nmer_repeats.cpp

来自「ncbi源码」· C++ 代码 · 共 291 行

/*
 * ===========================================================================
 * PRODUCTION $Log: nmer_repeats.cpp,v $
 * PRODUCTION Revision 1000.1  2004/06/01 20:55:10  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.6
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: nmer_repeats.cpp,v 1000.1 2004/06/01 20:55:10 gouriano Exp $
 * ===========================================================================
 *
 *                            PUBLIC DOMAIN NOTICE
 *               National Center for Biotechnology Information
 *
 *  This software/database is a "United States Government Work" under the
 *  terms of the United States Copyright Act.  It was written as part of
 *  the author's official duties as a United States Government employee and
 *  thus cannot be copyrighted.  This software/database is freely available
 *  to the public for use. The National Library of Medicine and the U.S.
 *  Government have not placed any restriction on its use or reproduction.
 *
 *  Although all reasonable efforts have been taken to ensure the accuracy
 *  and reliability of the software and data, the NLM and the U.S.
 *  Government do not and cannot warrant the performance or results that
 *  may be obtained by using this software or data. The NLM and the U.S.
 *  Government disclaim all warranties, express or implied, including
 *  warranties of performance, merchantability or fitness for any particular
 *  purpose.
 *
 *  Please cite the author in any work or product based on this material.
 *
 * ===========================================================================
 *
 * Authors:  Josh Cherry
 *
 * File Description:  gbench plugin for finding nmer repeats
 *
 */



#include <ncbi_pch.hpp>
#include "nmer_repeats.hpp"

#include <algo/sequence/find_pattern.hpp>
#include <gui/core/plugin_utils.hpp>
#include <gui/core/version.hpp>
#include <gui/dialogs/col/multi_col_dlg.hpp>
#include <gui/plugin/PluginCommandSet.hpp>
#include <gui/plugin/PluginInfo.hpp>
#include <gui/plugin/PluginReply.hpp>
#include <gui/plugin/PluginRequest.hpp>
#include <gui/plugin/PluginValueConstraint.hpp>
#include <gui/objutils/utils.hpp>
#include <objects/seqloc/Seq_interval.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>

BEGIN_NCBI_SCOPE
USING_SCOPE(objects);


CAlgoPlugin_NmerRepeats::~CAlgoPlugin_NmerRepeats()
{
}


// standard plugin announce bopilerplate
void CAlgoPlugin_NmerRepeats::GetInfo(CPluginInfo& info)
{
    info.Reset();
    
    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoPlugin_NmerRepeats", "Search/Find n-mer repeats",
                 "Find di-, tri-, and tetra-nucleotide repeats",
                 "");

    // command info
    CPluginCommandSet& cmds = info.SetCommands();
    CPluginCommand&    args = cmds.AddAlgoCommand(eAlgoCommand_run);
    args.AddArgument("locs", "Locations to evaluate",
                     CSeq_loc::GetTypeInfo(),
                     CPluginArg::TData::e_Array);
    args.SetConstraint("locs",
                       (*CPluginValueConstraint::CreateSeqMol(),
                        CSeq_inst::eMol_na,
                        CSeq_inst::eMol_dna,
                        CSeq_inst::eMol_rna));

    args.AddDefaultArgument("dinuc_min",
                            "Minimum number of dinuclotide units",
                            CPluginArg::eInteger, "5");
    args.AddDefaultArgument("trinuc_min",
                            "Minimum number of trinuclotide units",
                            CPluginArg::eInteger, "5");
    args.AddDefaultArgument("tetranuc_min",
                            "Minimum number of tetranuclotide units",
                            CPluginArg::eInteger, "5");
}


void CAlgoPlugin_NmerRepeats::RunCommand(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    _TRACE("CAlgoPlugin_NmerRepeats::Run()");

    vector<int> minima(5);
    minima[2] = args["dinuc_min"].AsInteger();
    minima[3] = args["trinuc_min"].AsInteger();
    minima[4] = args["tetranuc_min"].AsInteger();

    if ( !m_Dialog.get() ) {
        m_Dialog.reset(new CMultiColDlg());

        m_Dialog->SetWindowSize(600, 350);
        m_Dialog->SetTitle("n-mer Nucleotide Repeats");
        m_Dialog->SetLabel("Search results:");

        m_Dialog->SetColumn(0, "Sequence");
        m_Dialog->SetColumn(1, "Location");
        m_Dialog->SetColumn(2, "Position", FL_ALIGN_CENTER, 2.0);
        m_Dialog->SetColumn(3, "Repeat", FL_ALIGN_LEFT);
    }

    // clear any previous contents
    m_Dialog->SetRows(0);

    vector<TSeqPos> starts;
    vector<TSeqPos> ends;


    int row = 0;
    plugin_args::TLocList locs;
    GetArgValue(args["locs"], locs);
    ITERATE (plugin_args::TLocList, iter, locs) {
        const CSeq_loc&  loc = *iter->second;
        const IDocument& doc = *iter->first;

