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 enum analysis_modes { TransitiveNway, RadialNway, EntropyNway,
50 void save(boost::filesystem::path save_path="");
51 //! save the nway comparison
52 void save_muway(boost::filesystem::path save_path);
53 //! load a saved analysis directory
54 void load(boost::filesystem::path ana_path);
55 // ! return path to the where the analysis is stored
56 boost::filesystem::path get_analysis_path() const;
58 //! clear parameters and initialize data lists
61 //! set parameters from a file - 'mupa' ~ mussa parameters
62 void load_mupa_file(std::string para_file_path) { load_mupa_file(boost::filesystem::path(para_file_path));}
63 void load_mupa_file(boost::filesystem::path para_file_path);
65 // set parameters individually (eg from user input into gui classes)
67 void set_name(std::string a_name);
68 //! return name for this analysis
69 std::string get_name();
71 //! return number of sequences in this analyzis
72 /*! this returns either the_seqs.size() or seq_files.size()
73 * depending on which has data loaded in
74 * (silly delayed loading of sequence data)
78 //! set number of bases for this window size
79 void set_window(int a_window);
80 //! get number of bases for the sliding window
81 int get_window() const;
82 //! set number of bases that must match for a window to be saved
83 //! if threshold > soft_threshold this also sets soft_threshold
84 void set_threshold(int a_threshold);
85 //! get number of bases that must match for a window to be saved
86 int get_threshold() const;
87 //! sets the threshold used for computing the nway paths
88 //! must be in range [threshold..window size]
89 void set_soft_threshold(int sft_thres);
90 int get_soft_threshold() const;
92 void set_analysis_mode(enum analysis_modes new_ana_mode);
93 enum analysis_modes get_analysis_mode() const;
94 //! return a string name for an analysis mode
95 std::string get_analysis_mode_name() const;
96 //! return if we have unsaved changes
97 bool is_dirty() const;
98 //! is there anything loaded into this analysis?
101 //! return the refined paths found by the nway analysis.
102 const NwayPaths& paths() const;
104 //! given selected_paths, and view_paths, compute per base pair matches
105 //template <class IteratorT>
106 void createLocalAlignment(std::list<ConservedPath>::iterator begin,
107 std::list<ConservedPath>::iterator end,
108 std::list<ConservedPath::path_type>& result,
109 std::list<std::vector<bool> >& reversed);
111 //! run seqcomp and the nway filtering algorithm.
112 /*!analyze will run seqcomp and then the nway algorithm
113 * on whatever sequences have been loaded into this mussa instance.
114 * \throws mussa_analysis_error
117 /*! Run the nway filtering algorithm,
118 * this might be used when changing the soft threshhold?
122 //! appends a string sequence to the list of the_seqs
123 // void append_sequence(std::string a_seq);
124 //! appends a sequence to the list of the_seqs (makes copy)
125 void append_sequence(const Sequence& a_seq);
126 //! append a sequence to the list of seqs (shared)
127 void append_sequence(boost::shared_ptr<Sequence> a_seq);
129 //! Load a sequence from a fasta file and any annotations
130 /*! \param[in] seq_file the full path to the fasta file
131 * \param[in] annot_file the full path to an annotation file,
132 * if is an empty string, we won't bother loading anything
133 * \param[in] fasta_index specify which sequence in a multisequence fasta
135 * \param[in] sub_seq_start starting slice index to select a subsequence
136 * use 0 start from the beginning.
137 * \param[in] sub_seq_end ending slice index to select a subsequence
138 * use 0 to go to the end.
139 * \param[in] name sequence name, only used if not null
141 void load_sequence(boost::filesystem::path seq_file,
142 boost::filesystem::path annot_file,
143 int fasta_index, int sub_seq_start=0, int sub_seq_end=0,
144 std::string *name=0);
145 //! allow examining the sequences we have loaded
146 typedef std::vector<boost::shared_ptr<Sequence> > vector_sequence_type;
147 const vector_sequence_type& sequences() const;
149 // deprecated - support bridge for python version of mussa
150 // these save & load from the old file format
152 void load_old(char * load_file_path, int s_num);
154 // manage motif lists
155 //! add a motif it wont be applied until update_sequences_motif is called
156 void add_motif(const Sequence& motifs, const Color& colors);
157 //! add vector of motifs and colors to our motif collection
158 /*! this will automatically call update_sequences_motif
159 * this depends on sets and color maps being unique
160 * (aka if you add the same item more than once it doesn't
161 * increase the size of the data structure)
163 void set_motifs(const std::vector<Sequence>& motifs,
164 const std::vector<Color>& colors);
165 //! load motifs from an ifstream
166 /*! The file should look something like
167 * <sequence> <red> <green> <blue>
168 * where sequence is a string of IUPAC symbols
169 * and red,green,blue are a white space separated list of floats
170 * in the range [0.0, 1.0]
172 void load_motifs(std::istream &);
173 //! load a list of motifs from a file named filename
174 void load_motifs(boost::filesystem::path filename);
175 //! return our motifs;
176 const std::set<Sequence>& motifs() const;
178 //! return color mapper
179 boost::shared_ptr<AnnotationColors> colorMapper();
182 //! push motifs to our attached sequences
183 void update_sequences_motifs();
186 // parameters needed for a mussa analysis
187 //! name of this analysis. (will also be used when saving an analysis)
188 std::string analysis_name;
189 //! how many base pairs to include in a sliding window
191 //! how many base pairs need to match order to record a window as conserved
193 //! stores current filter used by GUI to change the connections shown
195 //! which nway comparison algorithm to use.
196 enum analysis_modes ana_mode;
198 //! should we append _w<window_size> to the saved analysis
200 //! should we append _t<threshold> to the saved analysis
204 vector_sequence_type the_seqs;
206 std::vector<std::vector<FLPs> > all_comps;
207 //! N-way data, ie the mussa results
211 std::set<Sequence> motif_sequences;
213 boost::shared_ptr<AnnotationColors> color_mapper;
214 //! path to our analysis
215 boost::filesystem::path analysis_path;
216 //! flag indicating if we have unsaved changes
220 //! runs all the seqcomps needed to support the nway comparison