readfeat.cpp

来自「ncbi源码」· C++ 代码 · 共 1,648 行 · 第 1/5 页

/*
 * ===========================================================================
 * PRODUCTION $Log: readfeat.cpp,v $
 * PRODUCTION Revision 1000.5  2004/06/01 19:46:24  gouriano
 * PRODUCTION PRODUCTION: UPGRADED [GCC34_MSVC7] Dev-tree R1.35
 * PRODUCTION
 * ===========================================================================
 */

/*
 * ===========================================================================
 *
 *                            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:  Jonathan Kans
 *
 * File Description:
 *   Feature table reader
 *
 */

#include <ncbi_pch.hpp>
#include <corelib/ncbistd.hpp>
#include <corelib/ncbithr.hpp>

#include <serial/iterator.hpp>
#include <serial/objistrasn.hpp>

// Objects includes
#include <objects/general/Int_fuzz.hpp>
#include <objects/general/Object_id.hpp>
#include <objects/general/Dbtag.hpp>

#include <objects/seqloc/Seq_id.hpp>
#include <objects/seqloc/Seq_loc.hpp>
#include <objects/seqloc/Seq_interval.hpp>
#include <objects/seqloc/Seq_point.hpp>

#include <objects/seq/Seq_annot.hpp>
#include <objects/seq/Annotdesc.hpp>
#include <objects/seq/Annot_descr.hpp>
#include <objects/seqfeat/SeqFeatData.hpp>

#include <objects/seqfeat/Seq_feat.hpp>
#include <objects/seqfeat/BioSource.hpp>
#include <objects/seqfeat/Org_ref.hpp>
#include <objects/seqfeat/OrgName.hpp>
#include <objects/seqfeat/SubSource.hpp>
#include <objects/seqfeat/OrgMod.hpp>
#include <objects/seqfeat/Gene_ref.hpp>
#include <objects/seqfeat/Cdregion.hpp>
#include <objects/seqfeat/Code_break.hpp>
#include <objects/seqfeat/Genetic_code.hpp>
#include <objects/seqfeat/Genetic_code_table.hpp>
#include <objects/seqfeat/RNA_ref.hpp>
#include <objects/seqfeat/Trna_ext.hpp>
#include <objects/seqfeat/Imp_feat.hpp>
#include <objects/seqfeat/Gb_qual.hpp>

#include <objtools/readers/readfeat.hpp>

#include <algorithm>

BEGIN_NCBI_SCOPE

BEGIN_objects_SCOPE // namespace ncbi::objects::

class /* NCBI_XOBJREAD_EXPORT */ CFeature_table_reader_imp
{
public:
    enum EQual {
        eQual_allele,
        eQual_anticodon,
        eQual_bond_type,
        eQual_bound_moiety,
        eQual_citation,
        eQual_clone,
        eQual_codon_start,
        eQual_cons_splice,
        eQual_db_xref,
        eQual_direction,
        eQual_EC_number,
        eQual_evidence,
        eQual_exception,
        eQual_frequency,
        eQual_function,
        eQual_gene,
        eQual_gene_desc,
        eQual_gene_syn,
        eQual_go_component,
        eQual_go_function,
        eQual_go_process,
        eQual_insertion_seq,
        eQual_label,
        eQual_locus_tag,
        eQual_macronuclear,
        eQual_map,
        eQual_MEDLINE,
        eQual_mod_base,
        eQual_muid,
        eQual_note,
        eQual_number,
        eQual_operon,
        eQual_organism,
        eQual_partial,
        eQual_PCR_conditions,
        eQual_phenotype,
        eQual_pmid,
        eQual_product,
        eQual_prot_desc,
        eQual_prot_note,
        eQual_protein_id,
        eQual_pseudo,
        eQual_PubMed,
        eQual_region_name,
        eQual_replace,
        eQual_rpt_family,
        eQual_rpt_type,
        eQual_rpt_unit,
        eQual_site_type,
        eQual_standard_name,
        eQual_transcript_id,
        eQual_transl_except,
        eQual_transl_table,
        eQual_translation,
        eQual_transposon,
        eQual_usedin
    };

    enum EOrgRef {
        eOrgRef_organism,
        eOrgRef_organelle,
        eOrgRef_div,
        eOrgRef_lineage,
        eOrgRef_gcode,
        eOrgRef_mgcode
    };

    typedef map< string, CSeqFeatData::ESubtype > TFeatReaderMap;
    typedef map< string, EQual > TQualReaderMap;
    typedef map< string, EOrgRef > TOrgRefReaderMap;
    typedef map< string, CBioSource::EGenome > TGenomeReaderMap;
    typedef map< string, CSubSource::ESubtype > TSubSrcReaderMap;
    typedef map< string, COrgMod::ESubtype > TOrgModReaderMap;
    typedef map< string, CSeqFeatData::EBond > TBondReaderMap;
    typedef map< string, CSeqFeatData::ESite > TSiteReaderMap;
    typedef map< string, int > TTrnaReaderMap;
    typedef vector< string > TSingleQualList;

