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 //! sequence track for mussa.
96 typedef SeqString::value_type value_type;
97 typedef SeqString::difference_type difference_type;
98 typedef SeqString::iterator iterator;
99 typedef SeqString::reverse_iterator reverse_iterator;
100 typedef SeqString::const_iterator const_iterator;
101 typedef SeqString::const_reverse_iterator const_reverse_iterator;
102 typedef SeqString::reference reference;
103 typedef SeqString::const_reference const_reference;
104 typedef SeqString::size_type size_type;
105 static const size_type npos = SeqString::npos;
106 enum strand_type { UnknownStrand, PlusStrand, MinusStrand, BothStrand };
107 enum alphabet_ref { reduced_dna_alphabet, reduced_rna_alphabet, reduced_nucleic_alphabet,
108 nucleic_alphabet, protein_alphabet };
110 Sequence(alphabet_ref a = reduced_nucleic_alphabet);
111 Sequence(const char* seq, alphabet_ref a = reduced_nucleic_alphabet);
112 Sequence(const std::string& seq, alphabet_ref a = reduced_nucleic_alphabet);
113 Sequence(const Sequence& seq);
114 Sequence(const Sequence *);
115 Sequence(const SeqSpanRef&, alphabet_ref a = reduced_nucleic_alphabet);
117 //! assignment to constant sequences
118 Sequence &operator=(const Sequence&);
120 friend std::ostream& operator<<(std::ostream&, const Sequence&);
121 friend bool operator<(const Sequence&, const Sequence&);
122 friend bool operator==(const Sequence&, const Sequence&);
123 friend bool operator!=(const Sequence&, const Sequence&);
124 const_reference operator[](size_type) const;
126 //! set sequence to a (sub)string containing nothing but AGCTN
127 void set_filtered_sequence(const std::string& seq,
130 size_type count=npos,
131 strand_type strand=UnknownStrand);
133 //! retrive element at specific position
134 const_reference at(size_type i) const { return seq->at(i); }
135 //! clear the sequence and its annotations
137 //! return a non-null terminated c pointer to the sequence data
138 const char *data() const { return seq->data(); }
140 const_iterator begin() const { return seq->begin(); }
142 const_iterator end() const { return seq->end(); }
143 //! is our sequence empty?
144 bool empty() const { return (seq) ? seq->empty() : true ; }
146 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
147 //! how many base pairs are there in our sequence
148 size_type size() const { return (seq) ? seq->size() : 0; }
149 //! alias for size (used by string)
150 size_type length() const { return size(); }
152 const_reverse_iterator rbegin() const { return seq->rbegin(); }
153 //! reverse end iterator
154 const_reverse_iterator rend() const { return seq->rend(); }
155 //! is our sequence empty?
156 //! start position relative to "base" sequence
157 size_type start() const { return seq->parentStart(); }
158 //! one past the last position relative to "base" sequence
159 size_type stop() const { return seq->parentStop(); }
161 //! return a subsequence, copying over any appropriate annotation
162 Sequence subseq(size_type start=0, size_type count = npos) const;
163 //! reverse a character
164 std::string create_reverse_map() const;
165 //! return a reverse compliment (this needs to be improved?)
166 Sequence rev_comp() const;
168 //! set sequence (filtered)
169 void set_sequence(const std::string &, alphabet_ref);
171 std::string get_sequence() const;
173 void set_species(const std::string &);
175 std::string get_species() const;
176 //! set the fasta header
177 void set_fasta_header(std::string header);
178 //! get the fasta header
179 std::string get_fasta_header() const;
180 //! get name (will return the first non-empty, of fasta_header, species)
181 std::string get_name() const;
182 //! return a reference to whichever alphabet we're currently representing
183 const Alphabet& get_alphabet() const;
184 //! return a reference to whichever alphabet we're currently representing
185 const Alphabet& get_alphabet(alphabet_ref) const;
187 //! load sequence from fasta file using the sequences current alphabet
188 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
189 int start_index=0, int end_index=0);
190 //! load sequence AGCT from fasta file
191 //! \throw mussa_load_error
192 //! \throw sequence_empty_error
193 //! \throw sequence_empty_file_error
194 void load_fasta(const boost::filesystem::path file_path,
197 int start_index=0, int end_index=0);
198 void load_fasta(std::istream& file,
199 int seq_num=1, int start_index=0, int end_index=0);
200 //! load sequence from stream
201 //! \throw mussa_load_error
202 //! \throw sequence_empty_error
203 //! \throw sequence_empty_file_error
204 void load_fasta(std::istream& file,
207 int start_index=0, int end_index=0);
208 //! load sequence annotations
209 //! \throws mussa_load_error
210 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
211 //! load sequence annotations
212 //! \throws mussa_load_error
213 void load_annot(std::fstream& data_stream, int start_index, int end_index);
214 //! parse annotation file
215 /*! \throws annotation_load_error
217 void parse_annot(std::string data, int start_index=0, int end_index=0);
218 //! add an annotation to our list of annotations
219 void add_annotation(const annot& a);
220 const std::list<annot>& annotations() const;
221 const std::list<motif>& 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);
234 void save(boost::filesystem::fstream &save_file);
235 void load_museq(boost::filesystem::path load_file_path, int seq_num);
239 //! which alphabet we're using
240 alphabet_ref alphabet;
241 //! strand orientation
248 //! store our oldstyle annotations
249 std::list<annot> annots;
250 //! a seperate list for motifs since we're currently not saving them
251 std::list<motif> motif_list;
253 //! copy over all our annotation children
254 void copy_children(Sequence &, size_type start, size_type count) const;
256 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
257 std::string rc_motif(std::string a_motif) const;
258 //! look for a string sequence type and and it to an annotation list
259 void add_string_annotation(std::string a_seq, std::string name);
261 // boost::serialization support
262 friend class boost::serialization::access;
263 template<class Archive>
264 void serialize(Archive& ar, const unsigned int /*version*/) {
265 ar & BOOST_SERIALIZATION_NVP(seq);
266 ar & BOOST_SERIALIZATION_NVP(alphabet);
267 ar & BOOST_SERIALIZATION_NVP(strand);
268 ar & BOOST_SERIALIZATION_NVP(header);
269 ar & BOOST_SERIALIZATION_NVP(species);
270 ar & BOOST_SERIALIZATION_NVP(annots);
271 ar & BOOST_SERIALIZATION_NVP(motif_list);
274 BOOST_CLASS_EXPORT(Sequence);