        // find the best ID for this bioseq
        try {
            CBioseq_Handle handle = doc.GetScope().GetBioseqHandle(loc);
            CSeqVector vec =
                handle.GetSequenceView(loc,
                                       CBioseq_Handle::eViewConstructed,
                                       CBioseq_Handle::eCoding_Iupac);

            string seq;
            vec.GetSeqData( (TSeqPos) 0, vec.size(), seq );

            string& id_str  = m_Dialog->SetCell(row, 0);
            string& loc_str = m_Dialog->SetCell(row, 1);

            const CSeq_id& best_id =
                sequence::GetId(handle, sequence::eGetId_Best);
            id_str.erase();
            best_id.GetLabel(&id_str);
            loc_str = CPluginUtils::GetLabel(loc, &doc.GetScope());

            CRef<CSeq_annot> annot(new CSeq_annot());

            string rep_unit;
            int rep_num;
            string rep_summary;

            for (unsigned int n = 2;  n <= 4;  ++n) {
                starts.clear();
                ends.clear();
                CFindPattern::FindNucNmerRepeats(seq, n, minima[n],
                                                 starts, ends);

                // preallocate rows in dialog for speed
                m_Dialog->SetRows(row + starts.size());

                for(unsigned int k = 0;  k < starts.size();  k++) {
                    rep_unit = seq.substr(starts[k], n);
                    
                    if (rep_unit.find_first_not_of(rep_unit.substr(0, 1), 1)
                        == string::npos) {
                        // skip it if it's a monomer repeat
                        continue;
                    }
                    
                    if (n == 4) {
                        // check whether it's a dimer repeat too
                        if (rep_unit[2] == rep_unit[0] &&
                            rep_unit[3] == rep_unit[1]) {
                            // if so, skip it
                            continue;
                        }
                    }

                    rep_num = (ends[k] - starts[k] + 1) / n;
                    rep_summary = "(" + rep_unit + ")" +
                        NStr::IntToString(rep_num);

                    string& pos_str = m_Dialog->SetCell(row, 2);
                    // 1-based indexing for dialog
                    pos_str = NStr::IntToString(starts[k] + 1) + " - "
                        + NStr::IntToString(ends[k] + 1);

                    m_Dialog->SetCell(row, 3) = rep_summary;
                        
                    ++row;


                    // create feature
                    CRef<CSeq_feat> feat(new CSeq_feat());

                    // set correct location
                    {{
                        CSeq_loc& floc = feat->SetLocation();
                        floc.SetInt().SetId().Assign(sequence::GetId(loc));
                        floc.SetInt().SetFrom(starts[k]);
                        floc.SetInt().SetTo(ends[k]);
                        floc.SetInt().SetStrand(eNa_strand_plus);

                        CRef<CSeq_loc> new_loc =
                            CSeqUtils::RemapChildToParent(loc, floc);
                        feat->SetLocation(*new_loc);
                    }}

                    // set feature data
                    feat->SetData().SetRegion() = "Repeat: " + rep_summary;
    
                    // save in annot
                    annot->SetData().SetFtable().push_back(feat);
                }
            }
            // add description to annot
            annot->AddName("n-mer Repeats");
            reply.AddObject(doc, *annot);
        }
        catch (CException& e) {
            string str = CPluginUtils::GetLabel(loc, &doc.GetScope());
            LOG_POST(Error << "Error processing location " << str
                     << ": " << e.what());
        }
#ifndef _DEBUG
        catch (...) {
            string str = CPluginUtils::GetLabel(loc, &doc.GetScope());
            LOG_POST(Error << "Error processing location " << str);
        }
#endif
    }


    //
    // prepare our dialog box
    //
    m_Dialog->SetRows(row);
    m_Dialog->Show();
    reply.AddAction(CPluginReplyAction::e_Add_to_document);
    reply.SetStatus(eMessageStatus_success);
}


END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: nmer_repeats.cpp,v $
 * Revision 1000.1  2004/06/01 20:55:10  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.6
 *
 * Revision 1.6  2004/05/21 22:27:46  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.5  2004/05/03 13:05:42  dicuccio
 * gui/utils --> gui/objutils where needed
 *
 * Revision 1.4  2004/03/05 17:35:37  dicuccio
 * Use sequence::GetId() instead of CSeq_id::GetStringDescr()
 *
 * Revision 1.3  2004/01/27 18:37:47  dicuccio
 * Code clean-up.  Use standard names for plugins.  Removed unnecessary #includes
 *
 * Revision 1.2  2004/01/07 15:50:37  dicuccio
 * Adjusted for API change in CPluginUtils::GetLabel().  Standardized exception
 * reporting in algorithms.
 *
 * Revision 1.1  2003/12/17 17:35:09  jcherry
 * Initial version
 *
 * ===========================================================================
 */