src/gromacs/gmxpreprocess/specbond.cpp

   1 /*
   2  * This file is part of the GROMACS molecular simulation package.
   3  *
   4  * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
   5  * Copyright (c) 2001-2004, The GROMACS development team.
   6  * Copyright (c) 2013,2014,2015,2016,2017, by the GROMACS development team, led by
   7  * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
   8  * and including many others, as listed in the AUTHORS file in the
   9  * top-level source directory and at http://www.gromacs.org.
  10  *
  11  * GROMACS is free software; you can redistribute it and/or
  12  * modify it under the terms of the GNU Lesser General Public License
  13  * as published by the Free Software Foundation; either version 2.1
  14  * of the License, or (at your option) any later version.
  15  *
  16  * GROMACS is distributed in the hope that it will be useful,
  17  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  18  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  19  * Lesser General Public License for more details.
  20  *
  21  * You should have received a copy of the GNU Lesser General Public
  22  * License along with GROMACS; if not, see
  23  * http://www.gnu.org/licenses, or write to the Free Software Foundation,
  24  * Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA.
  25  *
  26  * If you want to redistribute modifications to GROMACS, please
  27  * consider that scientific software is very special. Version
  28  * control is crucial - bugs must be traceable. We will be happy to
  29  * consider code for inclusion in the official distribution, but
  30  * derived work must not be called official GROMACS. Details are found
  31  * in the README & COPYING files - if they are missing, get the
  32  * official version at http://www.gromacs.org.
  33  *
  34  * To help us fund GROMACS development, we humbly ask that you cite
  35  * the research papers on the package. Check out http://www.gromacs.org.
  36  */
  37 #include "gmxpre.h"
  38
  39 #include "specbond.h"
  40
  41 #include <ctype.h>
  42 #include <string.h>
  43
  44 #include <cmath>
  45
  46 #include <algorithm>
  47
  48 #include "gromacs/fileio/pdbio.h"
  49 #include "gromacs/gmxpreprocess/pdb2top.h"
  50 #include "gromacs/math/vec.h"
  51 #include "gromacs/utility/cstringutil.h"
  52 #include "gromacs/utility/fatalerror.h"
  53 #include "gromacs/utility/smalloc.h"
  54 #include "gromacs/utility/strdb.h"
  55
  56 gmx_bool yesno(void)
  57 {
  58     char c;
  59
  60     do
  61     {
  62         c = toupper(fgetc(stdin));
  63     }
  64     while ((c != 'Y') && (c != 'N'));
  65
  66     return (c == 'Y');
  67 }
  68
  69 t_specbond *get_specbonds(int *nspecbond)
  70 {
  71     const char  *sbfile = "specbond.dat";
  72
  73     t_specbond  *sb = nullptr;
  74     char         r1buf[32], r2buf[32], a1buf[32], a2buf[32], nr1buf[32], nr2buf[32];
  75     double       length;
  76     int          nb1, nb2;
  77     char       **lines;
  78     int          nlines, i, n;
  79
  80     nlines = get_lines(sbfile, &lines);
  81     if (nlines > 0)
  82     {
  83         snew(sb, nlines);
  84     }
  85
  86     n = 0;
  87     for (i = 0; (i < nlines); i++)
  88     {
  89         if (sscanf(lines[i], "%s%s%d%s%s%d%lf%s%s",
  90                    r1buf, a1buf, &nb1, r2buf, a2buf, &nb2, &length, nr1buf, nr2buf) != 9)
  91         {
  92             fprintf(stderr, "Invalid line '%s' in %s\n", lines[i], sbfile);
  93         }
  94         else
  95         {
  96             sb[n].res1    = gmx_strdup(r1buf);
