src/gromacs/gmxpreprocess/hizzie.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) 2013,2014,2015,2017, 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 /* This file is completely threadsafe - keep it that way! */
  38 #include "gmxpre.h"
  39
  40 #include "hizzie.h"
  41
  42 #include <stdio.h>
  43 #include <string.h>
  44
  45 #include <cmath>
  46
  47 #include "gromacs/fileio/pdbio.h"
  48 #include "gromacs/gmxpreprocess/pdb2top.h"
  49 #include "gromacs/gmxpreprocess/toputil.h"
  50 #include "gromacs/math/functions.h"
  51 #include "gromacs/math/units.h"
  52 #include "gromacs/math/vec.h"
  53 #include "gromacs/topology/block.h"
  54 #include "gromacs/utility/arraysize.h"
  55 #include "gromacs/utility/cstringutil.h"
  56 #include "gromacs/utility/fatalerror.h"
  57 #include "gromacs/utility/smalloc.h"
  58
  59 static int in_strings(char *key, int nstr, const char **str)
  60 {
  61     int j;
  62
  63     for (j = 0; (j < nstr); j++)
  64     {
  65         if (strcmp(str[j], key) == 0)
  66         {
  67             return j;
  68         }
  69     }
  70
  71     return -1;
  72 }
  73
  74 static gmx_bool hbond(rvec x[], int i, int j, real distance)
  75 {
  76     real   tol = distance*distance;
  77     rvec   tmp;
  78
  79     rvec_sub(x[i], x[j], tmp);
  80
  81     return (iprod(tmp, tmp) < tol);
  82 }
  83
  84 static void chk_allhb(t_atoms *pdba, rvec x[], t_blocka *hb,
  85                       gmx_bool donor[], gmx_bool accept[], real dist)
  86 {
  87     int i, j, k, ii, natom;
  88
  89     natom = pdba->nr;
  90     snew(hb->index, natom+1);
  91     snew(hb->a, 6*natom);
  92     hb->nr  = natom;
  93     hb->nra = 6*natom;
  94
  95     k               = ii = 0;
  96     hb->index[ii++] = 0;
  97     for (i = 0; (i < natom); i++)
  98     {
  99         if (donor[i])
 100         {
 101             for (j = i+1; (j < natom); j++)
 102             {
 103                 if ((accept[j]) && (hbond(x, i, j, dist)))
 104                 {
 105                     hb->a[k++] = j;
 106                 }
 107             }
 108         }
 109         else if (accept[i])
 110         {
 111             for (j = i+1; (j < natom); j++)
 112             {
 113                 if ((donor[j]) && (hbond(x, i, j, dist)))
 114                 {
 115                     hb->a[k++] = j;
 116                 }
 117             }
 118         }
 119         hb->index[ii++] = k;
 120     }
 121     hb->nra = k;
 122 }
 123
 124 static void pr_hbonds(FILE *fp, t_blocka *hb, t_atoms *pdba)
 125 {
 126     int i, j, k, j0, j1;
 127
 128     fprintf(fp, "Dumping all hydrogen bonds!\n");
 129     for (i = 0; (i < hb->nr); i++)
 130     {
 131         j0 = hb->index[i];
 132         j1 = hb->index[i+1];
 133         for (j = j0; (j < j1); j++)
 134         {
 135             k = hb->a[j];
 136             fprintf(fp, "%5s%4d%5s - %5s%4d%5s\n",
 137                     *pdba->resinfo[pdba->atom[i].resind].name,
 138                     pdba->resinfo[pdba->atom[i].resind].nr, *pdba->atomname[i],
 139                     *pdba->resinfo[pdba->atom[k].resind].name,
 140                     pdba->resinfo[pdba->atom[k].resind].nr, *pdba->atomname[k]);
 141         }
 142     }
 143 }
 144
 145 static gmx_bool chk_hbonds(int i, t_atoms *pdba, rvec x[],
 146                            gmx_bool ad[], gmx_bool hbond[], rvec xh,
 147                            real angle, real dist)
 148 {
 149     gmx_bool bHB;
 150     int      j, aj, ri, natom;
 151     real     d2, dist2, a;
 152     rvec     nh, oh;
 153
 154     natom = pdba->nr;
 155     bHB   = FALSE;
 156     ri    = pdba->atom[i].resind;
 157     dist2 = gmx::square(dist);
