1 #ifndef _MUSSA_SEQUENCE_H_
2 #define _MUSSA_SEQUENCE_H_
3 // This file is part of the Mussa source distribution.
4 // http://mussa.caltech.edu/
5 // Contact author: Tristan De Buysscher, tristan@caltech.edu
7 // This program and all associated source code files are Copyright (C) 2005
8 // the California Institute of Technology, Pasadena, CA, 91125 USA. It is
9 // under the GNU Public License; please see the included LICENSE.txt
10 // file for more information, or contact Tristan directly.
13 // ----------------------------------------
14 // ---------- sequence.hh -----------
15 // ----------------------------------------
17 #include <boost/filesystem/path.hpp>
18 #include <boost/filesystem/fstream.hpp>
20 #include <boost/shared_ptr.hpp>
21 #include <boost/enable_shared_from_this.hpp>
25 #include "alphabet.hpp"
27 #include "seq_span.hpp"
29 // Sequence data class
31 /* The way that motifs are found currently doesn't really
32 * indicate that the match was a reverse compliment
37 //motif(int begin, int end, std::string type, std::string name);
38 //! this constructor is for when we're adding motifs to our annotations
39 motif(int begin, std::string motif);
48 friend bool operator==(const motif& left, const motif& right);
50 // boost::serialization support
52 friend class boost::serialization::access;
53 template<class Archive>
54 void serialize(Archive& ar, const unsigned int /*version*/) {
55 ar & BOOST_SERIALIZATION_NVP(begin);
56 ar & BOOST_SERIALIZATION_NVP(end);
57 ar & BOOST_SERIALIZATION_NVP(type);
58 ar & BOOST_SERIALIZATION_NVP(name);
59 ar & BOOST_SERIALIZATION_NVP(sequence);
64 typedef boost::shared_ptr<Sequence> SequenceRef;
66 //! sequence track for mussa.
72 typedef SeqString::value_type value_type;
73 typedef SeqString::difference_type difference_type;
74 typedef SeqString::iterator iterator;
75 typedef SeqString::reverse_iterator reverse_iterator;
76 typedef SeqString::const_iterator const_iterator;
77 typedef SeqString::const_reverse_iterator const_reverse_iterator;
78 typedef SeqString::reference reference;
79 typedef SeqString::const_reference const_reference;
80 typedef SeqString::size_type size_type;
81 static const size_type npos = SeqString::npos;
83 typedef std::list<motif> MotifList;
84 typedef boost::shared_ptr<MotifList> MotifListRef;
86 Sequence(AlphabetRef a = reduced_dna_alphabet);
87 Sequence(const char* seq,
88 AlphabetRef a = reduced_dna_alphabet,
89 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
90 Sequence(const std::string& seq,
91 AlphabetRef a = reduced_dna_alphabet,
92 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
93 //! make a new sequence, with the same SeqSpan
94 Sequence(const Sequence& seq);
95 //! make a new sequence, with the same SeqSpan
96 Sequence(const Sequence *);
97 //! Make a new sequence using a copy of SeqSpan
98 Sequence(const SequenceRef);
99 Sequence(const SeqSpanRef&);
101 //! assignment to constant sequences
102 Sequence &operator=(const Sequence&);
104 friend std::ostream& operator<<(std::ostream&, const Sequence&);
105 friend bool operator<(const Sequence&, const Sequence&);
106 friend bool operator==(const Sequence&, const Sequence&);
107 friend bool operator!=(const Sequence&, const Sequence&);
108 const_reference operator[](size_type) const;
110 //! set sequence to a (sub)string containing nothing but AGCTN
111 void set_filtered_sequence(const std::string& seq,
112 AlphabetRef a=reduced_dna_alphabet,
114 size_type count=npos,
115 SeqSpan::strand_type strand=SeqSpan::PlusStrand);
117 //! retrive element at specific position
118 const_reference at(size_type i) const { return seq->at(i); }
119 //! clear the sequence and its annotations
121 //! return a non-null terminated c pointer to the sequence data
122 const char *data() const { return seq->data(); }
124 const_iterator begin() const { return seq->begin(); }
126 const_iterator end() const { return seq->end(); }
127 //! is our sequence empty?
