#include <boost/filesystem/path.hpp>
#include <boost/filesystem/fstream.hpp>
-#include <list>
-#include <string>
-#include <vector>
+
+#include <boost/serialization/base_object.hpp>
+#include <boost/serialization/export.hpp>
+#include <boost/serialization/list.hpp>
+#include <boost/serialization/nvp.hpp>
+#include <boost/serialization/string.hpp>
+#include <boost/serialization/shared_ptr.hpp>
+#include <boost/serialization/utility.hpp>
+#include <boost/serialization/version.hpp>
+#include <boost/serialization/vector.hpp>
+
+#include <boost/shared_ptr.hpp>
+#include <boost/enable_shared_from_this.hpp>
+
#include <iostream>
-// Sequence data class
+#include "alphabet.hpp"
+#include "seq.hpp"
+#include "seq_span.hpp"
-//! Attach annotation information to a sequence track
-struct annot
-{
- annot();
- annot(int start, int end, std::string type, std::string name);
-
- int start;
- int end;
- std::string type;
- std::string name;
-};
+// Sequence data class
/* The way that motifs are found currently doesn't really
* indicate that the match was a reverse compliment
*/
-struct motif : public annot
+struct motif
{
+ motif();
+ //motif(int begin, int end, std::string type, std::string name);
+ //! this constructor is for when we're adding motifs to our annotations
+ motif(int begin, std::string motif);
+ ~motif();
+
+ int begin;
+ int end;
+ std::string type;
+ std::string name;
std::string sequence;
- //! this constructor is for when we're adding motifs to our annotations
- motif(int start, std::string motif);
+ friend bool operator==(const motif& left, const motif& right);
+
+ // boost::serialization support
+private:
+ friend class boost::serialization::access;
+ template<class Archive>
+ void serialize(Archive& ar, const unsigned int /*version*/) {
+ ar & BOOST_SERIALIZATION_NVP(begin);
+ ar & BOOST_SERIALIZATION_NVP(end);
+ ar & BOOST_SERIALIZATION_NVP(type);
+ ar & BOOST_SERIALIZATION_NVP(name);
+ ar & BOOST_SERIALIZATION_NVP(sequence);
+ }
};
+BOOST_CLASS_EXPORT(motif);
+
+class Sequence;
+typedef boost::shared_ptr<Sequence> SequenceRef;
//! sequence track for mussa.
-class Sequence
+class Sequence
{
- friend class ConnView;
- friend class SeqView;
- private:
- std::string sequence;
- std::string header;
- std::string species;
-
- std::list<annot> annots;
- //! a seperate list for motifs since we're currently not saving them
- std::list<motif> motif_list;
-
- void motif_scan(std::string a_motif, std::vector<int> * motif_match_starts);
- std::string rc_motif(std::string a_motif);
- //! look for a string sequence type and and it to an annotation list
- void add_string_annotation(std::string a_seq, std::string name);
-
- public:
- typedef std::string::iterator iterator;
- typedef std::string::const_iterator const_iterator;
- typedef std::string::size_type size_type;
-
- Sequence();
- Sequence(std::string seq);
- //! assignment to constant sequences
- Sequence &operator=(const Sequence&);
- Sequence &operator=(const std::string &);
- char operator[](int) const;
- friend std::ostream& operator<<(std::ostream& out, const Sequence& seq);
-
- //! set sequence to a (sub)string containing nothing but AGCTN
- void set_filtered_sequence(const std::string& seq,
- std::string::size_type start=0,
- std::string::size_type count=0);
-
- //! load sequence AGCT from fasta file
- /*! \throws mussa_load_error
- */
- void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
- int start_index=0, int end_index=0);
- //! load sequence annotations
- //! \throws mussa_load_error
- void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
- //! load sequence annotations
- //! \throws mussa_load_error
- //void load_annot(std::istream& data_stream, int start_index, int end_index);
- void parse_annot(std::string data, int start_index, int end_index);
- const std::list<annot>& annotations() const;
- const std::list<motif>& motifs() const;
- const std::string& get_species() const;
-
- // simple access functions
- void set_seq(const std::string& a_seq);
- const std::string& get_seq() const;
- const char *c_seq() const;
- std::string subseq(int start, int end) const;
- std::string rev_comp() const;
-
- // string-like functions
- iterator begin();
- const_iterator begin() const;
- iterator end();
- const_iterator end() const;
- bool empty() const;
- //! alias for size, (included as this is much like a string)
- std::string::size_type length() const;
