1 #ifndef _MUSSA_CLASS_H_
2 #define _MUSSA_CLASS_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 // ---------- mussa_class.hh -----------
15 // ----------------------------------------
18 #include <boost/filesystem/path.hpp>
19 #include <boost/shared_ptr.hpp>
27 #include "alg/annotation_colors.hpp"
28 #include "alg/mussa_callback.hpp"
29 #include "alg/nway_paths.hpp"
30 #include "alg/sequence.hpp"
32 std::string int_to_str(int an_int);
34 class Mussa : public QObject
39 //! call whatever signaling system we want
40 void progress(const std::string& description, int cur, int max);
43 typedef std::set<Sequence> motif_set;
44 enum analysis_modes { TransitiveNway, RadialNway, EntropyNway,
51 void save(boost::filesystem::path save_path="");
52 //! save the nway comparison
53 void save_muway(boost::filesystem::path save_path);
54 //! load a saved analysis directory
55 void load(boost::filesystem::path ana_path);
56 // ! return path to the where the analysis is stored
57 boost::filesystem::path get_analysis_path() const;
59 void set_analysis_path(boost::filesystem::path);
61 //! clear parameters and initialize data lists
64 //! set parameters from a file - 'mupa' ~ mussa parameters
65 void load_mupa_file(std::string para_file_path) { load_mupa_file(boost::filesystem::path(para_file_path));}
66 void load_mupa_file(boost::filesystem::path para_file_path);
68 // set parameters individually (eg from user input into gui classes)
70 void set_name(std::string a_name);
71 //! return name for this analysis
72 std::string get_name() const;
73 //! return a reasonable window title for this analysis
74 /*! this returns the "variable" portion for a title
76 std::string get_title() const;
78 //! return number of sequences in this analyzis
79 /*! this returns either the_seqs.size() or seq_files.size()
80 * depending on which has data loaded in
81 * (silly delayed loading of sequence data)
85 //! set number of bases for this window size
86 void set_window(int a_window);
87 //! get number of bases for the sliding window
88 int get_window() const;
89 //! set number of bases that must match for a window to be saved
90 //! if threshold > soft_threshold this also sets soft_threshold
91 void set_threshold(int a_threshold);
92 //! get number of bases that must match for a window to be saved
93 int get_threshold() const;
94 //! sets the threshold used for computing the nway paths
95 //! must be in range [threshold..window size]
96 void set_soft_threshold(int sft_thres);
97 int get_soft_threshold() const;
99 void set_analysis_mode(enum analysis_modes new_ana_mode);
100 enum analysis_modes get_analysis_mode() const;
101 //! return a string name for an analysis mode
102 std::string get_analysis_mode_name() const;
103 //! return if we have unsaved changes
104 bool is_dirty() const;
105 //! is there anything loaded into this analysis?
108 //! return the refined paths found by the nway analysis.
109 const NwayPaths& paths() const;
111 //! given selected_paths, and view_paths, compute per base pair matches
112 //template <class IteratorT>
113 void createLocalAlignment(std::list<ConservedPath>::iterator begin,
114 std::list<ConservedPath>::iterator end,
115 std::list<ConservedPath::path_type>& result,
116 std::list<std::vector<bool> >& reversed);
118 //! run seqcomp and the nway filtering algorithm.
119 /*!analyze will run seqcomp and then the nway algorithm
120 * on whatever sequences have been loaded into this mussa instance.
121 * \throws mussa_analysis_error
124 /*! Run the nway filtering algorithm,
125 * this might be used when changing the soft threshhold?
129 //! appends a string sequence to the list of the_seqs
130 // void append_sequence(std::string a_seq);
131 //! appends a sequence to the list of the_seqs (makes copy)
132 void append_sequence(const Sequence& a_seq);
133 //! append a sequence to the list of seqs (shared)
134 void append_sequence(boost::shared_ptr<Sequence> a_seq);
136 //! Load a sequence from a fasta file and any annotations
137 /*! \param[in] seq_file the full path to the fasta file
138 * \param[in] annot_file the full path to an annotation file,
139 * if is an empty string, we won't bother loading anything
140 * \param[in] fasta_index specify which sequence in a multisequence fasta
142 * \param[in] sub_seq_start starting slice index to select a subsequence
143 * use 0 start from the beginning.
144 * \param[in] sub_seq_end ending slice index to select a subsequence
145 * use 0 to go to the end.
146 * \param[in] name sequence name, only used if not null
148 void load_sequence(boost::filesystem::path seq_file,
149 boost::filesystem::path annot_file,
150 int fasta_index, int sub_seq_start=0, int sub_seq_end=0,
151 std::string *name=0);
152 //! allow examining the sequences we have loaded
153 typedef std::vector<boost::shared_ptr<Sequence> > vector_sequence_type;
154 const vector_sequence_type& sequences() const;
156 // deprecated - support bridge for python version of mussa
157 // these save & load from the old file format
159 void load_old(char * load_file_path, int s_num);
161 // manage motif lists
162 //! add a motif it wont be applied until update_sequences_motif is called
163 void add_motif(const Sequence& motifs, const Color& colors);
164 //! add vector of motifs and colors to our motif collection
165 /*! this will automatically call update_sequences_motif
166 * this depends on sets and color maps being unique
167 * (aka if you add the same item more than once it doesn't
168 * increase the size of the data structure)
170 void set_motifs(const std::vector<Sequence>& motifs,
171 const std::vector<Color>& colors);
172 /*! The file should look something like
173 * <sequence> <red> <green> <blue>
174 * where sequence is a string of IUPAC symbols
175 * and red,green,blue are a white space separated list of floats
176 * in the range [0.0, 1.0]
178 //! load a list of motifs from a file named filename
179 void load_motifs(boost::filesystem::path filename);
180 //! load motifs from an ifstream
181 /*! \sa Mussa::load_motifs(boost::filesystem::path)
183 void load_motifs(std::istream &);
184 //! return our motifs;
185 const motif_set& motifs() const;
187 //! return color mapper
188 boost::shared_ptr<AnnotationColors> colorMapper();
191 //! push motifs to our attached sequences
192 void update_sequences_motifs();
195 // parameters needed for a mussa analysis
196 //! name of this analysis. (will also be used when saving an analysis)
197 std::string analysis_name;
198 //! how many base pairs to include in a sliding window
200 //! how many base pairs need to match order to record a window as conserved
202 //! stores current filter used by GUI to change the connections shown
204 //! which nway comparison algorithm to use.
205 enum analysis_modes ana_mode;
207 //! should we append _w<window_size> to the saved analysis
209 //! should we append _t<threshold> to the saved analysis
213 vector_sequence_type the_seqs;
215 std::vector<std::vector<FLPs> > all_comps;
216 //! N-way data, ie the mussa results
220 motif_set motif_sequences;
222 boost::shared_ptr<AnnotationColors> color_mapper;
223 //! path to our analysis
224 boost::filesystem::path analysis_path;
225 //! flag indicating if we have unsaved changes
229 //! runs all the seqcomps needed to support the nway comparison
233 //! provide a simple name to point to our Mussa shared_ptr
234 typedef boost::shared_ptr<Mussa> MussaRef;