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>
34 // Sequence data class
36 //! Attach annotation information to a sequence track
40 annot(int begin, int end, std::string type, std::string name);
48 friend bool operator==(const annot& left, const annot& right);
50 // boost::serialization support
51 friend class boost::serialization::access;
52 template<class Archive>
53 void serialize(Archive& ar, const unsigned int /*version*/) {
54 ar & BOOST_SERIALIZATION_NVP(begin);
55 ar & BOOST_SERIALIZATION_NVP(end);
56 ar & BOOST_SERIALIZATION_NVP(type);
57 ar & BOOST_SERIALIZATION_NVP(name);
60 BOOST_CLASS_EXPORT(annot);
63 /* The way that motifs are found currently doesn't really
64 * indicate that the match was a reverse compliment
66 struct motif : public annot
70 motif() : annot(), sequence("") {};
71 //! this constructor is for when we're adding motifs to our annotations
72 motif(int begin, std::string motif);
75 // boost::serialization support
77 friend class boost::serialization::access;
78 template<class Archive>
79 void serialize(Archive& ar, const unsigned int /*version*/) {
80 ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(annot);
81 ar & BOOST_SERIALIZATION_NVP(sequence);
84 BOOST_CLASS_EXPORT(motif);
86 //! the only purpose of this class is that the shared_ptr template
87 //! functions need the serialization support to be in-class.
88 class seq_string : public std::string
91 friend class boost::serialization::access;
92 template<class Archive>
93 void serialize(Archive& ar, const unsigned int /*version*/) {
94 //ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(std::string);
95 ar & boost::serialization::make_nvp("bases",
96 boost::serialization::base_object<std::string>(*this)
101 //! sequence track for mussa.
105 typedef std::string::difference_type difference_type;
106 typedef std::string::iterator iterator;
107 typedef std::string::const_iterator const_iterator;
108 typedef std::string::reference reference;
109 typedef std::string::const_reference const_reference;
110 typedef std::string::size_type size_type;
111 static const size_type npos = std::string::npos;
112 enum strand_type { UnknownStrand, PlusStrand, MinusStrand, BothStrand };
114 // some standard dna alphabets
115 // Include nl (\012), and cr (\015) to make sequence parsing eol
116 // convention independent.
118 static const std::string dna_alphabet;
119 static const std::string rna_alphabet;
120 //! this is the general iupac alphabet for nucleotides
121 static const std::string nucleic_iupac_alphabet;
122 //! the protein alphabet
123 static const std::string protein_alphabet;
127 Sequence(const char* seq);
128 Sequence(const std::string& seq);
129 Sequence(const Sequence& seq);
130 //! assignment to constant sequences
131 Sequence &operator=(const Sequence&);
133 friend std::ostream& operator<<(std::ostream&, const Sequence&);
134 friend bool operator<(const Sequence&, const Sequence&);
135 friend bool operator==(const Sequence&, const Sequence&);
136 const_reference operator[](size_type) const;
138 //! set sequence to a (sub)string containing nothing but AGCTN
139 void set_filtered_sequence(const std::string& seq,
141 size_type count=npos,
142 strand_type strand=UnknownStrand);
144 //! retrive element at specific position
145 const_reference at(size_type n) const;
146 //! clear the sequence and its annotations
148 //! return c pointer to the sequence data
149 const char *c_str() const;
151 const_iterator begin() const;
153 const_iterator end() const;
154 //! is our sequence empty?
156 //! how many base pairs are there in our sequence
157 size_type size() const;
158 //! alias for size (used by string)
159 size_type length() const;
160 //! start position relative to "base" sequence
161 size_type start() const;
162 //! one past the last position relative to "base" sequence
163 size_type stop() const;
165 //! return a subsequence, copying over any appropriate annotation
166 Sequence subseq(int start=0, int count = std::string::npos);
167 //! return a reverse compliment (this needs to be improved?)
168 std::string rev_comp() const;
170 //! set sequence (filtered)
171 void set_sequence(const std::string &);
173 std::string get_sequence() const;
175 void set_species(const std::string &);
177 std::string get_species() const;
178 //! set the fasta header
179 void set_fasta_header(std::string header);
180 //! get the fasta header
181 std::string get_fasta_header() const;
182 //! get name (will return the first non-empty, of fasta_header, species)
183 std::string get_name() const;
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, int seq_num=1,
190 int start_index=0, int end_index=0);
191 //! load sequence from stream
192 //! \throw mussa_load_error
193 //! \throw sequence_empty_error
194 //! \throw sequence_empty_file_error
195 void load_fasta(std::iostream& file, int seq_num=1,
196 int start_index=0, int end_index=0);
197 //! load sequence annotations
198 //! \throws mussa_load_error
199 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
200 //! load sequence annotations
201 //! \throws mussa_load_error
202 void load_annot(std::fstream& data_stream, int start_index, int end_index);
203 bool parse_annot(std::string data, int start_index=0, int end_index=0);
204 //! add an annotation to our list of annotations
205 void add_annotation(const annot& a);
206 const std::list<annot>& annotations() const;
207 const std::list<motif>& motifs() const;
209 //! add a motif to our list of motifs
210 //! \throws motif_normalize_error if there's something wrong with a_motif
211 void add_motif(const Sequence& a_motif);
212 //! clear our list of found motifs
214 //! search a sequence for a_motif
215 //! \throws motif_normalize_error if there's something wrong with a_motif
216 std::vector<int> find_motif(const std::string& a_motif) const;
217 //! search a sequence for a_motif
218 //! \throws motif_normalize_error if there's something wrong with a_motif
219 std::vector<int> find_motif(const Sequence& a_motif) const;
220 //! convert IUPAC symbols to upperase
221 //! \throws motif_normalize_error if there is an invalid symbol
222 static std::string motif_normalize(const std::string& a_motif);
224 //! annotate the current sequence with other sequences
225 void find_sequences(std::list<Sequence>::iterator start,
226 std::list<Sequence>::iterator end);
228 void save(boost::filesystem::fstream &save_file);
229 void load_museq(boost::filesystem::path load_file_path, int seq_num);
234 //! hold a shared pointer to our sequence string
235 boost::shared_ptr<seq_string> seq;
236 //! start offset into the sequence
238 //! number of basepairs of the shared sequence we represent
240 //! strand orientation
247 //! store our oldstyle annotations
248 std::list<annot> annots;
249 //! a seperate list for motifs since we're currently not saving them
250 std::list<motif> motif_list;
252 void motif_scan(std::string a_motif, std::vector<int> * motif_match_starts) const;
253 std::string rc_motif(std::string a_motif) const;
254 //! look for a string sequence type and and it to an annotation list
255 void add_string_annotation(std::string a_seq, std::string name);
257 // boost::serialization support
258 friend class boost::serialization::access;
259 template<class Archive>
260 void serialize(Archive& ar, const unsigned int /*version*/) {
261 ar & BOOST_SERIALIZATION_NVP(parent);
262 ar & BOOST_SERIALIZATION_NVP(seq);
263 ar & BOOST_SERIALIZATION_NVP(seq_start);
264 ar & BOOST_SERIALIZATION_NVP(seq_count);
265 ar & BOOST_SERIALIZATION_NVP(strand);
266 ar & BOOST_SERIALIZATION_NVP(header);
267 ar & BOOST_SERIALIZATION_NVP(species);
268 ar & BOOST_SERIALIZATION_NVP(annots);
269 ar & BOOST_SERIALIZATION_NVP(motif_list);
272 BOOST_CLASS_EXPORT(Sequence);