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/list.hpp>
22 #include <boost/serialization/nvp.hpp>
23 #include <boost/serialization/string.hpp>
24 #include <boost/serialization/utility.hpp>
25 #include <boost/serialization/export.hpp>
32 // Sequence data class
34 //! Attach annotation information to a sequence track
38 annot(int begin, int end, std::string type, std::string name);
46 friend bool operator==(const annot& left, const annot& right);
48 // boost::serialization support
49 friend class boost::serialization::access;
50 template<class Archive>
51 void serialize(Archive& ar, const unsigned int /*version*/) {
52 ar & BOOST_SERIALIZATION_NVP(begin);
53 ar & BOOST_SERIALIZATION_NVP(end);
54 ar & BOOST_SERIALIZATION_NVP(type);
55 ar & BOOST_SERIALIZATION_NVP(name);
58 BOOST_CLASS_EXPORT(annot);
61 /* The way that motifs are found currently doesn't really
62 * indicate that the match was a reverse compliment
64 struct motif : public annot
68 motif() : annot(), sequence("") {};
69 //! this constructor is for when we're adding motifs to our annotations
70 motif(int begin, std::string motif);
73 // boost::serialization support
75 friend class boost::serialization::access;
76 template<class Archive>
77 void serialize(Archive& ar, const unsigned int /*version*/) {
78 ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(annot);
79 ar & BOOST_SERIALIZATION_NVP(sequence);
82 BOOST_CLASS_EXPORT(motif);
84 //! sequence track for mussa.
88 typedef std::string::difference_type difference_type;
89 typedef std::string::iterator iterator;
90 typedef std::string::const_iterator const_iterator;
91 typedef std::string::reference reference;
92 typedef std::string::const_reference const_reference;
93 typedef std::string::size_type size_type;
94 static const size_type npos = std::string::npos;
98 Sequence(const char* seq);
99 Sequence(const std::string& seq);
100 Sequence(const Sequence& seq);
101 //! assignment to constant sequences
102 Sequence &operator=(const Sequence&);
104 // dangerous since they create large copies
105 //operator std::string();
106 //operator std::string() const;
108 friend std::ostream& operator<<(std::ostream&, const Sequence&);
109 friend bool operator<(const Sequence&, const Sequence&);
110 friend bool operator==(const Sequence&, const Sequence&);
111 const_reference operator[](size_type) const;
113 //! set sequence to a (sub)string containing nothing but AGCTN
114 void set_filtered_sequence(const std::string& seq,
115 std::string::size_type start=0,
116 std::string::size_type count=0);
118 //! retrive element at specific position
119 const_reference at(size_type n) const;
120 //! clear the sequence and its annotations
124 const_iterator begin() const;
126 const_iterator end() const;
127 //! is our sequence empty?
129 //! how many base pairs are there in our sequence
130 size_type size() const;
131 //! alias for size (used by string)
132 size_type length() const;
134 //! return a subsequence, copying over any appropriate annotation
135 Sequence subseq(int start=0, int count = std::string::npos) const;
136 //! return a reverse compliment
137 std::string rev_comp() const;
139 //! set sequence (filtered)
140 void set_sequence(const std::string &);
142 std::string get_sequence() const;
144 void set_species(const std::string &);
146 std::string get_species() const;
147 //! set the fasta header
148 void set_fasta_header(std::string header);
149 //! get the fasta header
150 std::string get_fasta_header() const;
151 //! get name (will return the first non-empty, of fasta_header, species)
152 std::string get_name() const;
154 //! load sequence AGCT from fasta file
155 //! \throw mussa_load_error
156 //! \throw sequence_empty_error
157 //! \throw sequence_empty_file_error
158 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
159 int start_index=0, int end_index=0);
160 //! load sequence from stream
161 //! \throw mussa_load_error
162 //! \throw sequence_empty_error
163 //! \throw sequence_empty_file_error
164 void load_fasta(std::iostream& file, int seq_num=1,
165 int start_index=0, int end_index=0);
166 //! load sequence annotations
167 //! \throws mussa_load_error
168 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
169 //! load sequence annotations
170 //! \throws mussa_load_error
171 void load_annot(std::fstream& data_stream, int start_index, int end_index);
172 bool parse_annot(std::string data, int start_index=0, int end_index=0);
173 //! add an annotation to our list of annotations
174 void add_annotation(const annot& a);
175 const std::list<annot>& annotations() const;
176 const std::list<motif>& motifs() const;
178 //! add a motif to our list of motifs
179 //! \throws motif_normalize_error if there's something wrong with a_motif
180 void add_motif(const Sequence& a_motif);
181 //! clear our list of found motifs
183 //! search a sequence for a_motif
184 //! \throws motif_normalize_error if there's something wrong with a_motif
185 std::vector<int> find_motif(const std::string& a_motif) const;
186 //! search a sequence for a_motif
187 //! \throws motif_normalize_error if there's something wrong with a_motif
188 std::vector<int> find_motif(const Sequence& a_motif) const;
189 //! convert IUPAC symbols to upperase
190 //! \throws motif_normalize_error if there is an invalid symbol
191 static std::string motif_normalize(const std::string& a_motif);
193 //! annotate the current sequence with other sequences
194 void find_sequences(std::list<Sequence>::iterator start,
195 std::list<Sequence>::iterator end);
197 void save(boost::filesystem::fstream &save_file);
198 void load_museq(boost::filesystem::path load_file_path, int seq_num);
205 std::list<annot> annots;
206 //! a seperate list for motifs since we're currently not saving them
207 std::list<motif> motif_list;
209 void motif_scan(std::string a_motif, std::vector<int> * motif_match_starts) const;
210 std::string rc_motif(std::string a_motif) const;
211 //! look for a string sequence type and and it to an annotation list
212 void add_string_annotation(std::string a_seq, std::string name);
214 // boost::serialization support
215 friend class boost::serialization::access;
216 template<class Archive>
217 void serialize(Archive& ar, const unsigned int /*version*/) {
218 ar & BOOST_SERIALIZATION_NVP(seq);
219 ar & BOOST_SERIALIZATION_NVP(header);
220 ar & BOOST_SERIALIZATION_NVP(species);
221 ar & BOOST_SERIALIZATION_NVP(annots);
222 ar & BOOST_SERIALIZATION_NVP(motif_list);
225 BOOST_CLASS_EXPORT(Sequence);