- //! the number of base pairs in this sequence
- std::string::size_type size() const;
- void clear();
-
- void set_header(std::string header);
- std::string get_header() const;
- //! add a motif to our list of motifs
- //! \throws motif_normalize_error if there's something wrong with a_motif
- void add_motif(std::string a_motif);
- //! clear our list of found motifs
- void clear_motifs();
- //! search a sequence for a_motif
- //! \throws motif_normalize_error if there's something wrong with a_motif
- std::vector<int> find_motif(std::string a_motif);
- //! convert IUPAC symbols to upperase
- //! \throws motif_normalize_error if there is an invalid symbol
- static std::string motif_normalize(std::string a_motif);
-
- //! annotate the current sequence with other sequences
- void find_sequences(std::list<Sequence>::iterator start,
- std::list<Sequence>::iterator end);
-
- void save(boost::filesystem::fstream &save_file);
- void load_museq(boost::filesystem::path load_file_path, int seq_num);
+public:
+ typedef SeqString::value_type value_type;
+ typedef SeqString::difference_type difference_type;
+ typedef SeqString::iterator iterator;
+ typedef SeqString::reverse_iterator reverse_iterator;
+ typedef SeqString::const_iterator const_iterator;
+ typedef SeqString::const_reverse_iterator const_reverse_iterator;
+ typedef SeqString::reference reference;
+ typedef SeqString::const_reference const_reference;
+ typedef SeqString::size_type size_type;
+ static const size_type npos = SeqString::npos;
+
+ typedef std::list<motif> MotifList;
+ typedef boost::shared_ptr<MotifList> MotifListRef;
+
+ Sequence(AlphabetRef a = reduced_dna_alphabet);
+ Sequence(const char* seq,
+ AlphabetRef a = reduced_dna_alphabet,
+ SeqSpan::strand_type strand = SeqSpan::PlusStrand);
+ Sequence(const std::string& seq,
+ AlphabetRef a = reduced_dna_alphabet,
+ SeqSpan::strand_type strand = SeqSpan::PlusStrand);
+ //! make a new sequence, with the same SeqSpan
+ Sequence(const Sequence& seq);
+ //! make a new sequence, with the same SeqSpan
+ Sequence(const Sequence *);
+ //! Make a new sequence using a copy of SeqSpan
+ Sequence(const SequenceRef);
+ Sequence(const SeqSpanRef&);
+ ~Sequence();
+ //! assignment to constant sequences
+ Sequence &operator=(const Sequence&);
+
+ friend std::ostream& operator<<(std::ostream&, const Sequence&);
+ friend bool operator<(const Sequence&, const Sequence&);
+ friend bool operator==(const Sequence&, const Sequence&);
+ friend bool operator!=(const Sequence&, const Sequence&);
+ const_reference operator[](size_type) const;
+
+ //! set sequence to a (sub)string containing nothing but AGCTN
+ void set_filtered_sequence(const std::string& seq,
+ AlphabetRef a=reduced_dna_alphabet,
+ size_type start=0,
+ size_type count=npos,
+ SeqSpan::strand_type strand=SeqSpan::PlusStrand);
+
+ //! retrive element at specific position
+ const_reference at(size_type i) const { return seq->at(i); }
+ //! clear the sequence and its annotations
+ void clear();
+ //! return a non-null terminated c pointer to the sequence data
+ const char *data() const { return seq->data(); }
+ //! forward iterator
+ const_iterator begin() const { return seq->begin(); }
+ //! last iterator
+ const_iterator end() const { return seq->end(); }
+ //! is our sequence empty?
+ bool empty() const { return (seq) ? seq->empty() : true ; }
+ //! find first
+ size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
+ //! how many base pairs are there in our sequence
+ size_type size() const { return (seq) ? seq->size() : 0; }
+ //! alias for size (used by string)
+ size_type length() const { return size(); }
+ //! reverse iterator
+ const_reverse_iterator rbegin() const { return seq->rbegin(); }
+ //! reverse end iterator
+ const_reverse_iterator rend() const { return seq->rend(); }
+ //! is our sequence empty?
+ //! start position relative to "base" sequence
+ size_type start() const { return seq->parentStart(); }
+ //! one past the last position relative to "base" sequence
+ size_type stop() const { return seq->parentStop(); }
+
+ //! return a subsequence, copying over any appropriate annotation
+ Sequence subseq(size_type start=0,
+ size_type count = npos,
+ SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
+ //! reverse a character
+ std::string create_reverse_map() const;
+ //! return a reverse compliment (this needs to be improved?)
