genemark_loader.cpp

来自「ncbi源码」· C++ 代码 · 共 564 行 · 第 1/2 页

/*
 * ===========================================================================
 * PRODUCTION $Log: genemark_loader.cpp,v $
 * PRODUCTION Revision 1000.5  2004/06/01 20:58:40  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.35
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: genemark_loader.cpp,v 1000.5 2004/06/01 20:58:40 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: Dmitry Dernovoy
 *
 * File Description:
 *   CGeneMarkLoader - Plugin to load GeneMark's predictions 
 */

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

#include <gui/core/idocument.hpp>
#include <gui/core/version.hpp>
#include <gui/dialogs/file_browser.hpp>
#include <gui/plugin/PluginCommandSet.hpp>
#include <gui/plugin/PluginInfo.hpp>
#include <gui/plugin/PluginValue.hpp>

#include <objects/general/Int_fuzz.hpp>
#include <objects/general/Object_id.hpp>
#include <objects/seqfeat/Cdregion.hpp>
#include <objects/seqfeat/Feat_id.hpp>
#include <objects/seqfeat/Genetic_code.hpp>
#include <objects/seqfeat/SeqFeatData.hpp>
#include <objects/seqfeat/Seq_feat.hpp>
#include <objects/seqloc/Seq_interval.hpp>
#include <objects/seqloc/Seq_loc.hpp>

#define            GENEMARK_MAXLINE           200

BEGIN_NCBI_SCOPE


void CGeneMarkLoader::GetInfo(CPluginInfo& info)
{
    info.Reset();

    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CGeneMarkLoader",
                 "GeneMark\\/Glimmer output",
                 "Load the results from a GeneMark/Glimmer run", "");

    // command info
    CPluginCommandSet& cmds     = info.SetCommands();
    CPluginCommand& import_args = cmds.AddDataCommand(eDataCommand_import);
    import_args.AddArgument("document", "Document", CPluginArg::eDocument);
}


CGeneMarkLoader::CGeneMarkLoader()
{
}

CGeneMarkLoader::~CGeneMarkLoader()
{
}

void CGeneMarkLoader::Import(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    reply.SetStatus(eMessageStatus_failed);

    LOG_POST(Info << "CGeneMarkLoader::Load: start point.");

    IDocument* doc = const_cast<IDocument*> (&args["document"].AsDocument());
    if ( !doc ) {
        reply.SetStatus(eMessageStatus_failed);
        return;
    }



    try {
        LOG_POST(Info << "CGeneMarkLoader::Load: reading file...");

        CConstRef<CSeq_id> id(dynamic_cast<const CSeq_id*> (doc->GetObject()));
        if ( !id ) {
            _TRACE("CGeneMarkLoader::Load: can't get any reasonable seq_id");
            return;
        }

        string acc_text = id->GetSeqIdString(false) ;

        //   LOG_POST( Info << "CGeneMarkLoader:: accession: " << acc_text );
        //  _TRACE("CGeneMarkLoader:: accession: " << acc_text );

        string tmp_str =  string("(") + NStr::ToUpper(acc_text) + string(".fna.g*)");  //"(*.{gmHMM,gmark,glim_*})"

        string fname =
            NcbiFileBrowser("Open GeneMark/GeneMark.hmm/Glimmer output file...",
                            tmp_str.c_str(),
#if defined(NCBI_OS_MSWIN)
                            "\\\\Atlas\\b11\\tatiana\\Predictions\\");
#else
                            "/net/atlas/b11/tatiana/Predictions/");
#endif


        if ( fname.empty() ) {
            reply.SetStatus(eMessageStatus_ignored);
            return;
        }

        CBioseq_Handle handle = doc->GetScope().GetBioseqHandle(*id);
        if ( !handle ) {
            _TRACE("CGeneMarkLoader::Load: can't get bioseq handle");
            return;
        }

        CRef<CSeq_annot> annot;

        switch(x_RecognizeFormat(fname))
        {
        case fGeneMark:
            annot = x_LoadGeneMarkFile(fname, *id);
            break;
        case fGeneMarkHMM:
            annot = x_LoadGeneMarkHmmFile(fname, *id);
            break;
        case fGlimmer2:
            annot = x_LoadGlimmer2File(fname, *id);
            break;
        default:
            _TRACE("Unknown return code");
        case fUnknownFormat:
            _TRACE("Unsupported file format, file: "<< fname); 
            reply.SetStatus(eMessageStatus_failed);
            return;
        }