  97             sb[n].res2    = gmx_strdup(r2buf);
  98             sb[n].newres1 = gmx_strdup(nr1buf);
  99             sb[n].newres2 = gmx_strdup(nr2buf);
 100             sb[n].atom1   = gmx_strdup(a1buf);
 101             sb[n].atom2   = gmx_strdup(a2buf);
 102             sb[n].nbond1  = nb1;
 103             sb[n].nbond2  = nb2;
 104             sb[n].length  = length;
 105             n++;
 106         }
 107         sfree(lines[i]);
 108     }
 109     if (nlines > 0)
 110     {
 111         sfree(lines);
 112     }
 113     fprintf(stderr, "%d out of %d lines of %s converted successfully\n",
 114             n, nlines, sbfile);
 115
 116     *nspecbond = n;
 117
 118     return sb;
 119 }
 120
 121 void done_specbonds(int nsb, t_specbond sb[])
 122 {
 123     int i;
 124
 125     for (i = 0; (i < nsb); i++)
 126     {
 127         sfree(sb[i].res1);
 128         sfree(sb[i].res2);
 129         sfree(sb[i].atom1);
 130         sfree(sb[i].atom2);
 131         sfree(sb[i].newres1);
 132         sfree(sb[i].newres2);
 133     }
 134 }
 135
 136 static gmx_bool is_special(int nsb, t_specbond sb[], char *res, char *atom)
 137 {
 138     int i;
 139
 140     for (i = 0; (i < nsb); i++)
 141     {
 142         if (((strncmp(sb[i].res1, res, 3) == 0) &&
 143              (gmx_strcasecmp(sb[i].atom1, atom) == 0)) ||
 144             ((strncmp(sb[i].res2, res, 3) == 0) &&
 145              (gmx_strcasecmp(sb[i].atom2, atom) == 0)))
 146         {
 147             return TRUE;
 148         }
 149     }
 150     return FALSE;
 151 }
 152
 153 static gmx_bool is_bond(int nsb, t_specbond sb[], t_atoms *pdba, int a1, int a2,
 154                         real d, int *index_sb, gmx_bool *bSwap)
 155 {
 156     int   i;
 157     char *at1, *at2, *res1, *res2;
 158
 159     at1  = *pdba->atomname[a1];
 160     at2  = *pdba->atomname[a2];
 161     res1 = *pdba->resinfo[pdba->atom[a1].resind].name;
 162     res2 = *pdba->resinfo[pdba->atom[a2].resind].name;
 163
 164     if (debug)
 165     {
 166         fprintf(stderr, "Checking %s-%d %s-%d and %s-%d %s-%d: %g ",
 167                 res1, pdba->resinfo[pdba->atom[a1].resind].nr, at1, a1+1,
 168                 res2, pdba->resinfo[pdba->atom[a2].resind].nr, at2, a2+1, d);
 169     }
 170
 171     for (i = 0; (i < nsb); i++)
 172     {
 173         *index_sb = i;
 174         if (((strncmp(sb[i].res1, res1, 3) == 0)  &&
 175              (gmx_strcasecmp(sb[i].atom1, at1) == 0) &&
 176              (strncmp(sb[i].res2, res2, 3) == 0)  &&
 177              (gmx_strcasecmp(sb[i].atom2, at2) == 0)))
 178         {
 179             *bSwap = FALSE;
 180             if ((0.9*sb[i].length < d) && (1.1*sb[i].length > d))
 181             {
 182                 if (debug)
 183                 {
 184                     fprintf(stderr, "%g\n", sb[i].length);
 185                 }
 186                 return TRUE;
 187             }
 188         }
 189         if (((strncmp(sb[i].res1, res2, 3) == 0)  &&
 190              (gmx_strcasecmp(sb[i].atom1, at2) == 0) &&
 191              (strncmp(sb[i].res2, res1, 3) == 0)  &&
 192              (gmx_strcasecmp(sb[i].atom2, at1) == 0)))
 193         {
 194             *bSwap = TRUE;
 195             if ((0.9*sb[i].length < d) && (1.1*sb[i].length > d))
 196             {
 197                 if (debug)
 198                 {
 199                     fprintf(stderr, "%g\n", sb[i].length);
 200                 }
 201                 return TRUE;
 202             }
 203         }
 204     }
 205     if (debug)
 206     {
 207         fprintf(stderr, "\n");