128 bool empty() const { return (seq) ? seq->empty() : true ; }
130 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
131 //! how many base pairs are there in our sequence
132 size_type size() const { return (seq) ? seq->size() : 0; }
133 //! alias for size (used by string)
134 size_type length() const { return size(); }
136 const_reverse_iterator rbegin() const { return seq->rbegin(); }
137 //! reverse end iterator
138 const_reverse_iterator rend() const { return seq->rend(); }
139 //! is our sequence empty?
140 //! start position relative to "base" sequence
141 size_type start() const { return seq->parentStart(); }
142 //! one past the last position relative to "base" sequence
143 size_type stop() const { return seq->parentStop(); }
145 //! return a subsequence, copying over any appropriate annotation
146 Sequence subseq(size_type start=0,
147 size_type count = npos,
148 SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
149 //! reverse a character
150 std::string create_reverse_map() const;
151 //! return a reverse compliment (this needs to be improved?)
152 Sequence rev_comp() const;
154 //! set sequence (filtered)
155 void set_sequence(const std::string &, AlphabetRef);
157 std::string get_sequence() const;
159 void set_species(const std::string &);
161 std::string get_species() const;
162 //! set the fasta header
163 void set_fasta_header(std::string header);
164 //! get the fasta header
165 std::string get_fasta_header() const;
166 //! get name (will return the first non-empty, of fasta_header, species)
167 std::string get_name() const;
168 //! return a reference to whichever alphabet we're currently representing
169 const Alphabet& get_alphabet() const;
171 //! load sequence from fasta file using the sequences current alphabet
172 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
173 int start_index=0, int end_index=0);
174 //! load sequence AGCT from fasta file
175 //! \throw mussa_load_error
176 //! \throw sequence_empty_error
177 //! \throw sequence_empty_file_error
178 void load_fasta(const boost::filesystem::path file_path,
181 int start_index=0, int end_index=0);
182 void load_fasta(std::istream& file,
183 int seq_num=1, int start_index=0, int end_index=0);
184 //! load sequence from stream
185 //! \throw mussa_load_error
186 //! \throw sequence_empty_error
187 //! \throw sequence_empty_file_error
188 void load_fasta(std::istream& file,
191 int start_index=0, int end_index=0);
192 //! load sequence annotations
193 //! \throws mussa_load_error
194 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
195 //! load sequence annotations
196 //! \throws mussa_load_error
197 void load_annot(std::istream& data_stream, int start_index, int end_index);
198 //! parse annotation file
199 /*! \throws annotation_load_error
201 void parse_annot(std::string data, int start_index=0, int end_index=0);
202 //! add an annotation to our list of annotations
203 void add_annotation(const SeqSpanRef a);
204 //! add an annotation using tristan's mussa file paramenters
205 void add_annotation(std::string name, std::string type, size_type start, size_type stop);
206 //! create an initialized annotation with the "standard" types.
207 SeqSpanRef make_annotation(std::string name, std::string type, size_type start, size_type stop) const;
208 const SeqSpanRefList& annotations() const;
210 const MotifList& motifs() const;
212 //! add a motif to our list of motifs
213 void add_motif(const Sequence& a_motif);
214 //! clear our list of found motifs
216 //! search a sequence for a_motif
217 //! \throws motif_normalize_error if there's something wrong with a_motif
218 std::vector<int> find_motif(const Sequence& a_motif) const;
219 //! annotate the current sequence with other sequences
220 void find_sequences(std::list<Sequence>::iterator start,
221 std::list<Sequence>::iterator end);
222 SeqSpanRef seqspan() { return seq; }
224 void save(boost::filesystem::fstream &save_file);
225 //void load_museq(boost::filesystem::path load_file_path, int seq_num);
226 static SequenceRef load_museq(boost::filesystem::fstream& load_file);
235 //! store annotation regions
236 SeqSpanRefListRef annotation_list;
237 //! a seperate list for motifs since we're currently not saving them
238 MotifListRef motif_list;
240 //! copy over all our annotation children
241 void copy_children(Sequence &, size_type start, size_type count) const;
243 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
244 std::string rc_motif(std::string a_motif) const;
245 //! look for a string sequence type and and it to an annotation list
246 void add_string_annotation(std::string a_seq, std::string name);