2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 2014,2015,2016,2018,2019,2020, by the GROMACS development team, led by
5 * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl,
6 * and including many others, as listed in the AUTHORS file in the
7 * top-level source directory and at http://www.gromacs.org.
9 * GROMACS is free software; you can redistribute it and/or
10 * modify it under the terms of the GNU Lesser General Public License
11 * as published by the Free Software Foundation; either version 2.1
12 * of the License, or (at your option) any later version.
14 * GROMACS is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 * Lesser General Public License for more details.
19 * You should have received a copy of the GNU Lesser General Public
20 * License along with GROMACS; if not, see
21 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
22 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
24 * If you want to redistribute modifications to GROMACS, please
25 * consider that scientific software is very special. Version
26 * control is crucial - bugs must be traceable. We will be happy to
27 * consider code for inclusion in the official distribution, but
28 * derived work must not be called official GROMACS. Details are found
29 * in the README & COPYING files - if they are missing, get the
30 * official version at http://www.gromacs.org.
32 * To help us fund GROMACS development, we humbly ask that you cite
33 * the research papers on the package. Check out http://www.gromacs.org.
37 * Implements gmx::analysismodules::PairDistance.
39 * \author Teemu Murtola <teemu.murtola@gmail.com>
40 * \ingroup module_trajectoryanalysis
53 #include "gromacs/analysisdata/analysisdata.h"
54 #include "gromacs/analysisdata/modules/plot.h"
55 #include "gromacs/options/basicoptions.h"
56 #include "gromacs/options/filenameoption.h"
57 #include "gromacs/options/ioptionscontainer.h"
58 #include "gromacs/selection/nbsearch.h"
59 #include "gromacs/selection/selection.h"
60 #include "gromacs/selection/selectionoption.h"
61 #include "gromacs/trajectory/trajectoryframe.h"
62 #include "gromacs/trajectoryanalysis/analysissettings.h"
63 #include "gromacs/trajectoryanalysis/topologyinformation.h"
64 #include "gromacs/utility/arrayref.h"
65 #include "gromacs/utility/exceptions.h"
66 #include "gromacs/utility/stringutil.h"
71 namespace analysismodules
77 //! \addtogroup module_trajectoryanalysis
80 //! Enum value to store the selected value for `-type`.
81 enum class DistanceType : int
88 //! Enum value to store the selected value for `-refgrouping`/`-selgrouping`.
89 enum class GroupType : int
98 //! Strings corresponding to DistanceType.
99 const EnumerationArray<DistanceType, const char*> c_distanceTypeNames = { { "min", "max" } };
100 //! Strings corresponding to GroupType.
101 const EnumerationArray<GroupType, const char*> c_groupTypeNames = { { "all", "res", "mol",
105 * Implements `gmx pairdist` trajectory analysis module.
107 class PairDistance : public TrajectoryAnalysisModule
112 void initOptions(IOptionsContainer* options, TrajectoryAnalysisSettings* settings) override;
113 void initAnalysis(const TrajectoryAnalysisSettings& settings, const TopologyInformation& top) override;
115 TrajectoryAnalysisModuleDataPointer startFrames(const AnalysisDataParallelOptions& opt,
116 const SelectionCollection& selections) override;
117 void analyzeFrame(int frnr, const t_trxframe& fr, t_pbc* pbc, TrajectoryAnalysisModuleData* pdata) override;
119 void finishAnalysis(int nframes) override;
120 void writeOutput() override;
124 * Computed distances as a function of time.
126 * There is one data set for each selection in `sel_`.
127 * Within each data set, there is one column for each distance to be
128 * computed, as explained in the `-h` text.
130 AnalysisData distances_;
133 * Reference selection to compute distances to.
135 * mappedId() identifies the group (of type `refGroupType_`) into which
136 * each position belogs.
140 * Selections to compute distances from.
142 * mappedId() identifies the group (of type `selGroupType_`) into which
143 * each position belogs.
150 DistanceType distanceType_;
151 GroupType refGroupType_;
152 GroupType selGroupType_;
154 //! Number of groups in `refSel_`.
156 //! Maximum number of pairs of groups for one selection.
158 //! Initial squared distance for distance accumulation.
160 //! Cutoff squared for use in the actual calculation.
163 //! Neighborhood search object for the pair search.
164 AnalysisNeighborhood nb_;
166 // Copy and assign disallowed by base.
