nmer.cpp

来自「ncbi源码」· C++ 代码 · 共 278 行

/*
 * ===========================================================================
 * PRODUCTION $Log: nmer.cpp,v $
 * PRODUCTION Revision 1000.5  2004/06/01 20:55:08  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.24
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: nmer.cpp,v 1000.5 2004/06/01 20:55:08 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:
 *   CAlgoPlugin_Nmer -- wraps the algorithm to calculate n-mer frequencies
 */

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

#include <algo/sequence/nuc_prop.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 <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>


BEGIN_NCBI_SCOPE
USING_SCOPE(objects);



CAlgoPlugin_Nmer::~CAlgoPlugin_Nmer()
{
}


// standard plugin announcement boilerplate
void CAlgoPlugin_Nmer::GetInfo(CPluginInfo& info)
{
    info.Reset();

    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoPlugin_Nmer", "Composition/n-mer Frequencies",
                 "Determine n-mer frequencies",
                 "");

    // 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.AddArgument("nmer_size", "Size of n-mers",
                     CPluginArg::eInteger);
}


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

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

        m_Dialog->SetTitle("n-mer Frequencies");
        m_Dialog->SetLabel("Results of n-mer frequencies are as follows:");
        m_Dialog->SetColumn(0, "Sequence");
        m_Dialog->SetColumn(1, "Location", FL_ALIGN_LEFT, 2.0f);
        m_Dialog->SetColumn(2, "n-mer");
        m_Dialog->SetColumn(3, "Count", FL_ALIGN_CENTER);
    }

    int nmer_size = args["nmer_size"].AsInteger();  // the 'n' in nmer

    //
    // first, evaluate whole sequences
    //
    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);
            vector<int> table;
            CNucProp::CountNmers(vec, nmer_size, table);

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

            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());

            // preallocate rows in table for speed
            int row_count = 0;
            for(unsigned int k = 0;  k < table.size();  k++) {
                if (table[k] != 0) {
                    row_count++;
                }
            }
            m_Dialog->SetRows(row_count);

            for(unsigned int k = 0;  k < table.size();  k++) {
                if (table[k] == 0) {
                    continue;
                }

                CNucProp::Int2Nmer(k, nmer_size, m_Dialog->SetCell(row, 2));
                m_Dialog->SetCell(row, 3) = NStr::IntToString(table[k]);
                ++row;
            }
        }
        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->Show();
    reply.SetStatus(eMessageStatus_success);
}




END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: nmer.cpp,v $
 * Revision 1000.5  2004/06/01 20:55:08  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.24
 *
 * Revision 1.24  2004/05/21 22:27:46  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.23  2004/03/05 17:35:37  dicuccio
 * Use sequence::GetId() instead of CSeq_id::GetStringDescr()
 *
 * Revision 1.22  2004/01/27 18:37:46  dicuccio
 * Code clean-up.  Use standard names for plugins.  Removed unnecessary #includes
 *
 * Revision 1.21  2004/01/07 15:50:36  dicuccio
 * Adjusted for API change in CPluginUtils::GetLabel().  Standardized exception
 * reporting in algorithms.
 *
 * Revision 1.20  2003/11/24 15:45:26  dicuccio
 * Renamed CVersion to CPluginVersion
 *
 * Revision 1.19  2003/11/06 20:12:12  dicuccio
 * Cleaned up handling of USING_SCOPE - removed from all headers
 *
 * Revision 1.18  2003/11/04 17:49:23  dicuccio
 * Changed calling parameters for plugins - pass CPluginMessage instead of paired
 * CPluginCommand/CPluginReply
 *
 * Revision 1.17  2003/10/07 13:47:00  dicuccio
 * Renamed CPluginURL* to CPluginValue*
 *
 * Revision 1.16  2003/09/04 14:05:24  dicuccio
 * Use IDocument instead of CDocument
 *
 * Revision 1.15  2003/09/03 14:46:53  rsmith
 * change namespace name from args to plugin_args to avoid clashes with variable names.
 *
 * Revision 1.14  2003/08/21 12:03:07  dicuccio
 * Make use of new typedef in plugin_utils.hpp for argument values.
 *
 * Revision 1.13  2003/08/18 15:05:11  jcherry
 * Reflect changes in nuc_prop.?pp.  Pre-allocate dialog rows for speed.
 *
 * Revision 1.12  2003/08/05 17:03:55  dicuccio
 * Made multi-column output dialog a member variable - allows non-modal operation
 *
 * Revision 1.11  2003/07/28 11:51:48  dicuccio
 * Rewrote CTablePanel<> to be more flexible and better contained.  Added standard
 * multicolumn list dialog.  Deprecated use of COutputm_Dialog->
 *
 * Revision 1.10  2003/07/22 15:32:16  dicuccio
 * Changed to make use of new API in plugin_utils.hpp - GetArgValue()
 *
 * Revision 1.9  2003/07/21 19:32:53  dicuccio
 * Added constraints based on molecule type
 *
 * Revision 1.8  2003/07/14 11:12:57  shomrat
 * Plugin messageing system related changes
 *
 * Revision 1.7  2003/07/01 15:08:41  jcherry
 * Moved a bunch of stuff into CNucProp and CProtProp
 * Put these in c++/{src,include}/algo/sequence
 *
 * Revision 1.6  2003/06/26 15:33:40  dicuccio
 * Moved GetURLValue() from PluginURL.hpp to plugin_utils.hpp.  Fixed compilation
 * errors relating to missing #includes
 *
 * Revision 1.5  2003/06/25 17:02:57  dicuccio
 * Split CPluginHandle into a handle (pointer-to-implementation) and
 * implementation file.  Lots of #include file clean-ups.
 *
 * Revision 1.4  2003/06/20 14:52:36  dicuccio
 * Revised plugin registration - moved GetInfo() into the plugin handler
 *
 * Revision 1.3  2003/06/10 14:45:22  jcherry
 * Moved actual algorithms for n-mers to a separate class,
 * implemented in count_nmers.?pp
 *
 * Revision 1.2  2003/06/09 19:14:39  jcherry
 * Retabbified, etc., to comply with conventions
 *
 * Revision 1.1  2003/06/09 17:30:10  jcherry
 * Initial version
 *
 * ===========================================================================
 */