 208     }
 209     return FALSE;
 210 }
 211
 212 static void rename_1res(t_atoms *pdba, int resind, char *newres, gmx_bool bVerbose)
 213 {
 214     if (bVerbose)
 215     {
 216         printf("Using rtp entry %s for %s %d\n",
 217                newres,
 218                *pdba->resinfo[resind].name,
 219                pdba->resinfo[resind].nr);
 220     }
 221     /* this used to free *resname, which messes up the symtab! */
 222     snew(pdba->resinfo[resind].rtp, 1);
 223     *pdba->resinfo[resind].rtp = gmx_strdup(newres);
 224 }
 225
 226 int mk_specbonds(t_atoms *pdba, rvec x[], gmx_bool bInteractive,
 227                  t_ssbond **specbonds, gmx_bool bVerbose)
 228 {
 229     t_specbond *sb    = nullptr;
 230     t_ssbond   *bonds = nullptr;
 231     int         nsb;
 232     int         nspec, nbonds;
 233     int        *specp, *sgp;
 234     gmx_bool    bDoit, bSwap;
 235     int         i, j, b, e, e2;
 236     int         ai, aj, index_sb;
 237     real      **d;
 238     char        buf[10];
 239
 240     nbonds = 0;
 241     sb     = get_specbonds(&nsb);
 242
 243     if (nsb > 0)
 244     {
 245         snew(specp, pdba->nr);
 246         snew(sgp, pdba->nr);
 247
 248         nspec = 0;
 249         for (i = 0; (i < pdba->nr); i++)
 250         {
 251             /* Check if this atom is special and if it is not a double atom
 252              * in the input that still needs to be removed.
 253              */
 254             if (is_special(nsb, sb, *pdba->resinfo[pdba->atom[i].resind].name,
 255                            *pdba->atomname[i]) &&
 256                 !(nspec > 0 &&
 257                   pdba->atom[sgp[nspec-1]].resind == pdba->atom[i].resind &&
 258                   gmx_strcasecmp(*pdba->atomname[sgp[nspec-1]],
 259                                  *pdba->atomname[i]) == 0))
 260             {
 261                 specp[nspec] = pdba->atom[i].resind;
 262                 sgp[nspec]   = i;
 263                 nspec++;
 264             }
 265         }
 266         /* distance matrix d[nspec][nspec] */
 267         snew(d, nspec);
 268         for (i = 0; (i < nspec); i++)
 269         {
 270             snew(d[i], nspec);
 271         }
 272
 273         for (i = 0; (i < nspec); i++)
 274         {
 275             ai = sgp[i];
 276             for (j = 0; (j < nspec); j++)
 277             {
 278                 aj      = sgp[j];
 279                 d[i][j] = std::sqrt(distance2(x[ai], x[aj]));
 280             }
 281         }
 282         if (nspec > 1)
 283         {
 284 #define MAXCOL 7
 285             fprintf(stderr, "Special Atom Distance matrix:\n");
 286             for (b = 0; (b < nspec); b += MAXCOL)
 287             {
 288                 /* print resname/number column headings */
 289                 fprintf(stderr, "%8s%8s", "", "");
 290                 e = std::min(b+MAXCOL, nspec-1);
 291                 for (i = b; (i < e); i++)
 292                 {
 293                     sprintf(buf, "%s%d", *pdba->resinfo[pdba->atom[sgp[i]].resind].name,
 294                             pdba->resinfo[specp[i]].nr);
 295                     fprintf(stderr, "%8s", buf);
 296                 }
 297                 fprintf(stderr, "\n");
 298                 /* print atomname/number column headings */
 299                 fprintf(stderr, "%8s%8s", "", "");
 300                 e = std::min(b+MAXCOL, nspec-1);
 301                 for (i = b; (i < e); i++)
 302                 {
 303                     sprintf(buf, "%s%d", *pdba->atomname[sgp[i]], sgp[i]+1);
