectory, since the alignment is based on the geometry, not sequence.",
148 "The output of [TT]do_multiprot[tt] includes the rmsd and the number of residues",
149 "on which it was calculated.",
151 "An aligned trajectory file is generated with the [TT]-ox[tt] option.[PAR]",
152 "With the [TT]-cr[tt] option, the number of hits in the alignment is given",
153 "per residue. This number can be between 0 and the number of frames, and",
154 "indicates the structural conservation of this residue.[PAR]",
155 "If you do not have the [TT]multiprot[tt] program, get it. [TT]do_multiprot[tt] assumes",
156 "that the [TT]multiprot[tt] executable is [TT]/usr/local/bin/multiprot[tt]. If this is ",
157 "not the case, then you should set an environment variable [BB]MULTIPROT[bb]",
158 "pointing to the [TT]multiprot[tt] executable, e.g.: [PAR]",
159 "[TT]setenv MULTIPROT /usr/MultiProtInstall/multiprot.Linux[tt][PAR]",
160 "Note that at the current implementation only binary alignment (your",
161 "molecule to a reference) is supported. In addition, note that the ",
162 "by default [TT]multiprot[tt] aligns the two proteins on their C-alpha carbons.",
163 "and that this depends on the [TT]multiprot[tt] parameters which are not dealt ",
164 "with here. Thus, the C-alpha carbons is expected to give the same "
165 "results as choosing the whole protein and will be slightly faster.[PAR]",
166 "For information about [TT]multiprot[tt], see:",
167 "http://bioinfo3d.cs.tau.ac.il/MultiProt/.[PAR]"
169 static bool bVerbose;
171 { "-v", FALSE, etBOOL, {&bVerbose},
172 "HIDDENGenerate miles of useless information" }
175 const char *bugs[] = {
176 "The program is very slow, since multiprot is run externally"
180 t_trxstatus *trxout=NULL;
181 FILE *tapein,*fo,*frc,*tmpf,*out=NULL,*fres=NULL;
183 const char *fn="2_sol.res";
186 t_atoms *atoms,ratoms,useatoms;
191 int i,j,natoms,nratoms,nframe=0,model_nr=-1;
192 int cur_res,prev_res;
194 t_countres *countres=NULL;
200 char pdbfile[32],refpdb[256],title[256],rtitle[256],filemode[5];
202 char multiprot[256],*mptr;
206 bool bTrjout,bCountres;
207 const char *TrjoutFile=NULL;
209 static rvec translation={0,0,0},rotangles={0,0,0};
210 gmx_rmpbc_t gpbc=NULL;
213 { efTRX, "-f", NULL, ffREAD },
214 { efTPS, NULL, NULL, ffREAD },
215 { efNDX, NULL, NULL, ffOPTRD },
216 { efSTX, "-r", NULL , ffREAD },
217 { efXVG, "-o", "rmss", ffWRITE },
218 { efXVG, "-rc", "rescount", ffWRITE},
219 { efXVG, "-cr", "countres", ffOPTWR},
220 { efTRX, "-ox", "aligned", ffOPTWR }
222 #define NFILE asize(fnm)
224 CopyRight(stderr,argv[0]);
225 parse_common_args(&argc,argv,PCA_CAN_TIME | PCA_CAN_VIEW | PCA_TIME_UNIT,
226 NFILE,fnm, asize(pa),pa, asize(desc),desc,
227 asize(bugs),bugs,&oenv
229 fnRef=opt2fn("-r",NFILE,fnm);
230 bTrjout = opt2bSet("-ox",NFILE,fnm);
231 bCountres= opt2bSet("-cr",NFILE,fnm);
234 TrjoutFile = opt2fn_null("-ox",NFILE,fnm);
237 read_tps_conf(ftp2fn(efTPS,NFILE,fnm),title,&top,&ePBC,&xp,NULL,box,FALSE);
238 gpbc = gmx_rmpbc_init(&top.idef,ePBC,top.atoms.nr,box);
243 get_stx_coordnum(fnRef,&nratoms);
244 init_t_atoms(&ratoms,nratoms,TRUE);
246 read_stx_conf(fnRef,rtitle,&ratoms,xr,NULL,&ePBC,rbox);
249 fprintf(stderr,"Read %d atoms\n",atoms->nr);
250 fprintf(stderr,"Read %d reference atoms\n",ratoms.nr);
253 snew(countres,ratoms.nres);
256 for (i=0;i<ratoms.nr;i++) {
258 cur_res=ratoms.atom[i].resind;
259 if (cur_res != prev_res) {
260 countres[j].resnr=cur_res;
266 get_index(atoms,ftp2fn_null(efNDX,NFILE,fnm),1,&gnx,&index,&grpnm);
269 for(i=0; (i<gnx); i++) {
270 if (atoms->atom[index[i]].resind != nr0) {
271 nr0=atoms->atom[index[i]].resind;
275 fprintf(stderr,"There are %d residues in your selected group\n",nres);
277 strcpy(pdbfile,"ddXXXXXX");