+ Sequence rev_comp() const;
+
+ //! set sequence (filtered)
+ void set_sequence(const std::string &, AlphabetRef);
+ //! get sequence
+ std::string get_sequence() const;
+ //! set species name
+ void set_species(const std::string &);
+ //! get species name
+ std::string get_species() const;
+ //! set the fasta header
+ void set_fasta_header(std::string header);
+ //! get the fasta header
+ std::string get_fasta_header() const;
+ //! get name (will return the first non-empty, of fasta_header, species)
+ std::string get_name() const;
+ //! return a reference to whichever alphabet we're currently representing
+ const Alphabet& get_alphabet() const;
+
+ //! load sequence from fasta file using the sequences current alphabet
+ void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
+ int start_index=0, int end_index=0);
+ //! load sequence AGCT from fasta file
+ //! \throw mussa_load_error
+ //! \throw sequence_empty_error
+ //! \throw sequence_empty_file_error
+ void load_fasta(const boost::filesystem::path file_path,
+ AlphabetRef a,
+ int seq_num=1,
+ int start_index=0, int end_index=0);
+ void load_fasta(std::istream& file,
+ int seq_num=1, int start_index=0, int end_index=0);
+ //! load sequence from stream
+ //! \throw mussa_load_error
+ //! \throw sequence_empty_error
+ //! \throw sequence_empty_file_error
+ void load_fasta(std::istream& file,
+ AlphabetRef a,
+ int seq_num=1,
+ int start_index=0, int end_index=0);
+ //! load sequence annotations
+ //! \throws mussa_load_error
+ void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
+ //! load sequence annotations
+ //! \throws mussa_load_error
+ void load_annot(std::istream& data_stream, int start_index, int end_index);
+ //! parse annotation file
+ /*! \throws annotation_load_error
+ */
+ void parse_annot(std::string data, int start_index=0, int end_index=0);
+ //! add an annotation to our list of annotations
+ void add_annotation(const SeqSpanRef a);
+ //! add an annotation using tristan's mussa file paramenters
+ void add_annotation(std::string name, std::string type, size_type start, size_type stop);
+ //! create an initialized annotation with the "standard" types.
+ SeqSpanRef make_annotation(std::string name, std::string type, size_type start, size_type stop) const;
+ const SeqSpanRefList& annotations() const;
+
+ const MotifList& motifs() const;
+
+ //! add a motif to our list of motifs
+ void add_motif(const Sequence& a_motif);
+ //! clear our list of found motifs
+ void clear_motifs();
+ //! search a sequence for a_motif
+ //! \throws motif_normalize_error if there's something wrong with a_motif
+ std::vector<int> find_motif(const Sequence& a_motif) const;
+ //! annotate the current sequence with other sequences
+ void find_sequences(std::list<Sequence>::iterator start,
+ std::list<Sequence>::iterator end);
+ SeqSpanRef seqspan() { return seq; }
+
+ void save(boost::filesystem::fstream &save_file);
+ //void load_museq(boost::filesystem::path load_file_path, int seq_num);
+ static SequenceRef load_museq(boost::filesystem::fstream& load_file);
+
+protected:
+ SeqSpanRef seq;
+ //! fasta header
+ std::string header;
+ //! species name
+ std::string species;
+
+ //! store annotation regions
+ SeqSpanRefListRef annotation_list;
+ //! a seperate list for motifs since we're currently not saving them
+ MotifListRef motif_list;
+
+ //! copy over all our annotation children
+ void copy_children(Sequence &, size_type start, size_type count) const;
+
+ void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
+ std::string rc_motif(std::string a_motif) const;
+ //! look for a string sequence type and and it to an annotation list
+ void add_string_annotation(std::string a_seq, std::string name);
+
+ // boost::serialization support
+ friend class boost::serialization::access;
+ template<class Archive>
+ void serialize(Archive& ar, const unsigned int /*version*/) {
+ ar & BOOST_SERIALIZATION_NVP(seq);
+ ar & BOOST_SERIALIZATION_NVP(header);
+ ar & BOOST_SERIALIZATION_NVP(species);
+ ar & BOOST_SERIALIZATION_NVP(annotation_list);
+ ar & BOOST_SERIALIZATION_NVP(motif_list);
+ }
};
+BOOST_CLASS_EXPORT(Sequence);
#endif