cdd_search.cpp

来自「ncbi源码」· C++ 代码 · 共 337 行

/*
 * ===========================================================================
 * PRODUCTION $Log: cdd_search.cpp,v $
 * PRODUCTION Revision 1000.6  2004/06/01 20:54:50  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.13
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: cdd_search.cpp,v 1000.6 2004/06/01 20:54:50 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:  Mike DiCuccio
 *
 * File Description:
 *    CAlgoPlugin_CDDSearch - load sequence information from a file.
 */

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

#include <connect/ncbi_conn_stream.hpp>
#include <gui/core/doc_manager.hpp>
#include <gui/core/idocument.hpp>
#include <gui/core/version.hpp>
#include <gui/plugin/PluginCommandSet.hpp>
#include <gui/plugin/PluginInfo.hpp>
#include <gui/plugin/PluginRequest.hpp>
#include <gui/plugin/PluginValueConstraint.hpp>
#include <gui/utils/message_box.hpp>
#include <objects/seqalign/Dense_seg.hpp>
#include <objects/seqalign/Std_seg.hpp>
#include <objects/seqalign/Seq_align.hpp>
#include <objmgr/bioseq_handle.hpp>
#include <objmgr/object_manager.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>
#include <objtools/data_loaders/cdd/cdd.hpp>


BEGIN_NCBI_SCOPE
USING_SCOPE(objects);


CCddDataLoader* CAlgoPlugin_CDDSearch::sm_Loader = NULL;



CAlgoPlugin_CDDSearch::~CAlgoPlugin_CDDSearch()
{
}


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

    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoPlugin_CDDSearch",
                 "Search/Conserved Domains",
                 "Search for conserved domains in this protein",
                 "");

    // command info
    CPluginCommandSet& cmds     = info.SetCommands();
    CPluginCommand& args   = cmds.AddAlgoCommand(eAlgoCommand_run);
    args.AddArgument("id", "Sequence to search",
                     CSeq_id::GetTypeInfo());
    args.SetConstraint("id",
                       (*CPluginValueConstraint::CreateSeqMol(),
                        CSeq_inst::eMol_aa));

    args.AddDefaultArgument("expect", "Expect value",
                            CPluginArg::eDouble, "0.01");
    args.AddArgument("filter", "Filter results",
                     CPluginArg::eBoolean);
    args.AddDefaultArgument("type", "Type to retrieve",
                            CPluginArg::eString,
                            "Alignment");
    args.SetConstraint("type",
                       (*CPluginValueConstraint::CreateSet(),
                        "Alignment", "Feature table"));
}


//
// Save()
// Save the information in a given document into a file
//
void CAlgoPlugin_CDDSearch::RunCommand(CPluginMessage& msg)
{
    const CPluginCommand& args = msg.GetRequest().GetCommand();
    CPluginReply& reply = msg.SetReply();
    reply.SetStatus(eMessageStatus_failed);

    const CSeq_id&   id  = dynamic_cast<const CSeq_id&>(args["id"].AsObject());
    const IDocument* doc = args["id"].GetDocument();

    if ( !doc ) {
        NcbiMessageBox("No valid sequence found.");
        return;
    }

    bool   filter = args["filter"].AsBoolean();
    string type   = args["type"  ].AsString();
    double expect = args["expect"].AsDouble();

    string filter_str("T");
    if ( !filter ) {
        filter_str = "F";
    }

    if (type == "Alignment") {
        type = "LONLY";
    } else if (type == "Feature table") {
        type = "AONLY";
    }

    CScope& scope = doc->GetScope();

    //
    // retrieve our protein sequence
    // we must pass the sequence in IUPACaa format
    //
    CBioseq_Handle handle = scope.GetBioseqHandle(id);
    CSeqVector vec        = handle.GetSeqVector();
    vec.SetCoding(CSeq_data::e_Iupacaa);
    string data;
    vec.GetSeqData(0, vec.size(), data);

    //
    // compose our parameters
    //
    string params("SEQUENCE=");
    params += data + "&";
    params += "NOHTML&";
    params += "DATALIB=cdd&";
    params += "EXPECT=" + NStr::DoubleToString(expect) + "&";
    params += "GRAPH=1&";
    params += "FILTER=" + filter_str + "&";
    params += type;

    //
    // CConn_ServiceStream i/o
    //
    CConn_ServiceStream cdd_str("CddSearch");
    cdd_str << params;

    CRef<CSeq_annot> annot(new CSeq_annot());
    try {
        auto_ptr<CObjectIStream> is
            (CObjectIStream::Open(eSerial_AsnText, cdd_str));
        *is >> *annot;
    }
    catch (...) {
        // our read failed - it is likely that our output was not ASN.1
        string msg("No conserved domains were found for sequence\n");
        msg += sequence::GetTitle(handle);
        NcbiMessageBox(msg);
        return;
    }