 158     for (j = 0; (j < natom); j++)
 159     {
 160         /* Check whether the other atom is a donor/acceptor and not i */
 161         if ((ad[j]) && (j != i))
 162         {
 163             /* Check whether the other atom is on the same ring as well */
 164             if ((pdba->atom[j].resind != ri) ||
 165                 ((strcmp(*pdba->atomname[j], "ND1") != 0) &&
 166                  (strcmp(*pdba->atomname[j], "NE2") != 0)))
 167             {
 168                 aj  = j;
 169                 d2  = distance2(x[i], x[j]);
 170                 rvec_sub(x[i], xh, nh);
 171                 rvec_sub(x[aj], xh, oh);
 172                 a  = RAD2DEG * acos(cos_angle(nh, oh));
 173                 if ((d2 < dist2) && (a > angle))
 174                 {
 175                     if (debug)
 176                     {
 177                         fprintf(debug,
 178                                 "HBOND between %s%d-%s and %s%d-%s is %g nm, %g deg\n",
 179                                 *pdba->resinfo[pdba->atom[i].resind].name,
 180                                 pdba->resinfo[pdba->atom[i].resind].nr, *pdba->atomname[i],
 181                                 *pdba->resinfo[pdba->atom[aj].resind].name,
 182                                 pdba->resinfo[pdba->atom[aj].resind].nr, *pdba->atomname[aj],
 183                                 std::sqrt(d2), a);
 184                     }
 185                     hbond[i] = TRUE;
 186                     bHB      = TRUE;
 187                 }
 188             }
 189         }
 190     }
 191     return bHB;
 192 }
 193
 194 static void calc_ringh(rvec xattach, rvec xb, rvec xc, rvec xh)
 195 {
 196     rvec tab, tac;
 197     real n;
 198
 199     /* Add a proton on a ring to atom attach at distance 0.1 nm */
 200     rvec_sub(xattach, xb, tab);
 201     rvec_sub(xattach, xc, tac);
 202     rvec_add(tab, tac, xh);
 203     n = 0.1/norm(xh);
 204     svmul(n, xh, xh);
 205     rvec_inc(xh, xattach);
 206 }
 207
 208 void set_histp(t_atoms *pdba, rvec *x, real angle, real dist)
 209 {
 210     static const char *prot_acc[] = {
 211         "O", "OD1", "OD2", "OE1", "OE2", "OG", "OG1", "OH", "OW"
 212     };
 213 #define NPA asize(prot_acc)
 214     static const char *prot_don[] = {
 215         "N", "NH1", "NH2", "NE", "ND1", "ND2", "NE2", "NZ", "OG", "OG1", "OH", "NE1", "OW"
 216     };
 217 #define NPD asize(prot_don)
 218
 219     gmx_bool *donor, *acceptor;
 220     gmx_bool *hbond;
 221     gmx_bool  bHDd, bHEd;
 222     rvec      xh1, xh2;
 223     int       natom;
 224     int       i, j, nd, na, hisind, type = -1;
 225     int       nd1, ne2, cg, cd2, ce1;
 226     t_blocka *hb;
 227     char     *atomnm;
 228
 229     natom = pdba->nr;
 230
 231     i = 0;
 232     while (i < natom &&
 233            gmx_strcasecmp(*pdba->resinfo[pdba->atom[i].resind].name, "HIS") != 0)
 234     {
 235         i++;
 236     }
 237     if (natom == i)
 238     {
 239         return;
 240     }
 241
 242     /* A histidine residue exists that requires automated assignment, so
 243      * doing the analysis of donors and acceptors is worthwhile. */
 244     fprintf(stderr,
 245             "Analysing hydrogen-bonding network for automated assignment of histidine\n"
 246             " protonation.");
 247
 248     snew(donor, natom);
 249     snew(acceptor, natom);
 250     snew(hbond, natom);
 251     snew(hb, 1);
 252
 253     nd = na = 0;
 254     for (j = 0; (j < natom); j++)
 255     {
 256         if (in_strings(*pdba->atomname[j], NPA, prot_acc) != -1)
 257         {
 258             acceptor[j] = TRUE;
 259             na++;
 260         }
 261         if (in_strings(*pdba->atomname[j], NPD, prot_don) != -1)
