pattern2.cpp

来自「ncbi源码」· C++ 代码 · 共 380 行

/*
 * ===========================================================================
 * PRODUCTION $Log: pattern2.cpp,v $
 * PRODUCTION Revision 1000.1  2004/06/01 20:55:26  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.9
 * PRODUCTION
 * ===========================================================================
 */

/*  $Id: pattern2.cpp,v 1000.1 2004/06/01 20:55:26 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 searching against patterns from file
 *
 */


#include <ncbi_pch.hpp>
#include "pattern2.hpp"

#include <algo/sequence/find_pattern.hpp>
#include <corelib/ncbiapp.hpp>
#include <corelib/ncbireg.hpp>
#include <gui/core/plugin_utils.hpp>
#include <gui/utils/system_path.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 <objects/seqloc/Seq_interval.hpp>
#include <objmgr/seq_vector.hpp>
#include <objmgr/util/sequence.hpp>


BEGIN_NCBI_SCOPE
USING_SCOPE(objects);


CAlgoPlugin_NamedPatterns::~CAlgoPlugin_NamedPatterns()
{
}


// standard plugin announce bopilerplate
void CAlgoPlugin_NamedPatterns::GetInfo(CPluginInfo& info)
{
    info.Reset();
    
    // version info macro
    info.SetInfo(CPluginVersion::eMajor, CPluginVersion::eMinor, 0,
                 string(__DATE__) + " " + string(__TIME__),
                 "CAlgoPlugin_NamedPatterns",
                 "Search/Named Pattern Sets",
                 "Search a sequence against patterns from a file", "");


    // 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.AddDefaultArgument("patterns", "Pattern set", CPluginArg::eString,
                            "Load from file specified below");
    args.AddOptionalArgument("fname", "File containing patterns",
                              CPluginArg::eFile);
}


void CAlgoPlugin_NamedPatterns::FinalizeArgs(CPluginMessage& msg)
{
    CPluginCommand& args = msg.SetRequest().SetCommand();
    x_UpdateFileList();
    CPluginValueConstraint *constraint = CPluginValueConstraint::CreateSet();
    constraint->SetSet().push_back("Load from file specified below");
    typedef map<string, string> TMss;
    ITERATE (TMss, it, m_FileList) {
        constraint->SetSet().push_back(it->first);
    }
    args.SetConstraint("patterns", *constraint);
}


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

    
    if ( !m_Dialog.get() ) {
        m_Dialog.reset(new CMultiColDlg());
        m_Dialog->SetWindowSize(1000, 350);
        m_Dialog->SetTitle("Pattern Search Results");

        m_Dialog->SetColumn(0, "Sequence", FL_ALIGN_LEFT, 1.0f);
        m_Dialog->SetColumn(1, "Location", FL_ALIGN_LEFT, 1.0f);
        m_Dialog->SetColumn(2, "Position", FL_ALIGN_CENTER, 1.0f);
        m_Dialog->SetColumn(3, "Pattern ID", FL_ALIGN_LEFT, 3.0f);
        m_Dialog->SetColumn(4, "Pattern Description", FL_ALIGN_LEFT, 4.0f);
        m_Dialog->SetColumn(5, "Pattern", FL_ALIGN_LEFT, 3.0f);
        m_Dialog->SetColumn(6, "Matched Sequence", FL_ALIGN_LEFT, 2.0f);
    }

    m_Dialog->SetRows(0);  // to clear any previous contents
 
    vector<TSeqPos> starts;
    vector<TSeqPos> ends;

    string fname;

    // determine which file to load
    if (args["patterns"].AsString() == "Load from file specified below") {
        fname = args["fname"].AsFile();
    } else {
        fname = m_FileList[args["patterns"].AsString()];
    }

    CNcbiRegistry patterns;
    try {
        CNcbiIfstream is(fname.c_str());
        patterns.Read(is);
    }
    catch (exception& e) {
        LOG_POST(Error << "Couldn't load registry from file " <<
                 fname << ": " << e.what());
        reply.SetStatus(eMessageStatus_failed);
        return;
    }
    catch (...) {
        LOG_POST(Error << "Couldn't load registry from file " <<
                 fname);
        reply.SetStatus(eMessageStatus_failed);
        return;
    }

    //
    // first, evaluate whole sequences
    //
    int row = 0;
    plugin_args::TLocList locs;
    GetArgValue(args["locs"], locs);

    unsigned int patterns_searched;

    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());

            // a new feature table
            CRef<CSeq_annot> annot(new CSeq_annot());

            // iterate over patterns
            list<string> pat_ids;
            patterns.EnumerateSections(&pat_ids);
            patterns_searched = 0;