    //
    // fix our IDs
    //
    if (annot->GetData().IsFtable()) {
        if (annot->GetData().GetFtable().size() == 0) {
            annot.Reset();
        } else {

            NON_CONST_ITERATE (CSeq_annot::TData::TFtable, iter,
                               annot->SetData().SetFtable()) {
                CSeq_feat* feat = *iter;
                const CSeq_id& this_id = sequence::GetId(feat->SetLocation());
                if (this_id.IsLocal()) {
                    feat->SetLocation().SetId(id);
                }
            }
        }
    } else if (annot->GetData().IsAlign()) {
        if (annot->GetData().GetAlign().size() == 0) {
            annot.Reset();
        } else {
            NON_CONST_ITERATE (CSeq_annot::TData::TAlign, iter,
                               annot->SetData().SetAlign()) {
                CSeq_align& align = **iter;
                switch (align.GetSegs().Which()) {
                case CSeq_align::TSegs::e_Denseg:
                    {{
                        CDense_seg& seg = align.SetSegs().SetDenseg();
                        NON_CONST_ITERATE(CDense_seg::TIds, id_iter, seg.SetIds()) {
                            CSeq_id& this_id = **id_iter;
                            if (this_id.IsLocal()) {
                                this_id.Assign(id);
                            }
                        }
                    }}
                    break;

                case CSeq_align::TSegs::e_Std:
                    NON_CONST_ITERATE (CSeq_align::TSegs::TStd, iter,
                                       align.SetSegs().SetStd()) {
                        CStd_seg& seg = **iter;
                        NON_CONST_ITERATE(CStd_seg::TIds, id_iter, seg.SetIds()) {
                            CSeq_id& this_id = **id_iter;
                            if (this_id.IsLocal()) {
                                this_id.Assign(id);
                            }
                        }
                    }
                    break;

                default:
                    LOG_POST(Error << "unhandled seq-align type in CDDSearch");
                    break;
                }
            }
        }
    }

    if ( !annot ) {
        string msg("No conserved domains were found for sequence\n");
        msg += sequence::GetTitle(handle);
        NcbiMessageBox(msg);
        reply.SetStatus(eMessageStatus_failed);
        return;
    }

    //
    // return our object to the framework
    //
    reply.AddObject(*doc, *annot);
    reply.AddAction(CPluginReplyAction::e_Add_to_document);

    //
    // take care of our data loader
    if ( !sm_Loader ) {

        DEFINE_STATIC_FAST_MUTEX(s_Mutex);

        CFastMutexGuard LOCK(s_Mutex);
        if ( !sm_Loader ) {
            sm_Loader = new CCddDataLoader();
            CDocManager::GetObjectManager().RegisterDataLoader(*sm_Loader);
        }
    }

    scope.AddDataLoader("CDD_DataLoader");
    reply.SetStatus(eMessageStatus_success);
}



END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: cdd_search.cpp,v $
 * Revision 1000.6  2004/06/01 20:54:50  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.13
 *
 * Revision 1.13  2004/05/21 22:27:46  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.12  2004/04/16 14:43:15  dicuccio
 * Deleted dead code
 *
 * Revision 1.11  2004/03/24 13:57:19  friedman
 * Fixed mutex comment
 *
 * Revision 1.10  2004/03/23 19:55:18  friedman
 * Replaced 'static CFastMutex' with DEFINE_STATIC_FAST_MUTEX
 *
 * Revision 1.9  2004/03/16 15:37:22  vasilche
 * Added required include
 *
 * Revision 1.8  2004/01/27 18:37:32  dicuccio
 * Code clean-up.  Use standard names for plugins.  Removed unnecessary #includes
 *
 * Revision 1.7  2003/11/24 15:45:25  dicuccio
 * Renamed CVersion to CPluginVersion
 *
 * Revision 1.6  2003/11/18 17:47:57  dicuccio
 * Made data loader static and thread-safe.  Added standard processing for return values
 *
 * Revision 1.5  2003/11/06 20:12:12  dicuccio
 * Cleaned up handling of USING_SCOPE - removed from all headers
 *
 * Revision 1.4  2003/11/04 17:49:22  dicuccio
 * Changed calling parameters for plugins - pass CPluginMessage instead of paired
 * CPluginCommand/CPluginReply
 *
 * Revision 1.3  2003/10/20 18:30:56  dicuccio
 * Added CDD data laoder for retrieval of CDD alignment sequences.  Cleaned up
 * handling of imput streams
 *
 * Revision 1.2  2003/10/16 19:38:43  dicuccio
 * Removed old debugging code
 *
 * Revision 1.1  2003/10/16 12:56:35  dicuccio
 * Added CDD Search facility.  Cleaned up dll_register.cpp
 *
 * Revision 1.2  2003/10/07 13:47:05  dicuccio
 * Renamed CPluginURL* to CPluginValue*
 *
 * Revision 1.1  2003/09/16 14:06:48  dicuccio
 * Initial revision - split from CFileLoader
 *
 * ===========================================================================
 */