bioseq_handle.hpp

ncbi源码

/*
 * ===========================================================================
 * PRODUCTION $Log: bioseq_handle.hpp,v $
 * PRODUCTION Revision 1000.4  2004/06/01 19:21:17  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.56
 * PRODUCTION
 * ===========================================================================
 */

#ifndef BIOSEQ_HANDLE__HPP
#define BIOSEQ_HANDLE__HPP

/*  $Id: bioseq_handle.hpp,v 1000.4 2004/06/01 19:21:17 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.
*
* ===========================================================================
*
* Author: Aleksey Grichenko, Michael Kimelman, Eugene Vasilchenko
*
* File Description:
*
*/

#include <corelib/ncbistd.hpp>

#include <objects/seqloc/Na_strand.hpp>
#include <objects/seqset/Bioseq_set.hpp> // for EClass
#include <objects/seq/Seq_inst.hpp> // for EMol

#include <objmgr/seq_id_handle.hpp>
#include <objmgr/impl/heap_scope.hpp>

BEGIN_NCBI_SCOPE
BEGIN_SCOPE(objects)

class CDataSource;
class CSeqMap;
class CSeqVector;
class CScope;
class CSeq_id;
class CSeq_loc;
class CBioseq_Info;
class CTSE_Info;
class CSeq_entry;
class CSeq_annot;
class CSeqMatch_Info;
class CSynonymsSet;
class CBioseq_ScopeInfo;
class CSeq_id_ScopeInfo;

class CBioseq_Handle;
class CSeq_annot_Handle;
class CSeq_entry_Handle;
class CBioseq_EditHandle;
class CBioseq_set_EditHandle;
class CSeq_annot_EditHandle;
class CSeq_entry_EditHandle;

// Bioseq handle -- must be a copy-safe const type.
class NCBI_XOBJMGR_EXPORT CBioseq_Handle
{
public:
    // Default constructor
    CBioseq_Handle(void);

    // Get scope this handle belongs to
    CScope& GetScope(void) const;

    // Get id used to obtain this bioseq handle
    CConstRef<CSeq_id> GetSeqId(void) const;
    const CSeq_id_Handle& GetSeq_id_Handle(void) const;

    // Check if this id can be used to obtain this bioseq handle
    bool IsSynonym(const CSeq_id& id) const;
    CConstRef<CSynonymsSet> GetSynonyms(void) const;

    // Get parent Seq-entry handle
    CSeq_entry_Handle GetParentEntry(void) const;
    CSeq_entry_Handle GetSeq_entry_Handle(void) const;

    // Get top level Seq-entry handle
    CSeq_entry_Handle GetTopLevelEntry(void) const;

    // Get 'edit' version of handle
    CBioseq_EditHandle GetEditHandle(void) const;

    // Bioseq core -- using partially populated CBioseq
    typedef CConstRef<CBioseq> TBioseqCore;
    // Get bioseq core structure
    TBioseqCore GetBioseqCore(void) const;
    // Get the complete bioseq
    CConstRef<CBioseq> GetCompleteBioseq(void) const;

    // Deprecated interface.
    const CBioseq& GetBioseq(void) const;
    const CSeq_entry& GetTopLevelSeqEntry(void) const;