    // constructor
    CFeature_table_reader_imp(void);
    // destructor
    ~CFeature_table_reader_imp(void);

    // read 5-column feature table and return Seq-annot
    CRef<CSeq_annot> ReadSequinFeatureTable (CNcbiIstream& ifs,
                                             const string& seqid,
                                             const string& annotname,
                                             const CFeature_table_reader::TFlags flags);

    // create single feature from key
    CRef<CSeq_feat> CreateSeqFeat (const string& feat,
                                   CSeq_loc& location,
                                   const CFeature_table_reader::TFlags flags);

    // add single qualifier to feature
    void AddFeatQual (CRef<CSeq_feat> sfp,
                      const string& qual,
                      const string& val,
                      const CFeature_table_reader::TFlags flags);

private:
    // Prohibit copy constructor and assignment operator
    CFeature_table_reader_imp(const CFeature_table_reader_imp& value);
    CFeature_table_reader_imp& operator=(const CFeature_table_reader_imp& value);

    bool x_ParseFeatureTableLine (const string& line, Int4* startP, Int4* stopP,
                                  bool* partial5P, bool* partial3P, bool* ispointP,
                                  string& featP, string& qualP, string& valP, Int4 offset);

    bool x_AddIntervalToFeature (CRef<CSeq_feat> sfp, CSeq_loc_mix *mix,
                                 const string& seqid, Int4 start,
                                 Int4 stop, bool partial5, bool partial3);

    bool x_AddQualifierToFeature (CRef<CSeq_feat> sfp,
                                  const string& qual, const string& val);

    bool x_AddQualifierToGene     (CSeqFeatData& sfdata,
                                   EQual qtype, const string& val);
    bool x_AddQualifierToCdregion (CRef<CSeq_feat> sfp, CSeqFeatData& sfdata,
                                   EQual qtype, const string& val);
    bool x_AddQualifierToRna      (CSeqFeatData& sfdata,
                                   EQual qtype, const string& val);
    bool x_AddQualifierToImp      (CRef<CSeq_feat> sfp, CSeqFeatData& sfdata,
                                   EQual qtype, const string& qual, const string& val);
    bool x_AddQualifierToBioSrc   (CSeqFeatData& sfdata,
                                   EOrgRef rtype, const string& val);
    bool x_AddQualifierToBioSrc   (CSeqFeatData& sfdata,
                                   CSubSource::ESubtype stype, const string& val);
    bool x_AddQualifierToBioSrc   (CSeqFeatData& sfdata,
                                   COrgMod::ESubtype mtype, const string& val);

    int x_ParseTrnaString (const string& val);

    TFeatReaderMap    m_FeatKeys;
    TQualReaderMap    m_QualKeys;
    TOrgRefReaderMap  m_OrgRefKeys;
    TGenomeReaderMap  m_GenomeKeys;
    TSubSrcReaderMap  m_SubSrcKeys;
    TOrgModReaderMap  m_OrgModKeys;
    TBondReaderMap    m_BondKeys;
    TSiteReaderMap    m_SiteKeys;
    TTrnaReaderMap    m_TrnaKeys;
    TSingleQualList   m_SingleKeys;
};

auto_ptr<CFeature_table_reader_imp> CFeature_table_reader::sm_Implementation;

void CFeature_table_reader::x_InitImplementation()
{
    DEFINE_STATIC_FAST_MUTEX(s_Implementation_mutex);

    CFastMutexGuard   LOCK(s_Implementation_mutex);
    if ( !sm_Implementation.get() ) {
        sm_Implementation.reset(new CFeature_table_reader_imp());
    }
}


typedef struct featinit {
    const char *           key;
    CSeqFeatData::ESubtype subtype;
} FeatInit;