 304                     fprintf(stderr, "%8s", buf);
 305                 }
 306                 fprintf(stderr, "\n");
 307                 /* print matrix */
 308                 e = std::min(b+MAXCOL, nspec);
 309                 for (i = b+1; (i < nspec); i++)
 310                 {
 311                     sprintf(buf, "%s%d", *pdba->resinfo[pdba->atom[sgp[i]].resind].name,
 312                             pdba->resinfo[specp[i]].nr);
 313                     fprintf(stderr, "%8s", buf);
 314                     sprintf(buf, "%s%d", *pdba->atomname[sgp[i]], sgp[i]+1);
 315                     fprintf(stderr, "%8s", buf);
 316                     e2 = std::min(i, e);
 317                     for (j = b; (j < e2); j++)
 318                     {
 319                         fprintf(stderr, " %7.3f", d[i][j]);
 320                     }
 321                     fprintf(stderr, "\n");
 322                 }
 323             }
 324         }
 325
 326         snew(bonds, nspec);
 327
 328         for (i = 0; (i < nspec); i++)
 329         {
 330             ai = sgp[i];
 331             for (j = i+1; (j < nspec); j++)
 332             {
 333                 aj = sgp[j];
 334                 /* Ensure creation of at most nspec special bonds to avoid overflowing bonds[] */
 335                 if (nbonds < nspec && is_bond(nsb, sb, pdba, ai, aj, d[i][j], &index_sb, &bSwap))
 336                 {
 337                     fprintf(stderr, "%s %s-%d %s-%d and %s-%d %s-%d%s",
 338                             bInteractive ? "Link" : "Linking",
 339                             *pdba->resinfo[pdba->atom[ai].resind].name,
 340                             pdba->resinfo[specp[i]].nr,
 341                             *pdba->atomname[ai], ai+1,
 342                             *pdba->resinfo[pdba->atom[aj].resind].name,
 343                             pdba->resinfo[specp[j]].nr,
 344                             *pdba->atomname[aj], aj+1,
 345                             bInteractive ? " (y/n) ?" : "...\n");
 346                     bDoit = bInteractive ? yesno() : TRUE;
 347
 348                     if (bDoit)
 349                     {
 350                         /* Store the residue numbers in the bonds array */
 351                         bonds[nbonds].res1 = specp[i];
 352                         bonds[nbonds].res2 = specp[j];
 353                         bonds[nbonds].a1   = gmx_strdup(*pdba->atomname[ai]);
 354                         bonds[nbonds].a2   = gmx_strdup(*pdba->atomname[aj]);
 355                         /* rename residues */
 356                         if (bSwap)
 357                         {
 358                             rename_1res(pdba, specp[i], sb[index_sb].newres2, bVerbose);
 359                             rename_1res(pdba, specp[j], sb[index_sb].newres1, bVerbose);
 360                         }
 361                         else
 362                         {
 363                             rename_1res(pdba, specp[i], sb[index_sb].newres1, bVerbose);
 364                             rename_1res(pdba, specp[j], sb[index_sb].newres2, bVerbose);
 365                         }
 366                         nbonds++;
 367                     }
 368                 }
 369             }
 370         }
 371
 372         for (i = 0; (i < nspec); i++)
 373         {
 374             sfree(d[i]);
 375         }
 376         sfree(d);
 377         sfree(sgp);
 378         sfree(specp);
 379
 380         done_specbonds(nsb, sb);
 381         sfree(sb);
 382     }
 383
 384     *specbonds = bonds;
 385
 386     return nbonds;
 387 }