    //////////////////////////////////////////////////////////////////
    // Bioseq members
    // id
    typedef vector<CSeq_id_Handle> TId;
    bool IsSetId(void) const;
    bool CanGetId(void) const;
    const TId& GetId(void) const;
    // descr
    typedef CSeq_descr TDescr;
    bool IsSetDescr(void) const;
    bool CanGetDescr(void) const;
    const TDescr& GetDescr(void) const;
    // inst
    typedef CSeq_inst TInst;
    bool IsSetInst(void) const;
    bool CanGetInst(void) const;
    const TInst& GetInst(void) const;
    // inst.repr
    typedef TInst::TRepr TInst_Repr;
    bool IsSetInst_Repr(void) const;
    bool CanGetInst_Repr(void) const;
    TInst_Repr GetInst_Repr(void) const;
    // inst.mol
    typedef TInst::TMol TInst_Mol;
    bool IsSetInst_Mol(void) const;
    bool CanGetInst_Mol(void) const;
    TInst_Mol GetInst_Mol(void) const;
    // inst.length
    typedef TInst::TLength TInst_Length;
    bool IsSetInst_Length(void) const;
    bool CanGetInst_Length(void) const;
    TInst_Length GetInst_Length(void) const;
    TSeqPos GetBioseqLength(void) const; // try to calculate it if not set
    // inst.fuzz
    typedef TInst::TFuzz TInst_Fuzz;
    bool IsSetInst_Fuzz(void) const;
    bool CanGetInst_Fuzz(void) const;
    const TInst_Fuzz& GetInst_Fuzz(void) const;
    // inst.topology
    typedef TInst::TTopology TInst_Topology;
    bool IsSetInst_Topology(void) const;
    bool CanGetInst_Topology(void) const;
    TInst_Topology GetInst_Topology(void) const;
    // inst.strand
    typedef TInst::TStrand TInst_Strand;
    bool IsSetInst_Strand(void) const;
    bool CanGetInst_Strand(void) const;
    TInst_Strand GetInst_Strand(void) const;
    // inst.seq-data
    typedef TInst::TSeq_data TInst_Seq_data;
    bool IsSetInst_Seq_data(void) const;
    bool CanGetInst_Seq_data(void) const;
    const TInst_Seq_data& GetInst_Seq_data(void) const;
    // inst.ext
    typedef TInst::TExt TInst_Ext;
    bool IsSetInst_Ext(void) const;
    bool CanGetInst_Ext(void) const;
    const TInst_Ext& GetInst_Ext(void) const;
    // inst.hist
    typedef TInst::THist TInst_Hist;
    bool IsSetInst_Hist(void) const;
    bool CanGetInst_Hist(void) const;
    const TInst_Hist& GetInst_Hist(void) const;
    // annot
    bool HasAnnots(void) const;

    //////////////////////////////////////////////////////////////////
    // Old interface:

    // Go up to a certain complexity level (or the nearest level of the same
    // priority if the required class is not found):
    // level   class
    // 0       not-set (0) ,
    // 3       nuc-prot (1) ,       -- nuc acid and coded proteins
    // 2       segset (2) ,         -- segmented sequence + parts
    // 2       conset (3) ,         -- constructed sequence + parts
    // 1       parts (4) ,          -- parts for 2 or 3
    // 1       gibb (5) ,           -- geninfo backbone
    // 1       gi (6) ,             -- geninfo
    // 5       genbank (7) ,        -- converted genbank
    // 3       pir (8) ,            -- converted pir
    // 4       pub-set (9) ,        -- all the seqs from a single publication
    // 4       equiv (10) ,         -- a set of equivalent maps or seqs
    // 3       swissprot (11) ,     -- converted SWISSPROT
    // 3       pdb-entry (12) ,     -- a complete PDB entry
    // 4       mut-set (13) ,       -- set of mutations
    // 4       pop-set (14) ,       -- population study
    // 4       phy-set (15) ,       -- phylogenetic study
    // 4       eco-set (16) ,       -- ecological sample study
    // 4       gen-prod-set (17) ,  -- genomic products, chrom+mRNa+protein
    // 4       wgs-set (18) ,       -- whole genome shotgun project
    // 0       other (255)
    CSeq_entry_Handle GetComplexityLevel(CBioseq_set::EClass cls) const;
    // Return level with exact complexity, or empty handle if not found.
    CSeq_entry_Handle GetExactComplexityLevel(CBioseq_set::EClass cls) const;

    // Get some values from core:
    CSeq_inst::TMol GetBioseqMolType(void) const;

    // Get sequence map.
    const CSeqMap& GetSeqMap(void) const;

