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 /* The way that motifs are found currently doesn't really
42 * indicate that the match was a reverse compliment
47 //motif(int begin, int end, std::string type, std::string name);
48 //! this constructor is for when we're adding motifs to our annotations
49 motif(int begin, std::string motif);
58 friend bool operator==(const motif& left, const motif& right);
60 // boost::serialization support
62 friend class boost::serialization::access;
63 template<class Archive>
64 void serialize(Archive& ar, const unsigned int /*version*/) {
65 ar & BOOST_SERIALIZATION_NVP(begin);
66 ar & BOOST_SERIALIZATION_NVP(end);
67 ar & BOOST_SERIALIZATION_NVP(type);
68 ar & BOOST_SERIALIZATION_NVP(name);
69 ar & BOOST_SERIALIZATION_NVP(sequence);
72 BOOST_CLASS_EXPORT(motif);
75 typedef boost::shared_ptr<Sequence> SequenceRef;
77 //! sequence track for mussa.
81 typedef SeqString::value_type value_type;
82 typedef SeqString::difference_type difference_type;
83 typedef SeqString::iterator iterator;
84 typedef SeqString::reverse_iterator reverse_iterator;
85 typedef SeqString::const_iterator const_iterator;
86 typedef SeqString::const_reverse_iterator const_reverse_iterator;
87 typedef SeqString::reference reference;
88 typedef SeqString::const_reference const_reference;
89 typedef SeqString::size_type size_type;
90 static const size_type npos = SeqString::npos;
92 typedef std::list<motif> MotifList;
93 typedef boost::shared_ptr<MotifList> MotifListRef;
95 Sequence(AlphabetRef a = reduced_dna_alphabet);
96 Sequence(const char* seq,
97 AlphabetRef a = reduced_dna_alphabet,
98 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
99 Sequence(const std::string& seq,
100 AlphabetRef a = reduced_dna_alphabet,
101 SeqSpan::strand_type strand = SeqSpan::PlusStrand);
102 //! make a new sequence, with the same SeqSpan
103 Sequence(const Sequence& seq);
104 //! make a new sequence, with the same SeqSpan
105 Sequence(const Sequence *);
106 //! Make a new sequence using a copy of SeqSpan
107 Sequence(const SequenceRef);
108 Sequence(const SeqSpanRef&);
110 //! assignment to constant sequences
111 Sequence &operator=(const Sequence&);
113 friend std::ostream& operator<<(std::ostream&, const Sequence&);
114 friend bool operator<(const Sequence&, const Sequence&);
115 friend bool operator==(const Sequence&, const Sequence&);
116 friend bool operator!=(const Sequence&, const Sequence&);
117 const_reference operator[](size_type) const;
119 //! set sequence to a (sub)string containing nothing but AGCTN
120 void set_filtered_sequence(const std::string& seq,
121 AlphabetRef a=reduced_dna_alphabet,
123 size_type count=npos,
124 SeqSpan::strand_type strand=SeqSpan::PlusStrand);
126 //! retrive element at specific position
127 const_reference at(size_type i) const { return seq->at(i); }
128 //! clear the sequence and its annotations
130 //! return a non-null terminated c pointer to the sequence data
131 const char *data() const { return seq->data(); }
133 const_iterator begin() const { return seq->begin(); }
135 const_iterator end() const { return seq->end(); }
136 //! is our sequence empty?
137 bool empty() const { return (seq) ? seq->empty() : true ; }
139 size_type find_first_not_of(const std::string& q, size_type index=0) { return seq->find_first_not_of(q, index); }
140 //! how many base pairs are there in our sequence
141 size_type size() const { return (seq) ? seq->size() : 0; }
142 //! alias for size (used by string)
143 size_type length() const { return size(); }
145 const_reverse_iterator rbegin() const { return seq->rbegin(); }
146 //! reverse end iterator
147 const_reverse_iterator rend() const { return seq->rend(); }
148 //! is our sequence empty?
