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/serialization/base_object.hpp>
21 #include <boost/serialization/export.hpp>
22 #include <boost/serialization/list.hpp>
23 #include <boost/serialization/nvp.hpp>
24 #include <boost/serialization/string.hpp>
25 #include <boost/serialization/shared_ptr.hpp>
26 #include <boost/serialization/utility.hpp>
27 #include <boost/serialization/version.hpp>
28 #include <boost/serialization/vector.hpp>
30 #include <boost/shared_ptr.hpp>
31 #include <boost/enable_shared_from_this.hpp>
35 #include "alphabet.hpp"
37 #include "seq_span.hpp"
39 // Sequence data class
41 /* The way that motifs are found currently doesn't really
42 * indicate that the match was a reverse compliment
47 //motif(int begin, int end, std::string type, std::string name);
48 //! this constructor is for when we're adding motifs to our annotations
49 motif(int begin, std::string motif);
58 friend bool operator==(const motif& left, const motif& right);
60 // boost::serialization support
62 friend class boost::serialization::access;
63 template<class Archive>
64 void serialize(Archive& ar, const unsigned int /*version*/) {
65 ar & BOOST_SERIALIZATION_NVP(begin);
66 ar & BOOST_SERIALIZATION_NVP(end);
67 ar & BOOST_SERIALIZATION_NVP(type);
68 ar & BOOST_SERIALIZATION_NVP(name);
69 ar & BOOST_SERIALIZATION_NVP(sequence);
72 BOOST_CLASS_EXPORT(motif);
75 typedef boost::shared_ptr<Sequence> SequenceRef;
77 //! sequence track for mussa.
83 typedef SeqString::value_type value_type;
84 typedef SeqString::difference_type difference_type;
85 typedef SeqString::iterator iterator;
86 typedef SeqString::reverse_iterator reverse_iterator;
87 typedef SeqString::const_iterator const_iterator;
88 typedef SeqString::const_reverse_iterator const_reverse_iterator;
89 typedef SeqString::reference reference;
90 typedef SeqString::const_reference const_reference;
91 typedef SeqString::size_type size_type;
92 static const size_type npos = SeqString::npos;
94 typedef std::list<motif> MotifList;
95 typedef boost::shared_ptr<MotifList> MotifListRef;
97 Sequence(AlphabetRef a = reduced_dna_alphabet);
98 Sequence(const char* seq,
99 AlphabetRef a = reduced_dna_alphabet,
100 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
101 Sequence(const std::string& seq,
102 AlphabetRef a = reduced_dna_alphabet,
103 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
104 //! make a new sequence, with the same SeqSpan
105 Sequence(const Sequence& seq);
106 //! make a new sequence, with the same SeqSpan
107 Sequence(const Sequence *);
108 //! Make a new sequence using a copy of SeqSpan
109 Sequence(const SequenceRef);
110 Sequence(const SeqSpanRef&);
112 //! assignment to constant sequences
113 Sequence &operator=(const Sequence&);
115 friend std::ostream& operator<<(std::ostream&, const Sequence&);
116 friend bool operator<(const Sequence&, const Sequence&);
117 friend bool operator==(const Sequence&, const Sequence&);
118 friend bool operator!=(const Sequence&, const Sequence&);
119 const_reference operator[](size_type) const;
121 //! set sequence to a (sub)string containing nothing but AGCTN
122 void set_filtered_sequence(const std::string& seq,
123 AlphabetRef a=reduced_dna_alphabet,
125 size_type count=npos,
126 SeqSpan::strand_type strand=SeqSpan::PlusStrand);
128 //! retrive element at specific position
129 const_reference at(size_type i) const { return seq->at(i); }
130 //! clear the sequence and its annotations
132 //! return a non-null terminated c pointer to the sequence data
133 const char *data() const { return seq->data(); }
135 const_iterator begin() const { return seq->begin(); }
137 const_iterator end() const { return seq->end(); }
138 //! is our sequence empty?
139 bool empty() const { return (seq) ? seq->empty() : true ; }
141 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
142 //! how many base pairs are there in our sequence
143 size_type size() const { return (seq) ? seq->size() : 0; }
144 //! alias for size (used by string)
145 size_type length() const { return size(); }
147 const_reverse_iterator rbegin() const { return seq->rbegin(); }
148 //! reverse end iterator
149 const_reverse_iterator rend() const { return seq->rend(); }
150 //! is our sequence empty?