169 PairDistance::PairDistance() :
171 distanceType_(DistanceType::Min),
172 refGroupType_(GroupType::All),
173 selGroupType_(GroupType::All),
179 registerAnalysisDataset(&distances_, "dist");
183 void PairDistance::initOptions(IOptionsContainer* options, TrajectoryAnalysisSettings* settings)
185 static const char* const desc[] = {
186 "[THISMODULE] calculates pairwise distances between one reference",
187 "selection (given with [TT]-ref[tt]) and one or more other selections",
188 "(given with [TT]-sel[tt]). It can calculate either the minimum",
189 "distance (the default), or the maximum distance (with",
190 "[TT]-type max[tt]). Distances to each selection provided with",
191 "[TT]-sel[tt] are computed independently.[PAR]",
192 "By default, the global minimum/maximum distance is computed.",
193 "To compute more distances (e.g., minimum distances to each residue",
194 "in [TT]-ref[tt]), use [TT]-refgrouping[tt] and/or [TT]-selgrouping[tt]",
195 "to specify how the positions within each selection should be",
197 "Computed distances are written to the file specified with [TT]-o[tt].",
198 "If there are N groups in [TT]-ref[tt] and M groups in the first",
199 "selection in [TT]-sel[tt], then the output contains N*M columns",
200 "for the first selection. The columns contain distances like this:",
201 "r1-s1, r2-s1, ..., r1-s2, r2-s2, ..., where rn is the n'th group",
202 "in [TT]-ref[tt] and sn is the n'th group in the other selection.",
203 "The distances for the second selection comes as separate columns",
204 "after the first selection, and so on. If some selections are",
205 "dynamic, only the selected positions are used in the computation",
206 "but the same number of columns is always written out. If there",
207 "are no positions contributing to some group pair, then the cutoff",
208 "value is written (see below).[PAR]",
209 "[TT]-cutoff[tt] sets a cutoff for the computed distances.",
210 "If the result would contain a distance over the cutoff, the cutoff",
211 "value is written to the output file instead. By default, no cutoff",
212 "is used, but if you are not interested in values beyond a cutoff,",
213 "or if you know that the minimum distance is smaller than a cutoff,",
214 "you should set this option to allow the tool to use grid-based",
215 "searching and be significantly faster.[PAR]",
216 "If you want to compute distances between fixed pairs,",
217 "[gmx-distance] may be a more suitable tool."
220 settings->setHelpText(desc);
222 options->addOption(FileNameOption("o")
227 .defaultBasename("dist")
228 .description("Distances as function of time"));
231 DoubleOption("cutoff").store(&cutoff_).description("Maximum distance to consider"));
232 options->addOption(EnumOption<DistanceType>("type")
233 .store(&distanceType_)
234 .enumValue(c_distanceTypeNames)
235 .description("Type of distances to calculate"));
237 EnumOption<GroupType>("refgrouping")
238 .store(&refGroupType_)
239 .enumValue(c_groupTypeNames)
240 .description("Grouping of -ref positions to compute the min/max over"));
242 EnumOption<GroupType>("selgrouping")
243 .store(&selGroupType_)
244 .enumValue(c_groupTypeNames)
245 .description("Grouping of -sel positions to compute the min/max over"));
247 options->addOption(SelectionOption("ref").store(&refSel_).required().description(
248 "Reference positions to calculate distances from"));
249 options->addOption(SelectionOption("sel").storeVector(&sel_).required().multiValue().description(
250 "Positions to calculate distances for"));
253 //! Helper function to initialize the grouping for a selection.
254 int initSelectionGroups(Selection* sel, const gmx_mtop_t* top, GroupType type)
256 e_index_t indexType = INDEX_UNKNOWN;
257 // If the selection type is INDEX_UNKNOWN (e.g. a position not associated
258 // with a set of particles), don't overwrite the selection type.