 262         {
 263             donor[j] = TRUE;
 264             nd++;
 265         }
 266     }
 267     fprintf(stderr, " %d donors and %d acceptors were found.\n", nd, na);
 268     chk_allhb(pdba, x, hb, donor, acceptor, dist);
 269     if (debug)
 270     {
 271         pr_hbonds(debug, hb, pdba);
 272     }
 273     fprintf(stderr, "There are %d hydrogen bonds\n", hb->nra);
 274
 275     /* Now do the HIS stuff */
 276     hisind = -1;
 277     while (i < natom)
 278     {
 279         if (gmx_strcasecmp(*pdba->resinfo[pdba->atom[i].resind].name, "HIS") != 0)
 280         {
 281             i++;
 282         }
 283         else
 284         {
 285             if (pdba->atom[i].resind != hisind)
 286             {
 287                 hisind = pdba->atom[i].resind;
 288
 289                 /* Find the  atoms in the ring */
 290                 nd1 = ne2 = cg = cd2 = ce1 = -1;
 291                 while (i < natom && pdba->atom[i].resind == hisind)
 292                 {
 293                     atomnm = *pdba->atomname[i];
 294                     if (strcmp(atomnm, "CD2") == 0)
 295                     {
 296                         cd2 = i;
 297                     }
 298                     else if (strcmp(atomnm, "CG") == 0)
 299                     {
 300                         cg  = i;
 301                     }
 302                     else if (strcmp(atomnm, "CE1") == 0)
 303                     {
 304                         ce1 = i;
 305                     }
 306                     else if (strcmp(atomnm, "ND1") == 0)
 307                     {
 308                         nd1 = i;
 309                     }
 310                     else if (strcmp(atomnm, "NE2") == 0)
 311                     {
 312                         ne2 = i;
 313                     }
 314
 315                     i++;
 316                 }
 317
 318                 if (!((cg == -1 ) || (cd2 == -1) || (ce1 == -1) ||
 319                       (nd1 == -1) || (ne2 == -1)))
 320                 {
 321                     calc_ringh(x[nd1], x[cg], x[ce1], xh1);
 322                     calc_ringh(x[ne2], x[ce1], x[cd2], xh2);
 323
 324                     bHDd = chk_hbonds(nd1, pdba, x, acceptor, hbond, xh1, angle, dist);
 325                     chk_hbonds(nd1, pdba, x, donor, hbond, xh1, angle, dist);
 326                     bHEd = chk_hbonds(ne2, pdba, x, acceptor, hbond, xh2, angle, dist);
 327                     chk_hbonds(ne2, pdba, x, donor, hbond, xh2, angle, dist);
 328
 329                     if (bHDd)
 330                     {
 331                         if (bHEd)
 332                         {
 333                             type = ehisH;
 334                         }
 335                         else
 336                         {
 337                             type = ehisA;
 338                         }
 339                     }
 340                     else
 341                     {
 342                         type = ehisB;
 343                     }
 344                     fprintf(stderr, "Will use %s for residue %d\n",
 345                             hh[type], pdba->resinfo[hisind].nr);
 346                 }
 347                 else
 348                 {
 349                     gmx_fatal(FARGS, "Incomplete ring in HIS%d",
 350                               pdba->resinfo[hisind].nr);
 351                 }
 352
 353                 snew(pdba->resinfo[hisind].rtp, 1);
 354                 *pdba->resinfo[hisind].rtp = gmx_strdup(hh[type]);
 355             }
 356         }
 357     }
 358     done_blocka(hb);
 359     sfree(hb);
 360     sfree(donor);
 361     sfree(acceptor);
 362     sfree(hbond);
 363 }