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 //! Attach annotation information to a sequence track
45 annot(int begin, int end, std::string type, std::string name);
53 friend bool operator==(const annot& left, const annot& right);
55 // boost::serialization support
56 friend class boost::serialization::access;
57 template<class Archive>
58 void serialize(Archive& ar, const unsigned int /*version*/) {
59 ar & BOOST_SERIALIZATION_NVP(begin);
60 ar & BOOST_SERIALIZATION_NVP(end);
61 ar & BOOST_SERIALIZATION_NVP(type);
62 ar & BOOST_SERIALIZATION_NVP(name);
65 BOOST_CLASS_EXPORT(annot);
68 /* The way that motifs are found currently doesn't really
69 * indicate that the match was a reverse compliment
71 struct motif : public annot
75 motif() : annot(), sequence("") {};
76 //! this constructor is for when we're adding motifs to our annotations
77 motif(int begin, std::string motif);
80 // boost::serialization support
82 friend class boost::serialization::access;
83 template<class Archive>
84 void serialize(Archive& ar, const unsigned int /*version*/) {
85 ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(annot);
86 ar & BOOST_SERIALIZATION_NVP(sequence);
89 BOOST_CLASS_EXPORT(motif);
92 typedef boost::shared_ptr<Sequence> SequenceRef;
94 //! sequence track for mussa.
98 typedef SeqString::value_type value_type;
99 typedef SeqString::difference_type difference_type;
100 typedef SeqString::iterator iterator;
101 typedef SeqString::reverse_iterator reverse_iterator;
102 typedef SeqString::const_iterator const_iterator;
103 typedef SeqString::const_reverse_iterator const_reverse_iterator;
104 typedef SeqString::reference reference;
105 typedef SeqString::const_reference const_reference;
106 typedef SeqString::size_type size_type;
107 static const size_type npos = SeqString::npos;
109 Sequence(AlphabetRef a = reduced_dna_alphabet);
110 Sequence(const char* seq,
111 AlphabetRef a = reduced_dna_alphabet,
112 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
113 Sequence(const std::string& seq,
114 AlphabetRef a = reduced_dna_alphabet,
115 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
116 //! make a new sequence, with the same SeqSpan
117 Sequence(const Sequence& seq);
118 //! make a new sequence, with the same SeqSpan
119 Sequence(const Sequence *);
120 //! Make a new sequence using a copy of SeqSpan
121 Sequence(const SequenceRef);
122 Sequence(const SeqSpanRef&);
124 //! assignment to constant sequences
125 Sequence &operator=(const Sequence&);
127 friend std::ostream& operator<<(std::ostream&, const Sequence&);
128 friend bool operator<(const Sequence&, const Sequence&);
129 friend bool operator==(const Sequence&, const Sequence&);
130 friend bool operator!=(const Sequence&, const Sequence&);
131 const_reference operator[](size_type) const;
133 //! set sequence to a (sub)string containing nothing but AGCTN
134 void set_filtered_sequence(const std::string& seq,
135 AlphabetRef a=reduced_dna_alphabet,
137 size_type count=npos,
138 SeqSpan::strand_type strand=SeqSpan::PlusStrand);
140 //! retrive element at specific position
141 const_reference at(size_type i) const { return seq->at(i); }
142 //! clear the sequence and its annotations
144 //! return a non-null terminated c pointer to the sequence data
145 const char *data() const { return seq->data(); }
147 const_iterator begin() const { return seq->begin(); }
149 const_iterator end() const { return seq->end(); }
150 //! is our sequence empty?
151 bool empty() const { return (seq) ? seq->empty() : true ; }
153 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
154 //! how many base pairs are there in our sequence
155 size_type size() const { return (seq) ? seq->size() : 0; }
156 //! alias for size (used by string)
157 size_type length() const { return size(); }
159 const_reverse_iterator rbegin() const { return seq->rbegin(); }
160 //! reverse end iterator
161 const_reverse_iterator rend() const { return seq->rend(); }
162 //! is our sequence empty?