259 if (sel->type() != INDEX_UNKNOWN)
263 case GroupType::All: indexType = INDEX_ALL; break;
264 case GroupType::Residue: indexType = INDEX_RES; break;
265 case GroupType::Molecule: indexType = INDEX_MOL; break;
266 case GroupType::None: indexType = INDEX_ATOM; break;
267 case GroupType::Count: GMX_THROW(InternalError("Invalid GroupType"));
270 return sel->initOriginalIdsToGroup(top, indexType);
274 void PairDistance::initAnalysis(const TrajectoryAnalysisSettings& settings, const TopologyInformation& top)
276 refGroupCount_ = initSelectionGroups(&refSel_, top.mtop(), refGroupType_);
279 distances_.setDataSetCount(sel_.size());
280 for (size_t i = 0; i < sel_.size(); ++i)
282 const int selGroupCount = initSelectionGroups(&sel_[i], top.mtop(), selGroupType_);
283 const int columnCount = refGroupCount_ * selGroupCount;
284 maxGroupCount_ = std::max(maxGroupCount_, columnCount);
285 distances_.setColumnCount(i, columnCount);
288 if (!fnDist_.empty())
290 AnalysisDataPlotModulePointer plotm(new AnalysisDataPlotModule(settings.plotSettings()));
291 plotm->setFileName(fnDist_);
292 if (distanceType_ == DistanceType::Max)
294 plotm->setTitle("Maximum distance");
298 plotm->setTitle("Minimum distance");
300 // TODO: Figure out and add a descriptive subtitle and/or a longer
301 // title and/or better legends based on the grouping and the reference
303 plotm->setXAxisIsTime();
304 plotm->setYLabel("Distance (nm)");
305 for (size_t g = 0; g < sel_.size(); ++g)
307 plotm->appendLegend(sel_[g].name());
309 distances_.addModule(plotm);
312 nb_.setCutoff(cutoff_);
316 initialDist2_ = std::numeric_limits<real>::max();
320 initialDist2_ = cutoff_ * cutoff_;
322 if (distanceType_ == DistanceType::Max)
326 cutoff2_ = cutoff_ * cutoff_;
330 * Temporary memory for use within a single-frame calculation.
332 class PairDistanceModuleData : public TrajectoryAnalysisModuleData
336 * Reserves memory for the frame-local data.
338 PairDistanceModuleData(TrajectoryAnalysisModule* module,
339 const AnalysisDataParallelOptions& opt,
340 const SelectionCollection& selections,
342 const Selection& refSel,
344 TrajectoryAnalysisModuleData(module, opt, selections)
346 distArray_.resize(maxGroupCount);
347 countArray_.resize(maxGroupCount);
348 refCountArray_.resize(refGroupCount);
349 if (!refSel.isDynamic())
351 initRefCountArray(refSel);
355 void finish() override { finishDataHandles(); }
358 * Computes the number of positions in each group in \p refSel
359 * and stores them into `refCountArray_`.
361 void initRefCountArray(const Selection& refSel)
363 std::fill(refCountArray_.begin(), refCountArray_.end(), 0);
365 while (refPos < refSel.posCount())
367 const int refIndex = refSel.position(refPos).mappedId();
368 const int startPos = refPos;
370 while (refPos < refSel.posCount() && refSel.position(refPos).mappedId() == refIndex)
374 refCountArray_[refIndex] = refPos - startPos;
379 * Squared distance between each group
381 * One entry for each group pair for the current selection.
382 * Enough memory is allocated to fit the largest calculation selection.
383 * This is needed to support neighborhood searching, which may not
384 * return the pairs in order: for each group pair, we need to search
385 * through all the position pairs and update this array to find the
386 * minimum/maximum distance between them.
388 std::vector<real> distArray_;
390 * Number of pairs within the cutoff that have contributed to the value
393 * This is needed to identify whether there were any pairs inside the
394 * cutoff and whether there were additional pairs outside the cutoff
395 * that were not covered by the neihborhood search.
397 std::vector<int> countArray_;
399 * Number of positions within each reference group.
401 * This is used to more efficiently compute the total number of pairs
402 * (for comparison with `countArray_`), as otherwise these numbers
403 * would need to be recomputed for each selection.
405 std::vector<int> refCountArray_;
408 TrajectoryAnalysisModuleDataPointer PairDistance::startFrames(const AnalysisDataParallelOptions& opt,
409 const SelectionCollection& selections)
411 return TrajectoryAnalysisModuleDataPointer(new PairDistanceModuleData(
412 this, opt, selections, refGroupCount_, refSel_, maxGroupCount_));
415 void PairDistance::analyzeFrame(int frnr, const t_trxframe& fr, t_pbc* pbc, TrajectoryAnalysisModuleData* pdata)
417 AnalysisDataHandle dh = pdata->dataHandle(distances_);
418 const Selection& refSel = TrajectoryAnalysisModuleData::parallelSelection(refSel_);
419 const SelectionList& sel = TrajectoryAnalysisModuleData::parallelSelections(sel_);
420 PairDistanceModuleData& frameData = *static_cast<PairDistanceModuleData*>(pdata);
421 std::vector<real>& distArray = frameData.distArray_;
422 std::vector<int>& countArray = frameData.countArray_;
424 if (cutoff_ > 0.0 && refSel.isDynamic())
426 // Count the number of reference positions in each group, so that
427 // this does not need to be computed again for each selection.
