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>
26 #include "alg/annotation_colors.hpp"
27 #include "alg/mussa_callback.hpp"
28 #include "alg/nway_paths.hpp"
29 #include "alg/sequence.hpp"
31 std::string int_to_str(int an_int);
33 class Mussa : public QObject
38 //! call whatever signaling system we want
39 void progress(const std::string& description, int cur, int max);
42 enum analysis_modes { TransitiveNway, RadialNway, EntropyNway,
49 //! save the nway comparison
50 void save_muway(boost::filesystem::path save_path);
51 //! load a saved analysis directory
52 void load(boost::filesystem::path ana_path);
54 //! clear parameters and initialize data lists
57 //! set parameters from a file - 'mupa' ~ mussa parameters
58 void load_mupa_file(std::string para_file_path) { load_mupa_file(boost::filesystem::path(para_file_path));}
59 void load_mupa_file(boost::filesystem::path para_file_path);
61 // set parameters individually (eg from user input into gui classes)
63 void set_name(std::string a_name);
64 //! return name for this analysis
65 std::string get_name();
67 //! return number of sequences in this analyzis
68 /*! this returns either the_seqs.size() or seq_files.size()
69 * depending on which has data loaded in
70 * (silly delayed loading of sequence data)
74 //! set number of bases for this window size
75 void set_window(int a_window);
76 //! get number of bases for the sliding window
77 int get_window() const;
78 //! set number of bases that must match for a window to be saved
79 //! if threshold > soft_threshold this also sets soft_threshold
80 void set_threshold(int a_threshold);
81 //! get number of bases that must match for a window to be saved
82 int get_threshold() const;
83 //! sets the threshold used for computing the nway paths
84 //! must be in range [threshold..window size]
85 void set_soft_threshold(int sft_thres);
86 int get_soft_threshold() const;
88 void set_analysis_mode(enum analysis_modes new_ana_mode);
89 enum analysis_modes get_analysis_mode() const;
90 //! return a string name for an analysis mode
91 std::string get_analysis_mode_name() const;
93 //! return the refined paths found by the nway analysis.
94 const NwayPaths& paths() const;
96 //! given selected_paths, and view_paths, compute per base pair matches
97 //template <class IteratorT>
98 void createLocalAlignment(std::list<ConservedPath>::iterator begin,
99 std::list<ConservedPath>::iterator end,
100 std::list<ConservedPath::path_type>& result,
101 std::list<std::vector<bool> >& reversed);
103 //! run seqcomp and the nway filtering algorithm.
104 /*!analyze will run seqcomp and then the nway algorithm
105 * on whatever sequences have been loaded into this mussa instance.
106 * \throws mussa_analysis_error
109 /*! Run the nway filtering algorithm,
110 * this might be used when changing the soft threshhold?
114 //! appends a string sequence to the list of the_seqs
115 // void append_sequence(std::string a_seq);
116 //! appends a sequence to the list of the_seqs
117 void append_sequence(Sequence a_seq);
119 //! Load a sequence from a fasta file and any annotations
120 /*! \param[in] seq_file the full path to the fasta file
121 * \param[in] annot_file the full path to an annotation file,
122 * if is an empty string, we won't bother loading anything
123 * \param[in] fasta_index specify which sequence in a multisequence fasta
125 * \param[in] sub_seq_start starting slice index to select a subsequence
126 * use 0 start from the beginning.
127 * \param[in] sub_seq_end ending slice index to select a subsequence
128 * use 0 to go to the end.
130 void load_sequence(boost::filesystem::path seq_file,
131 boost::filesystem::path annot_file,
132 int fasta_index, int sub_seq_start=0, int sub_seq_end=0);
133 //! allow examining the sequences we have loaded
134 const std::vector<Sequence>& sequences() const;
136 // deprecated - support bridge for python version of mussa
137 // these save & load from the old file format
139 void load_old(char * load_file_path, int s_num);
141 // manage motif lists
142 //! add vector of motifs and colors to our motif collection
143 /*! this depends on sets and color maps being unique
144 * (aka if you add the same item more than once it doesn't
145 * increase the size of the data structure
147 void add_motifs(const std::vector<std::string>& motifs,
148 const std::vector<Color>& colors);
149 //! load motifs from an ifstream
150 /*! The file should look something like
151 * <sequence> <red> <green> <blue>
152 * where sequence is a string of IUPAC symbols
153 * and red,green,blue are a white space separated list of floats
154 * in the range [0.0, 1.0]
156 void load_motifs(std::istream &);
157 //! load a list of motifs from a file named filename
158 void load_motifs(boost::filesystem::path filename);
159 //! return our motifs;
160 const std::set<std::string>& motifs() const;
162 //! return color mapper
163 AnnotationColors& colorMapper();
166 //! push motifs to our attached sequences
167 void update_sequences_motifs();
170 // parameters needed for a mussa analysis
171 //! name of this analysis. (will also be used when saving an analysis)
172 std::string analysis_name;
173 //! how many base pairs to include in a sliding window
175 //! how many base pairs need to match order to record a window as conserved
178 //! which nway comparison algorithm to use.
179 enum analysis_modes ana_mode;
181 //! should we append _w<window_size> to the saved analysis
183 //! should we append _t<threshold> to the saved analysis
187 std::vector<Sequence> the_seqs;
189 std::vector<std::vector<FLPs> > all_comps;
190 //! N-way data, ie the mussa results
194 std::set<std::string> motif_sequences;
196 AnnotationColors color_mapper;
199 //! runs all the seqcomps needed to support the nway comparison