149 //! start position relative to "base" sequence
150 size_type start() const { return seq->parentStart(); }
151 //! one past the last position relative to "base" sequence
152 size_type stop() const { return seq->parentStop(); }
154 //! return a subsequence, copying over any appropriate annotation
155 Sequence subseq(size_type start=0,
156 size_type count = npos,
157 SeqSpan::strand_type strand = SeqSpan::SameStrand) const;
158 //! reverse a character
159 std::string create_reverse_map() const;
160 //! return a reverse compliment (this needs to be improved?)
161 Sequence rev_comp() const;
163 //! set sequence (filtered)
164 void set_sequence(const std::string &, AlphabetRef);
166 std::string get_sequence() const;
168 void set_species(const std::string &);
170 std::string get_species() const;
171 //! set the fasta header
172 void set_fasta_header(std::string header);
173 //! get the fasta header
174 std::string get_fasta_header() const;
175 //! get name (will return the first non-empty, of fasta_header, species)
176 std::string get_name() const;
177 //! return a reference to whichever alphabet we're currently representing
178 const Alphabet& get_alphabet() const;
180 //! load sequence from fasta file using the sequences current alphabet
181 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
182 int start_index=0, int end_index=0);
183 //! load sequence AGCT from fasta file
184 //! \throw mussa_load_error
185 //! \throw sequence_empty_error
186 //! \throw sequence_empty_file_error
187 void load_fasta(const boost::filesystem::path file_path,
190 int start_index=0, int end_index=0);
191 void load_fasta(std::istream& file,
192 int seq_num=1, int start_index=0, int end_index=0);
193 //! load sequence from stream
194 //! \throw mussa_load_error
195 //! \throw sequence_empty_error
196 //! \throw sequence_empty_file_error
197 void load_fasta(std::istream& file,
200 int start_index=0, int end_index=0);
201 //! load sequence annotations
202 //! \throws mussa_load_error
203 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
204 //! load sequence annotations
205 //! \throws mussa_load_error
206 void load_annot(std::fstream& data_stream, int start_index, int end_index);
207 //! parse annotation file
208 /*! \throws annotation_load_error
210 void parse_annot(std::string data, int start_index=0, int end_index=0);
211 //! add an annotation to our list of annotations
212 void add_annotation(const SeqSpanRef a);
213 //! add an annotation using tristan's mussa file paramenters
214 void add_annotation(std::string name, std::string type, size_type start, size_type stop);
215 //! create an initialized annotation with the "standard" types.
216 SeqSpanRef make_annotation(std::string name, std::string type, size_type start, size_type stop) const;
217 const SeqSpanRefList& annotations() const;
219 const MotifList& motifs() const;
221 //! add a motif to our list of motifs
222 void add_motif(const Sequence& a_motif);
223 //! clear our list of found motifs
225 //! search a sequence for a_motif
226 //! \throws motif_normalize_error if there's something wrong with a_motif
227 std::vector<int> find_motif(const Sequence& a_motif) const;
228 //! annotate the current sequence with other sequences
229 void find_sequences(std::list<Sequence>::iterator start,
230 std::list<Sequence>::iterator end);
232 void save(boost::filesystem::fstream &save_file);
233 void load_museq(boost::filesystem::path load_file_path, int seq_num);
242 //! store annotation regions
243 SeqSpanRefListRef annotation_list;
244 //! a seperate list for motifs since we're currently not saving them
245 MotifListRef motif_list;
247 //! copy over all our annotation children
248 void copy_children(Sequence &, size_type start, size_type count) const;
250 void motif_scan(const Sequence& a_motif, std::vector<int> * motif_match_starts) const;
251 std::string rc_motif(std::string a_motif) const;
252 //! look for a string sequence type and and it to an annotation list
253 void add_string_annotation(std::string a_seq, std::string name);
255 // boost::serialization support
256 friend class boost::serialization::access;
257 template<class Archive>
258 void serialize(Archive& ar, const unsigned int /*version*/) {
259 ar & BOOST_SERIALIZATION_NVP(seq);
260 ar & BOOST_SERIALIZATION_NVP(header);
261 ar & BOOST_SERIALIZATION_NVP(species);
262 ar & BOOST_SERIALIZATION_NVP(annotation_list);
263 ar & BOOST_SERIALIZATION_NVP(motif_list);
266 BOOST_CLASS_EXPORT(Sequence);