151 //! start position relative to "base" sequence
152 size_type start() const { return seq->parentStart(); }
153 //! one past the last position relative to "base" sequence
154 size_type stop() const { return seq->parentStop(); }
156 //! return a subsequence, copying over any appropriate annotation
157 Sequence subseq(size_type start=0,
158 size_type count = npos,
159 SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
160 //! reverse a character
161 std::string create_reverse_map() const;
162 //! return a reverse compliment (this needs to be improved?)
163 Sequence rev_comp() const;
165 //! set sequence (filtered)
166 void set_sequence(const std::string &, AlphabetRef);
168 std::string get_sequence() const;
170 void set_species(const std::string &);
172 std::string get_species() const;
173 //! set the fasta header
174 void set_fasta_header(std::string header);
175 //! get the fasta header
176 std::string get_fasta_header() const;
177 //! get name (will return the first non-empty, of fasta_header, species)
178 std::string get_name() const;
179 //! return a reference to whichever alphabet we're currently representing
180 const Alphabet& get_alphabet() const;
182 //! load sequence from fasta file using the sequences current alphabet
183 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
184 int start_index=0, int end_index=0);
185 //! load sequence AGCT from fasta file
186 //! \throw mussa_load_error
187 //! \throw sequence_empty_error
188 //! \throw sequence_empty_file_error
189 void load_fasta(const boost::filesystem::path file_path,
192 int start_index=0, int end_index=0);
193 void load_fasta(std::istream& file,
194 int seq_num=1, int start_index=0, int end_index=0);
195 //! load sequence from stream
196 //! \throw mussa_load_error
197 //! \throw sequence_empty_error
198 //! \throw sequence_empty_file_error
199 void load_fasta(std::istream& file,
202 int start_index=0, int end_index=0);
203 //! load sequence annotations
204 //! \throws mussa_load_error
205 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
206 //! load sequence annotations
207 //! \throws mussa_load_error
208 void load_annot(std::istream& data_stream, int start_index, int end_index);
209 //! parse annotation file
210 /*! \throws annotation_load_error
212 void parse_annot(std::string data, int start_index=0, int end_index=0);
213 //! add an annotation to our list of annotations
214 void add_annotation(const SeqSpanRef a);
215 //! add an annotation using tristan's mussa file paramenters
216 void add_annotation(std::string name, std::string type, size_type start, size_type stop);
217 //! create an initialized annotation with the "standard" types.
218 SeqSpanRef make_annotation(std::string name, std::string type, size_type start, size_type stop) const;
219 const SeqSpanRefList& annotations() const;
221 const MotifList& motifs() const;
223 //! add a motif to our list of motifs
224 void add_motif(const Sequence& a_motif);
225 //! clear our list of found motifs
227 //! search a sequence for a_motif
228 //! \throws motif_normalize_error if there's something wrong with a_motif
229 std::vector<int> find_motif(const Sequence& a_motif) const;
230 //! annotate the current sequence with other sequences
231 void find_sequences(std::list<Sequence>::iterator start,
232 std::list<Sequence>::iterator end);
233 SeqSpanRef seqspan() { return seq; }
235 void save(boost::filesystem::fstream &save_file);
236 //void load_museq(boost::filesystem::path load_file_path, int seq_num);
237 static SequenceRef load_museq(boost::filesystem::fstream& load_file);
246 //! store annotation regions
247 SeqSpanRefListRef annotation_list;
248 //! a seperate list for motifs since we're currently not saving them
249 MotifListRef motif_list;
251 //! copy over all our annotation children
252 void copy_children(Sequence &, size_type start, size_type count) const;
254 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
255 std::string rc_motif(std::string a_motif) const;
256 //! look for a string sequence type and and it to an annotation list
257 void add_string_annotation(std::string a_seq, std::string name);
259 // boost::serialization support
260 friend class boost::serialization::access;
261 template<class Archive>
262 void serialize(Archive& ar, const unsigned int /*version*/) {
263 ar & BOOST_SERIALIZATION_NVP(seq);
264 ar & BOOST_SERIALIZATION_NVP(header);
265 ar & BOOST_SERIALIZATION_NVP(species);
266 ar & BOOST_SERIALIZATION_NVP(annotation_list);
267 ar & BOOST_SERIALIZATION_NVP(motif_list);
271 //BOOST_CLASS_EXPORT(Sequence);
272 //BOOST_CLASS_EXPORT_GUID(Sequence, "Sequence")