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.
85 class Sequence : public std::string
91 std::list<annot> annots;
92 //! a seperate list for motifs since we're currently not saving them
93 std::list<motif> motif_list;
95 void motif_scan(std::string a_motif, std::vector<int> * motif_match_starts);
96 std::string rc_motif(std::string a_motif);
97 //! look for a string sequence type and and it to an annotation list
98 void add_string_annotation(std::string a_seq, std::string name);
100 // boost::serialization support
101 friend class boost::serialization::access;
102 template<class Archive>
103 void serialize(Archive& ar, const unsigned int /*version*/) {
104 ar & boost::serialization::make_nvp(
106 boost::serialization::base_object<std::string >(*this)
108 ar & BOOST_SERIALIZATION_NVP(header);
109 ar & BOOST_SERIALIZATION_NVP(species);
110 ar & BOOST_SERIALIZATION_NVP(annots);
111 ar & BOOST_SERIALIZATION_NVP(motif_list);
117 Sequence(const char* seq);
118 Sequence(const std::string& seq);
119 Sequence(const Sequence& seq);
120 //! assignment to constant sequences
121 Sequence &operator=(const Sequence&);
123 //! set sequence to a (sub)string containing nothing but AGCTN
124 void set_filtered_sequence(const std::string& seq,
125 std::string::size_type start=0,
126 std::string::size_type count=0);
128 //! load sequence AGCT from fasta file
129 //! \throw mussa_load_error
130 //! \throw sequence_empty_error
131 //! \throw sequence_empty_file_error
132 void load_fasta(const boost::filesystem::path file_path, int seq_num=1,
133 int start_index=0, int end_index=0);
134 //! load sequence from stream
135 //! \throw mussa_load_error
136 //! \throw sequence_empty_error
137 //! \throw sequence_empty_file_error
138 void load_fasta(std::iostream& file, int seq_num=1,
139 int start_index=0, int end_index=0);
140 //! load sequence annotations
141 //! \throws mussa_load_error
142 void load_annot(const boost::filesystem::path file_path, int start_index, int end_index);
143 //! load sequence annotations
144 //! \throws mussa_load_error
145 void load_annot(std::fstream& data_stream, int start_index, int end_index);
146 bool parse_annot(std::string data, int start_index=0, int end_index=0);
147 //! add an annotation to our list of annotations
148 void add_annotation(const annot& a);
149 const std::list<annot>& annotations() const;
150 const std::list<motif>& motifs() const;
152 //! return a subsequence, copying over any appropriate annotation
153 Sequence subseq(int start=0, int count = std::string::npos) const;
154 //! return a reverse compliment
155 std::string rev_comp() const;
157 //! clear the sequence and its annotations
161 void set_species(const std::string &);
163 std::string get_species() const;
164 //! set the fasta header
165 void set_fasta_header(std::string header);
166 //! get the fasta header
167 std::string get_fasta_header() const;
168 //! get name (will return the first non-empty, of fasta_header, species)
169 std::string get_name() const;
171 //! add a motif to our list of motifs
172 //! \throws motif_normalize_error if there's something wrong with a_motif
173 void add_motif(const Sequence& a_motif);
174 //! clear our list of found motifs
176 //! search a sequence for a_motif
177 //! \throws motif_normalize_error if there's something wrong with a_motif
178 std::vector<int> find_motif(std::string a_motif);
179 //! convert IUPAC symbols to upperase
180 //! \throws motif_normalize_error if there is an invalid symbol
181 static std::string motif_normalize(std::string a_motif);
183 //! annotate the current sequence with other sequences
184 void find_sequences(std::list<Sequence>::iterator start,
185 std::list<Sequence>::iterator end);
187 void save(boost::filesystem::fstream &save_file);
188 void load_museq(boost::filesystem::path load_file_path, int seq_num);
191 BOOST_CLASS_EXPORT(Sequence);