predict_signal_seq.cpp

来自「ncbi源码」· C++ 代码 · 共 296 行

/*
 * ===========================================================================
 * PRODUCTION $Log: predict_signal_seq.cpp,v $
 * PRODUCTION Revision 1000.5  2004/06/01 20:55:33  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.14
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: predict_signal_seq.cpp,v 1000.5 2004/06/01 20:55:33 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 predicting protein signal sequences
 *
 */


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

#include <algo/sequence/signal_seq.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/PluginRequest.hpp>
#include <gui/plugin/PluginValueConstraint.hpp>
#include <gui/objutils/utils.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>


BEGIN_NCBI_SCOPE
USING_SCOPE(objects);



CAlgoPlugin_PredictSignalSeq::~CAlgoPlugin_PredictSignalSeq()
{
}


// standard plugin announce bopilerplate
void CAlgoPlugin_PredictSignalSeq::GetInfo(CPluginInfo& info)
{
    info.Reset();
    
    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoPlugin_PredictSignalSeq",
                 "Protein analysis/Predict secretory signal sequence",
                 "Predict N-terminal export signal sequence "
                 "from a protein sequence",
                 "");

    // 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_aa));

    args.AddDefaultArgument("max_pos", "Maximum position of cleavage site",
                            CPluginArg::eInteger, "40");

    args.AddDefaultArgument("domain", "Phylogenetic domain",
                            CPluginArg::eString, "Eukaryotic");
    args.SetConstraint("domain", (*CPluginValueConstraint::CreateSet(),
                                "Eukaryotic",
                                  "Bacterial"));
}


void CAlgoPlugin_PredictSignalSeq::RunCommand(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    _TRACE("CAlgoPlugin_PredictSignalSeq::RunCommand()");
    
    if ( !m_Dialog.get() ) {
        m_Dialog.reset(new CMultiColDlg());

        m_Dialog->SetWindowSize(800, 450);
        m_Dialog->SetTitle("Signal Sequence Prediction");
        
        m_Dialog->SetColumn(0, "Sequence", FL_ALIGN_LEFT, 0.5f);
        m_Dialog->SetColumn(1, "Location", FL_ALIGN_LEFT, 0.5f);
        m_Dialog->SetColumn(2, "Sig. Seq. Predicted?",
                            FL_ALIGN_CENTER, 0.4f);
        m_Dialog->SetColumn(3, "Max. Score", FL_ALIGN_CENTER, 0.5f);
        m_Dialog->SetColumn(4, "Cut After", FL_ALIGN_CENTER, 0.5f);
    }

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

    int row = 0;
    plugin_args::TLocList locs;
    GetArgValue(args["locs"], locs);

    int max_pos = args["max_pos"].AsInteger() - 1; // convert to 0-based

    // figure out the domain requested
    CSignalSeq::EDomain domain;
    if (args["domain"].AsString() == "Bacterial") {
        domain = CSignalSeq::eBacterial;
    } else {
        domain = CSignalSeq::eEukaryotic;
    }

    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);

            // get sequence vector
            CSeqVector vec =
                handle.GetSequenceView(loc,
                                       CBioseq_Handle::eViewConstructed,
                                       CBioseq_Handle::eCoding_Ncbi);


            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());

            // places to store results
            double score;
            TSeqPos pos;

            // do the prediction
            CSignalSeq::Predict(vec, domain, max_pos, pos, score);

            // report results in the dialog
            m_Dialog->SetCell(row, 2) = score >= 3.5 ?
                "yes" : "no";
            m_Dialog->SetCell(row, 3) = NStr::DoubleToString(score);
            m_Dialog->SetCell(row, 4) = NStr::IntToString(pos + 1);

            if (score >= 3.5) {  // if a sig. seq. is predicted
                // a new feature table
                CRef<CSeq_annot> annot(new CSeq_annot());

                // add description to annot
                annot->AddName("Predicted signal sequence");
                string comment = string("Predicted signal sequence using ")
                    + args["domain"].AsString() +" weights";
                annot->AddComment(comment);

                //
                // add feature to annot
                //

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

                feat->SetExp_ev(CSeq_feat::eExp_ev_not_experimental);

                // set up the location
                feat->SetLocation().SetInt().SetFrom(0);
                feat->SetLocation().SetInt().SetTo(pos);
                feat->SetLocation().SetInt().SetStrand(sequence::GetStrand(loc));
                feat->SetLocation().SetId(sequence::GetId(loc));
                feat->SetLocation
                    (*CSeqUtils::RemapChildToParent(loc, feat->GetLocation()));
                feat->SetData().SetSite(CSeqFeatData::eSite_signal_peptide);
                
                // save in annot
                annot->SetData().SetFtable().push_back(feat);

                // attach annot to doc
                reply.AddObject(doc, *annot);
            }
        }
        catch (exception& e) {
            LOG_POST(Error << e.what());
            string str = CPluginUtils::GetLabel(loc, &doc.GetScope());
            LOG_POST(Error << "Error processing location " << str);
        }
#ifndef _DEBUG
        catch (...) {
            string str = CPluginUtils::GetLabel(loc, &doc.GetScope());
            LOG_POST(Error << "Error processing location " << str);
        }
#endif
    }

    // update all views
    //CDocManager::UpdateAllViews();

    //
    // prepare our dialog box
    //
    m_Dialog->SetLabel(string("Signal Sequence Prediction"
                              " Using ") 
                       + args["domain"].AsString() + " Weights");
    m_Dialog->Show();
    reply.SetStatus(eMessageStatus_success);
    reply.AddAction(CPluginReplyAction::e_Add_to_document);
}


END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: predict_signal_seq.cpp,v $
 * Revision 1000.5  2004/06/01 20:55:33  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.14
 *
 * Revision 1.14  2004/05/21 22:27:47  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.13  2004/05/03 13:05:42  dicuccio
 * gui/utils --> gui/objutils where needed
 *
 * Revision 1.12  2004/03/05 17:35:37  dicuccio
 * Use sequence::GetId() instead of CSeq_id::GetStringDescr()
 *
 * Revision 1.11  2004/01/27 18:38:04  dicuccio
 * Code clean-up.  Use standard names for plugins.  Removed unnecessary #includes
 *
 * Revision 1.10  2004/01/07 15:50:38  dicuccio
 * Adjusted for API change in CPluginUtils::GetLabel().  Standardized exception
 * reporting in algorithms.
 *
 * Revision 1.9  2003/11/24 15:45:27  dicuccio
 * Renamed CVersion to CPluginVersion
 *
 * Revision 1.8  2003/11/18 17:48:37  dicuccio
 * Added standard processing of return values
 *
 * Revision 1.7  2003/11/06 20:12:12  dicuccio
 * Cleaned up handling of USING_SCOPE - removed from all headers
 *
 * Revision 1.6  2003/11/04 17:49:23  dicuccio
 * Changed calling parameters for plugins - pass CPluginMessage instead of paired
 * CPluginCommand/CPluginReply
 *
 * Revision 1.5  2003/10/27 17:46:49  dicuccio
 * Removed dead #includes
 *
 * Revision 1.4  2003/10/14 16:24:37  dicuccio
 * Correctly remap new feature locations through the parent location to the master
 * sequence
 *
 * Revision 1.3  2003/10/07 13:47:00  dicuccio
 * Renamed CPluginURL* to CPluginValue*
 *
 * Revision 1.2  2003/09/25 17:21:35  jcherry
 * Added name to annot
 *
 * Revision 1.1  2003/09/10 15:31:34  jcherry
 * Initial version
 *
 * ===========================================================================
 */