        // save the object in the reply for framework processing
        reply.AddObject(*doc, *annot);
        reply.AddAction(CPluginReplyAction::e_Add_to_document);
        reply.SetStatus(eMessageStatus_success);
    }

    catch (CException& e) {
        LOG_POST(Info << "failed to read GeneMark file: " << e.what());
        _TRACE("failed to read GeneMark file: " << e.what() );
    }
#ifndef _DEBUG
    catch (...) {
        _TRACE("failed to read GeneMark file: unknown error");
    }
#endif
}


CGeneMarkLoader::EFileFormat
CGeneMarkLoader::x_RecognizeFormat(const string& fname)
{
    CNcbiIfstream istr( fname.c_str() );
    char buf[ GENEMARK_MAXLINE +1 ];

    istr.getline(buf, GENEMARK_MAXLINE);

    if( strstr(buf,"r=-1.") || strstr(buf,"r=-0.") ) return fGlimmer2;

    for(int i=1; (i < 30) && !istr.eof() ; ++i)
    {
        istr.getline(buf, GENEMARK_MAXLINE);
        if(strstr(buf, "Strand"))
        {
            if(strstr(buf,"Frame"))    return fGeneMark;
            if(strstr(buf,"RightEnd")) return fGeneMarkHMM;
        }
    }
    return fUnknownFormat;
}

CRef<CSeq_annot>
CGeneMarkLoader::x_LoadGlimmer2File(const string&  fname,
                                    const CSeq_id& id)
{
    CNcbiIfstream istr(fname.c_str());

    CRef<CSeq_annot> annot( new CSeq_annot() );
    annot->AddName("Glimmer2 predictions");

    list< CRef<CSeq_feat> >& ftable = annot->SetData().SetFtable(); 

    char buf[ GENEMARK_MAXLINE +1 ];

    while( istr.getline(buf, GENEMARK_MAXLINE) )
    {
        CNcbiIstrstream Lstr( buf );

        int     gene_number = 0, glim_from = 0, glim_to = 0;
        Lstr >> gene_number;
        Lstr >> glim_from;
        Lstr >> glim_to;

        char strand = 'U';
        for( ; (strand != '-') && (strand != '+'); Lstr >> strand );

        TSeqPos LeftEnd, RightEnd;

        if(strand == '-')
        {
            LeftEnd  = glim_to - 3;
            RightEnd = glim_from;
        }else{                         // direct
            LeftEnd  = glim_from;
            RightEnd = glim_to + 3;
        }

        if(LeftEnd > RightEnd) // Glimmer prediction over zero-point
        {
            // can't handle yet
            LOG_POST(Info << "CGeneMarkLoader::Can't handle Glimmer's prediction over zero-point, skipped.");
            continue;
        }

        CRef<CSeq_feat> feat(new CSeq_feat());

        feat->SetComment() = "Glimmer2 pred #" + NStr::IntToString(gene_number) ;

        CSeq_interval& floc = feat->SetLocation().SetInt();

        floc.SetFrom(LeftEnd- 1);
        floc.SetTo(RightEnd - 1); 
        floc.SetStrand((strand == '-') ? eNa_strand_minus : eNa_strand_plus);

        floc.SetId().Assign(id);    // floc.SetId().SetGi( NStr::StringToInt(m_SeqId) );

        CSeqFeatData& fdata = feat->SetData();
        CCdregion& cdreg = fdata.SetCdregion();
        cdreg.SetFrame(CCdregion::eFrame_one);

        list< CRef< CGenetic_code::C_E > >& gcode = cdreg.SetCode().Set();
        CRef< CGenetic_code::C_E > ce(new CGenetic_code::C_E);
        ce->SetId(11);                                        // TSE=1; seq=1; feat=1
        gcode.push_back(ce);

        ftable.push_back(feat);
    }
    return annot;
}

CRef<CSeq_annot>
CGeneMarkLoader::x_LoadGeneMarkHmmFile(const string& fname,
                                       const CSeq_id& id)
{
    CNcbiIfstream istr(fname.c_str());

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