163 //! start position relative to "base" sequence
164 size_type start() const { return seq->parentStart(); }
165 //! one past the last position relative to "base" sequence
166 size_type stop() const { return seq->parentStop(); }
168 //! return a subsequence, copying over any appropriate annotation
169 Sequence subseq(size_type start=0,
170 size_type count = npos,
171 SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
172 //! reverse a character
173 std::string create_reverse_map() const;
174 //! return a reverse compliment (this needs to be improved?)
175 Sequence rev_comp() const;
177 //! set sequence (filtered)
178 void set_sequence(const std::string &, AlphabetRef);
180 std::string get_sequence() const;
182 void set_species(const std::string &);
184 std::string get_species() const;
185 //! set the fasta header
186 void set_fasta_header(std::string header);
187 //! get the fasta header
188 std::string get_fasta_header() const;
189 //! get name (will return the first non-empty, of fasta_header, species)
190 std::string get_name() const;
191 //! return a reference to whichever alphabet we're currently representing
192 const Alphabet& get_alphabet() const;
194 //! load sequence from fasta file using the sequences current alphabet
195 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
196 int start_index=0, int end_index=0);
197 //! load sequence AGCT from fasta file
198 //! \throw mussa_load_error
199 //! \throw sequence_empty_error
200 //! \throw sequence_empty_file_error
201 void load_fasta(const boost::filesystem::path file_path,
204 int start_index=0, int end_index=0);
205 void load_fasta(std::istream& file,
206 int seq_num=1, int start_index=0, int end_index=0);
207 //! load sequence from stream
208 //! \throw mussa_load_error
209 //! \throw sequence_empty_error
210 //! \throw sequence_empty_file_error
211 void load_fasta(std::istream& file,
214 int start_index=0, int end_index=0);
215 //! load sequence annotations
216 //! \throws mussa_load_error
217 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
218 //! load sequence annotations
219 //! \throws mussa_load_error
220 void load_annot(std::fstream& data_stream, int start_index, int end_index);
221 //! parse annotation file
222 /*! \throws annotation_load_error
224 void parse_annot(std::string data, int start_index=0, int end_index=0);
225 //! add an annotation to our list of annotations
226 void add_annotation(const annot& a);
227 const std::list<annot>& annotations() const;
228 const std::list<motif>& motifs() const;
230 //! add a motif to our list of motifs
231 void add_motif(const Sequence& a_motif);
232 //! clear our list of found motifs
234 //! search a sequence for a_motif
235 //! \throws motif_normalize_error if there's something wrong with a_motif
236 std::vector<int> find_motif(const Sequence& a_motif) const;
237 //! annotate the current sequence with other sequences
238 void find_sequences(std::list<Sequence>::iterator start,
239 std::list<Sequence>::iterator end);
241 void save(boost::filesystem::fstream &save_file);
242 void load_museq(boost::filesystem::path load_file_path, int seq_num);
251 //! store our oldstyle annotations
252 std::list<annot> annots;
253 //! a seperate list for motifs since we're currently not saving them
254 std::list<motif> motif_list;
256 //! copy over all our annotation children
257 void copy_children(Sequence &, size_type start, size_type count) const;
259 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
260 std::string rc_motif(std::string a_motif) const;
261 //! look for a string sequence type and and it to an annotation list
262 void add_string_annotation(std::string a_seq, std::string name);
264 // boost::serialization support
265 friend class boost::serialization::access;
266 template<class Archive>
267 void serialize(Archive& ar, const unsigned int /*version*/) {
268 ar & BOOST_SERIALIZATION_NVP(seq);
269 ar & BOOST_SERIALIZATION_NVP(header);
270 ar & BOOST_SERIALIZATION_NVP(species);
271 ar & BOOST_SERIALIZATION_NVP(annots);
272 ar & BOOST_SERIALIZATION_NVP(motif_list);
275 BOOST_CLASS_EXPORT(Sequence);