279 if ((tmpf = fopen(pdbfile,"w")) == NULL) {
280 sprintf(pdbfile,"%ctmp%cfilterXXXXXX",DIR_SEPARATOR,DIR_SEPARATOR);
282 if ((tmpf = fopen(pdbfile,"w")) == NULL) {
283 gmx_fatal(FARGS,"Can not open tmp file %s",pdbfile);
291 strcpy(refpdb,"ddXXXXXX");
293 strcat(refpdb,".pdb");
294 write_sto_conf(refpdb,rtitle,&ratoms,xr,NULL,ePBC,rbox);
297 strcpy(refpdb,fnRef);
300 if ((mptr=getenv("MULTIPROT")) == NULL) {
301 mptr="/usr/local/bin/multiprot";
303 if (!gmx_fexist(mptr)) {
304 gmx_fatal(FARGS,"MULTIPROT executable (%s) does not exist (use setenv MULTIPROT)",
307 sprintf (multiprot,"%s %s %s > /dev/null %s",
308 mptr, refpdb, pdbfile, "2> /dev/null");
311 fprintf(stderr,"multiprot cmd='%s'\n",multiprot);
313 if (!read_first_frame(oenv,&status,ftp2fn(efTRX,NFILE,fnm),&fr,TRX_READ_X))
314 gmx_fatal(FARGS,"Could not read a frame from %s",ftp2fn(efTRX,NFILE,fnm));
320 outftp=fn2ftp(TrjoutFile);
322 fprintf(stderr,"Will write %s: %s\n",ftp2ext(ftp),ftp2desc(outftp));
323 strcpy(filemode,"w");
330 trxout = open_trx(TrjoutFile,filemode);
335 /* Make atoms struct for output in GRO or PDB files */
336 /* get memory for stuff to go in pdb file */
337 init_t_atoms(&useatoms,nout,FALSE);
338 sfree(useatoms.resinfo);
339 useatoms.resinfo=atoms->resinfo;
340 for(i=0;(i<nout);i++) {
341 useatoms.atomname[i]=atoms->atomname[i];
342 useatoms.atom[i]=atoms->atom[i];
343 useatoms.nres=max(useatoms.nres,useatoms.atom[i].resind+1);
346 out=gmx_ffopen(TrjoutFile,filemode);
351 if (natoms > atoms->nr) {
352 gmx_fatal(FARGS,"\nTrajectory does not match topology!");
355 gmx_fatal(FARGS,"\nTrajectory does not match selected group!");
358 fo = xvgropen(opt2fn("-o",NFILE,fnm),"RMSD","Time (ps)","RMSD (nm)",oenv);
359 frc = xvgropen(opt2fn("-rc",NFILE,fnm),"Number of Residues in the alignment","Time (ps)","Residues",oenv);
362 t = output_env_conv_time(oenv,fr.time);
363 gmx_rmpbc(gpbc,natoms,fr.box,fr.x);
364 tapein=gmx_ffopen(pdbfile,"w");
365 write_pdbfile_indexed(tapein,NULL,atoms,fr.x,ePBC,fr.box,' ',-1,gnx,index,NULL,TRUE);
369 process_multiprot_output(fn, &rmsd, &nres2,rotangles,translation,bCountres,countres);
370 fprintf(fo,"%12.7f",t);
371 fprintf(fo," %12.7f\n",rmsd);
372 fprintf(frc,"%12.7f",t);
373 fprintf(frc,"%12d\n",nres2);
375 rotate_conf(natoms,fr.x,NULL,rotangles[XX],rotangles[YY],rotangles[ZZ]);
376 for(i=0; i<natoms; i++) {
377 rvec_inc(fr.x[i],translation);
384 write_trxframe(trxout,&fr,NULL);
389 sprintf(out_title,"Generated by do_multiprot : %s t= %g %s",
390 title,output_env_conv_time(oenv,fr.time),output_env_get_time_unit(oenv));
393 write_hconf_p(out,out_title,&useatoms,prec2ndec(fr.prec),
397 fprintf(out,"REMARK GENERATED BY DO_MULTIPROT\n");
398 sprintf(out_title,"%s t= %g %s",title,output_env_conv_time(oenv,fr.time),output_env_get_time_unit(oenv));
399 /* if reading from pdb, we want to keep the original
400 model numbering else we write the output frame
401 number plus one, because model 0 is not allowed in pdb */
402 if (ftp==efPDB && fr.step > model_nr) {
408 write_pdbfile(out,out_title,&useatoms,fr.x,ePBC,fr.box,' ',model_nr,NULL,TRUE);
411 fr.title = out_title;
412 fr.bTitle = (nframe == 0);
416 write_g96_conf(out,&fr,-1,NULL);
422 } while(read_next_frame(oenv,status,&fr));
424 fres= xvgropen(opt2fn("-cr",NFILE,fnm),"Number of frames in which the residues are aligned to","Residue","Number",oenv);
425 for (i=0;i<ratoms.nres;i++) {
426 fprintf(fres,"%10d %12d\n",countres[i].resnr,countres[i].count);
432 fprintf(stderr,"\n");
434 if (trxout != NULL) {
437 else if (out != NULL) {
440 view_all(oenv,NFILE, fnm);
445 free_t_atoms(&ratoms,TRUE);
447 if (outftp==efPDB || outftp==efGRO || outftp==efG96) {
448 free_t_atoms(&useatoms,TRUE);