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;
107 Sequence(AlphabetRef a = reduced_dna_alphabet);
108 Sequence(const char* seq,
109 AlphabetRef a = reduced_dna_alphabet,
110 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
111 Sequence(const std::string& seq,
112 AlphabetRef a = reduced_dna_alphabet,
113 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
114 Sequence(const Sequence& seq);
115 Sequence(const Sequence *);
116 Sequence(const SeqSpanRef&);
118 //! assignment to constant sequences
119 Sequence &operator=(const Sequence&);
121 friend std::ostream& operator<<(std::ostream&, const Sequence&);
122 friend bool operator<(const Sequence&, const Sequence&);
123 friend bool operator==(const Sequence&, const Sequence&);
124 friend bool operator!=(const Sequence&, const Sequence&);
125 const_reference operator[](size_type) const;
127 //! set sequence to a (sub)string containing nothing but AGCTN
128 void set_filtered_sequence(const std::string& seq,
129 AlphabetRef a=reduced_dna_alphabet,
131 size_type count=npos,
132 SeqSpan::strand_type strand=SeqSpan::PlusStrand);
134 //! retrive element at specific position
135 const_reference at(size_type i) const { return seq->at(i); }
136 //! clear the sequence and its annotations
138 //! return a non-null terminated c pointer to the sequence data
139 const char *data() const { return seq->data(); }
141 const_iterator begin() const { return seq->begin(); }
143 const_iterator end() const { return seq->end(); }
144 //! is our sequence empty?
145 bool empty() const { return (seq) ? seq->empty() : true ; }
147 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
148 //! how many base pairs are there in our sequence
149 size_type size() const { return (seq) ? seq->size() : 0; }
150 //! alias for size (used by string)
151 size_type length() const { return size(); }
153 const_reverse_iterator rbegin() const { return seq->rbegin(); }
154 //! reverse end iterator
155 const_reverse_iterator rend() const { return seq->rend(); }
156 //! is our sequence empty?
157 //! start position relative to "base" sequence
158 size_type start() const { return seq->parentStart(); }
159 //! one past the last position relative to "base" sequence
160 size_type stop() const { return seq->parentStop(); }
162 //! return a subsequence, copying over any appropriate annotation
163 Sequence subseq(size_type start=0,
164 size_type count = npos,
165 SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
166 //! reverse a character
167 std::string create_reverse_map() const;
168 //! return a reverse compliment (this needs to be improved?)
169 Sequence rev_comp() const;
171 //! set sequence (filtered)
172 void set_sequence(const std::string &, AlphabetRef);
174 std::string get_sequence() const;
176 void set_species(const std::string &);
178 std::string get_species() const;
179 //! set the fasta header
180 void set_fasta_header(std::string header);
181 //! get the fasta header
182 std::string get_fasta_header() const;
183 //! get name (will return the first non-empty, of fasta_header, species)
184 std::string get_name() const;
185 //! return a reference to whichever alphabet we're currently representing
186 const Alphabet& get_alphabet() const;
188 //! load sequence from fasta file using the sequences current alphabet
189 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
190 int start_index=0, int end_index=0);
191 //! load sequence AGCT from fasta file
192 //! \throw mussa_load_error
193 //! \throw sequence_empty_error
194 //! \throw sequence_empty_file_error
195 void load_fasta(const boost::filesystem::path file_path,
198 int start_index=0, int end_index=0);
199 void load_fasta(std::istream& file,
200 int seq_num=1, int start_index=0, int end_index=0);
201 //! load sequence from stream
202 //! \throw mussa_load_error
203 //! \throw sequence_empty_error
204 //! \throw sequence_empty_file_error
205 void load_fasta(std::istream& file,
208 int start_index=0, int end_index=0);
209 //! load sequence annotations
210 //! \throws mussa_load_error
211 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
212 //! load sequence annotations
213 //! \throws mussa_load_error
214 void load_annot(std::fstream& data_stream, int start_index, int end_index);
215 //! parse annotation file
216 /*! \throws annotation_load_error
218 void parse_annot(std::string data, int start_index=0, int end_index=0);
219 //! add an annotation to our list of annotations
220 void add_annotation(const annot& a);
221 const std::list<annot>& annotations() const;
222 const std::list<motif>& motifs() const;
224 //! add a motif to our list of motifs
225 void add_motif(const Sequence& a_motif);
226 //! clear our list of found motifs
228 //! search a sequence for a_motif
229 //! \throws motif_normalize_error if there's something wrong with a_motif
230 std::vector<int> find_motif(const Sequence& a_motif) const;
231 //! annotate the current sequence with other sequences
232 void find_sequences(std::list<Sequence>::iterator start,
233 std::list<Sequence>::iterator end);
235 void save(boost::filesystem::fstream &save_file);
236 void load_museq(boost::filesystem::path load_file_path, int seq_num);
245 //! store our oldstyle annotations
246 std::list<annot> annots;
247 //! a seperate list for motifs since we're currently not saving them
248 std::list<motif> motif_list;
250 //! copy over all our annotation children
251 void copy_children(Sequence &, size_type start, size_type count) const;
253 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
254 std::string rc_motif(std::string a_motif) const;
255 //! look for a string sequence type and and it to an annotation list
256 void add_string_annotation(std::string a_seq, std::string name);
258 // boost::serialization support
259 friend class boost::serialization::access;
260 template<class Archive>
261 void serialize(Archive& ar, const unsigned int /*version*/) {
262 ar & BOOST_SERIALIZATION_NVP(seq);
263 ar & BOOST_SERIALIZATION_NVP(header);
264 ar & BOOST_SERIALIZATION_NVP(species);
265 ar & BOOST_SERIALIZATION_NVP(annots);
266 ar & BOOST_SERIALIZATION_NVP(motif_list);
269 BOOST_CLASS_EXPORT(Sequence);