static FeatInit feat_key_to_subtype [] = {
    { "-10_signal",         CSeqFeatData::eSubtype_10_signal          },
    { "-35_signal",         CSeqFeatData::eSubtype_35_signal          },
    { "3'clip",             CSeqFeatData::eSubtype_3clip              },
    { "3'UTR",              CSeqFeatData::eSubtype_3UTR               },
    { "5'clip",             CSeqFeatData::eSubtype_5clip              },
    { "5'UTR",              CSeqFeatData::eSubtype_5UTR               },
    { "attenuator",         CSeqFeatData::eSubtype_attenuator         },
    { "Bond",               CSeqFeatData::eSubtype_bond               },
    { "CAAT_signal",        CSeqFeatData::eSubtype_CAAT_signal        },
    { "CDS",                CSeqFeatData::eSubtype_cdregion           },
    { "Cit",                CSeqFeatData::eSubtype_pub                },
    { "Comment",            CSeqFeatData::eSubtype_comment            },
    { "conflict",           CSeqFeatData::eSubtype_conflict           },
    { "C_region",           CSeqFeatData::eSubtype_C_region           },
    { "D-loop",             CSeqFeatData::eSubtype_D_loop             },
    { "D_segment",          CSeqFeatData::eSubtype_D_segment          },
    { "enhancer",           CSeqFeatData::eSubtype_enhancer           },
    { "exon",               CSeqFeatData::eSubtype_exon               },
    { "GC_signal",          CSeqFeatData::eSubtype_GC_signal          },
    { "gene",               CSeqFeatData::eSubtype_gene               },
    { "Het",                CSeqFeatData::eSubtype_het                },
    { "iDNA",               CSeqFeatData::eSubtype_iDNA               },
    { "intron",             CSeqFeatData::eSubtype_intron             },
    { "J_segment",          CSeqFeatData::eSubtype_J_segment          },
    { "LTR",                CSeqFeatData::eSubtype_LTR                },
    { "mat_peptide",        CSeqFeatData::eSubtype_mat_peptide_aa     },
    { "mat_peptide_nt",     CSeqFeatData::eSubtype_mat_peptide        },
    { "misc_binding",       CSeqFeatData::eSubtype_misc_binding       },
    { "misc_difference",    CSeqFeatData::eSubtype_misc_difference    },
    { "misc_feature",       CSeqFeatData::eSubtype_misc_feature       },
    { "misc_recomb",        CSeqFeatData::eSubtype_misc_recomb        },
    { "misc_RNA",           CSeqFeatData::eSubtype_otherRNA           },
    { "misc_signal",        CSeqFeatData::eSubtype_misc_signal        },
    { "misc_structure",     CSeqFeatData::eSubtype_misc_structure     },
    { "modified_base",      CSeqFeatData::eSubtype_modified_base      },
    { "mRNA",               CSeqFeatData::eSubtype_mRNA               },
    { "NonStdRes",          CSeqFeatData::eSubtype_non_std_residue    },
    { "Num",                CSeqFeatData::eSubtype_num                },
    { "N_region",           CSeqFeatData::eSubtype_N_region           },
    { "old_sequence",       CSeqFeatData::eSubtype_old_sequence       },
    { "operon",             CSeqFeatData::eSubtype_operon             },
    { "oriT",               CSeqFeatData::eSubtype_oriT               },
    { "polyA_signal",       CSeqFeatData::eSubtype_polyA_signal       },
    { "polyA_site",         CSeqFeatData::eSubtype_polyA_site         },
    { "precursor_RNA",      CSeqFeatData::eSubtype_preRNA             },
    { "pre_RNA",            CSeqFeatData::eSubtype_preRNA             },
    { "preprotein",         CSeqFeatData::eSubtype_preprotein         },
    { "primer_bind",        CSeqFeatData::eSubtype_primer_bind        },
    { "prim_transcript",    CSeqFeatData::eSubtype_prim_transcript    },
    { "promoter",           CSeqFeatData::eSubtype_promoter           },
    { "Protein",            CSeqFeatData::eSubtype_prot               },
    { "protein_bind",       CSeqFeatData::eSubtype_protein_bind       },
    { "RBS",                CSeqFeatData::eSubtype_RBS                },
    { "REFERENCE",          CSeqFeatData::eSubtype_pub                },
    { "Region",             CSeqFeatData::eSubtype_region             },
    { "repeat_region",      CSeqFeatData::eSubtype_repeat_region      },
    { "repeat_unit",        CSeqFeatData::eSubtype_repeat_unit        },
    { "rep_origin",         CSeqFeatData::eSubtype_rep_origin         },
    { "rRNA",               CSeqFeatData::eSubtype_rRNA               },
    { "Rsite",              CSeqFeatData::eSubtype_rsite              },
    { "satellite",          CSeqFeatData::eSubtype_satellite          },
    { "scRNA",              CSeqFeatData::eSubtype_scRNA              },
    { "SecStr",             CSeqFeatData::eSubtype_psec_str           },
    { "sig_peptide",        CSeqFeatData::eSubtype_sig_peptide_aa     },
    { "sig_peptide_nt",     CSeqFeatData::eSubtype_sig_peptide        },
    { "Site",               CSeqFeatData::eSubtype_site               },
    { "Site-ref",           CSeqFeatData::eSubtype_site_ref           },
    { "snoRNA",             CSeqFeatData::eSubtype_snoRNA             },
    { "snRNA",              CSeqFeatData::eSubtype_snRNA              },
    { "source",             CSeqFeatData::eSubtype_biosrc             },
    { "Src",                CSeqFeatData::eSubtype_biosrc             },
    { "stem_loop",          CSeqFeatData::eSubtype_stem_loop          },
    { "STS",                CSeqFeatData::eSubtype_STS                },
    { "S_region",           CSeqFeatData::eSubtype_S_region           },
    { "TATA_signal",        CSeqFeatData::eSubtype_TATA_signal        },