428 // This is needed only if it is possible that the neighborhood search
429 // does not cover all the pairs, hence the cutoff > 0.0 check.
430 // If refSel is static, then the array contents are static as well,
431 // and it has been initialized in the constructor of the data object.
432 frameData.initRefCountArray(refSel);
434 const std::vector<int>& refCountArray = frameData.refCountArray_;
436 AnalysisNeighborhoodSearch nbsearch = nb_.initSearch(pbc, refSel);
437 dh.startFrame(frnr, fr.time);
438 for (size_t g = 0; g < sel.size(); ++g)
440 const int columnCount = distances_.columnCount(g);
441 std::fill(distArray.begin(), distArray.begin() + columnCount, initialDist2_);
442 std::fill(countArray.begin(), countArray.begin() + columnCount, 0);
444 // Accumulate the number of position pairs within the cutoff and the
445 // min/max distance for each group pair.
446 AnalysisNeighborhoodPairSearch pairSearch = nbsearch.startPairSearch(sel[g]);
447 AnalysisNeighborhoodPair pair;
448 while (pairSearch.findNextPair(&pair))
450 const SelectionPosition& refPos = refSel.position(pair.refIndex());
451 const SelectionPosition& selPos = sel[g].position(pair.testIndex());
452 const int refIndex = refPos.mappedId();
453 const int selIndex = selPos.mappedId();
454 const int index = selIndex * refGroupCount_ + refIndex;
455 const real r2 = pair.distance2();
456 if (distanceType_ == DistanceType::Min)
458 if (distArray[index] > r2)
460 distArray[index] = r2;
465 if (distArray[index] < r2)
467 distArray[index] = r2;
473 // If it is possible that positions outside the cutoff (or lack of
474 // them) affects the result, then we need to check whether there were
475 // any. This is necessary for two cases:
476 // - With max distances, if there are pairs outside the cutoff, then
477 // the computed distance should be equal to the cutoff instead of
478 // the largest distance that was found above.
479 // - With either distance type, if all pairs are outside the cutoff,
480 // then countArray must be updated so that the presence flag
481 // in the output data reflects the dynamic selection status, not
482 // whether something was inside the cutoff or not.
486 // Loop over groups in this selection (at start, selPos is always
487 // the first position in the next group).
488 while (selPos < sel[g].posCount())
490 // Count the number of positions in this group.
491 const int selIndex = sel[g].position(selPos).mappedId();
492 const int startPos = selPos;
494 while (selPos < sel[g].posCount() && sel[g].position(selPos).mappedId() == selIndex)
498 const int count = selPos - startPos;
499 // Check all group pairs that contain this group.
500 for (int i = 0; i < refGroupCount_; ++i)
502 const int index = selIndex * refGroupCount_ + i;
503 const int totalCount = refCountArray[i] * count;
504 // If there were positions outside the cutoff,
505 // update the distance if necessary and the count.
506 if (countArray[index] < totalCount)
508 if (distanceType_ == DistanceType::Max)
510 distArray[index] = cutoff2_;
512 countArray[index] = totalCount;
518 // Write the computed distances to the output data.
520 for (int i = 0; i < columnCount; ++i)
522 if (countArray[i] > 0)
524 dh.setPoint(i, std::sqrt(distArray[i]));
528 // If there are no contributing positions, write out the cutoff
530 dh.setPoint(i, cutoff_, false);
537 void PairDistance::finishAnalysis(int /*nframes*/) {}
539 void PairDistance::writeOutput() {}
545 const char PairDistanceInfo::name[] = "pairdist";
546 const char PairDistanceInfo::shortDescription[] =
547 "Calculate pairwise distances between groups of positions";
549 TrajectoryAnalysisModulePointer PairDistanceInfo::create()
551 return TrajectoryAnalysisModulePointer(new PairDistance);
554 } // namespace analysismodules