            ITERATE (list<string>, pat_id, pat_ids) {

                if (*pat_id == "-") {
                    continue;
                }
                patterns_searched++;
                string pattern = patterns.Get(*pat_id, "pattern");
                string desc = patterns.Get(*pat_id, "description");
                CFindPattern::Find(seq, pattern, starts, ends);

                for(unsigned int k = 0;  k < starts.size();  k++) {
                    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) = *pat_id;
                    m_Dialog->SetCell(row, 4) = desc;
                    m_Dialog->SetCell(row, 5) = pattern;

                    m_Dialog->SetCell(row, 6) =
                        seq.substr(starts[k], ends[k] - starts[k] + 1);
                    ++row;
                }

                //
                // add features to annot
                //

                for( unsigned int k = 0;  k < starts.size();  k++) {
                    // 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]);
                    feat->SetLocation
                        (*CSeqUtils::RemapChildToParent(loc, floc));

                    // set feature data
                    feat->SetData().SetRegion() = "Pattern match:: "
                        + desc + "; " + *pat_id;
    
                    // save in annot
                    annot->SetData().SetFtable().push_back(feat);
                }
            }

            // add description to annot
            annot->SetName("Pattern matches");

            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->SetLabel(string("A search against ")
                 + NStr::IntToString(patterns_searched)
                 + " patterns produced:");
    m_Dialog->Show();
    reply.SetStatus(eMessageStatus_success);
    reply.AddAction(CPluginReplyAction::e_Add_to_document);
}


void CAlgoPlugin_NamedPatterns::x_UpdateFileList(void)
{
    string dir;

    m_FileList.clear();

    CNcbiApplication* app = CNcbiApplication::Instance();
    _ASSERT(app);
    CNcbiRegistry& registry = app->GetConfig();
    
    if ( (dir = registry.Get("Patterns", "PatternPath")).empty() ) {
        registry.Set("Patterns", "PatternPath", "<std>, <home>",
                     CNcbiRegistry::ePersistent, " default external_path");
    }
    dir = registry.Get("Patterns", "PatternPath");
    
    list<string> paths;
    NStr::Split(dir, ", \t\n\r", paths);

    ITERATE (list<string>, iter, paths) {
        string dir_name;
        if (*iter == "<std>"  ||  *iter == "<home>") {
            dir_name = CSystemPath::ResolvePath(*iter, "etc/patterns");
        } else {
            dir_name = CSystemPath::ResolvePath(*iter, "");
        }
        if ( dir_name.empty() ) {
            continue;
        }
        
        CDir dir(dir_name);
        if ( !dir.Exists() ) {
            continue;
        }

        CDir::TEntries entries = dir.GetEntries("*.ini");
        ITERATE (CDir::TEntries, entry_iter, entries) {
            if ( !(*entry_iter)->IsFile() ) {
                continue;
            }
            
            string full_path = (*entry_iter)->GetPath();

            CNcbiIfstream reg_stream(full_path.c_str());
            CNcbiRegistry my_reg(reg_stream);
            list<string> sections;

            string name = my_reg.Get("-", "name");
            if (!name.empty()) {
                m_FileList[name] = full_path;
            }
        }
    }
}

END_NCBI_SCOPE


/*
 * ===========================================================================
 * $Log: pattern2.cpp,v $
 * Revision 1000.1  2004/06/01 20:55:26  gouriano
 * PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.9
 *
 * Revision 1.9  2004/05/21 22:27:47  gorelenk
 * Added PCH ncbi_pch.hpp
 *
 * Revision 1.8  2004/05/03 13:05:42  dicuccio
 * gui/utils --> gui/objutils where needed
 *
 * Revision 1.7  2004/03/05 17:35:37  dicuccio
 * Use sequence::GetId() instead of CSeq_id::GetStringDescr()
 *
 * Revision 1.6  2004/02/17 20:35:25  rsmith
 * moved core/settings.[ch]pp and core/system_path.[ch]pp to config and utils, respectively.
 *
 * Revision 1.5  2004/02/04 15:27:48  jcherry
 * patterns->etc/patterns
 *
 * Revision 1.4  2004/01/27 18:37:56  dicuccio
 * Code clean-up.  Use standard names for plugins.  Removed unnecessary #includes
 *
 * Revision 1.3  2004/01/21 23:44:45  jcherry
 * Present a menu that includes sets of patterns from all *.ini files
 * in certain directories (uses new FinalizeArgs() API)
 *
 * 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  2003/12/16 22:22:16  jcherry
 * Initial version
 *
 * ===========================================================================
 */