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