    // CSeqVector constructor flags
    enum EVectorCoding {
        eCoding_NotSet, // Use original coding - DANGEROUS! - may change
        eCoding_Ncbi,   // Set coding to binary coding (Ncbi4na or Ncbistdaa)
        eCoding_Iupac   // Set coding to printable coding (Iupacna or Iupacaa)
    };
    enum EVectorStrand {
        eStrand_Plus,   // Plus strand
        eStrand_Minus   // Minus strand
    };

    // Get sequence: Iupacna or Iupacaa if use_iupac_coding is true
    CSeqVector GetSeqVector(EVectorCoding coding,
                            ENa_strand strand = eNa_strand_plus) const;
    CSeqVector GetSeqVector(ENa_strand strand = eNa_strand_plus) const;
    CSeqVector GetSeqVector(EVectorCoding coding, EVectorStrand strand) const;
    CSeqVector GetSeqVector(EVectorStrand strand) const;

    // Sequence filtering: get a seq-vector for a part of the sequence.
    // The part shown depends oon the mode selected. If the location
    // contains references to other sequences they are ignored (unlike
    // CBioseq constructor, which constructs a bioseq using all references
    // from a location). Strand information from "location" is ingored
    // when creating merged or excluded views. If "minus_strand" is true,
    // the result is reverse-complement.
    enum ESequenceViewMode {
        eViewConstructed,    // Do not merge or reorder intervals
        eViewMerged,         // Merge overlapping intervals, sort by location
        eViewExcluded        // Show intervals not included in the seq-loc
    };
    CSeqVector GetSequenceView(const CSeq_loc& location,
                               ESequenceViewMode mode,
                               EVectorCoding coding = eCoding_Ncbi,
                               ENa_strand strand = eNa_strand_plus) const;


    CConstRef<CSeqMap> GetSeqMapByLocation(const CSeq_loc& location,
                                           ESequenceViewMode mode) const;

    CRef<CSeq_loc> MapLocation(const CSeq_loc& loc) const;

    // Utility methods/operators

    // Comparison
    bool operator== (const CBioseq_Handle& h) const;
    bool operator!= (const CBioseq_Handle& h) const;
    bool operator<  (const CBioseq_Handle& h) const;

    // Check
    operator bool(void)  const;
    bool operator!(void) const;

    // these methods are for cross scope move only.
    CBioseq_EditHandle CopyTo(const CSeq_entry_EditHandle& entry,
                              int index = -1) const;
    CBioseq_EditHandle CopyTo(const CBioseq_set_EditHandle& seqset,
                              int index = -1) const;
    CBioseq_EditHandle CopyToSeq(const CSeq_entry_EditHandle& entry) const;

protected:
    friend class CScope_Impl;
    friend class CSynonymsSet;

    CBioseq_Handle(const CSeq_id_Handle& id, CBioseq_ScopeInfo* bioseq_info);
    const CBioseq_ScopeInfo& x_GetScopeInfo(void) const;

    CHeapScope          m_Scope;
    CSeq_id_Handle      m_Seq_id;
    CConstRef<CObject>  m_Bioseq_Info;

public: // non-public section
    const CBioseq_Info& x_GetInfo(void) const;
};


class NCBI_XOBJMGR_EXPORT CBioseq_EditHandle : public CBioseq_Handle
{
public:
    CBioseq_EditHandle(void);
    
    CSeq_entry_EditHandle GetParentEntry(void) const;

    // Modification functions

    //////////////////////////////////////////////////////////////////
    // Bioseq members
    // id
    void ResetId(void) const;
    bool AddId(const CSeq_id_Handle& id) const;
    bool RemoveId(const CSeq_id_Handle& id) const;
    // descr
    void SetDescr(TDescr& v) const;
    // inst
    void SetInst(TInst& v) const;
    // inst.repr
    void SetInst_Repr(TInst_Repr v) const;
    // inst.mol
    void SetInst_Mol(TInst_Mol v) const;