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 // ----------------------------------------
16 #include <boost/filesystem/path.hpp>
24 #include "alg/annotation_colors.hpp"
25 #include "alg/nway_paths.hpp"
26 #include "alg/sequence.hpp"
28 std::string int_to_str(int an_int);
32 friend class ConnWindow;
34 enum analysis_modes { TransitiveNway, RadialNway, EntropyNway,
41 //! save the nway comparison
42 void save_muway(boost::filesystem::path save_path);
43 //! load a saved analysis directory
44 void load(boost::filesystem::path ana_path);
46 //! clear parameters and initialize data lists
49 //! set parameters from a file - 'mupa' ~ mussa parameters
50 void load_mupa_file(std::string para_file_path) { load_mupa_file(boost::filesystem::path(para_file_path));}
51 void load_mupa_file(boost::filesystem::path para_file_path);
53 // set parameters individually (eg from user input into gui classes)
55 void set_name(std::string a_name);
56 //! return name for this analysis
57 std::string get_name();
59 //! return number of sequences in this analyzis
60 /*! this returns either the_seqs.size() or seq_files.size()
61 * depending on which has data loaded in
62 * (silly delayed loading of sequence data)
65 //! set number of bases for this window size
66 void set_window(int a_window);
67 //! get number of bases for the sliding window
68 int get_window() const;
69 //! set number of bases that must match for a window to be saved
70 //! if threshold > soft_threshold this also sets soft_threshold
71 void set_threshold(int a_threshold);
72 //! get number of bases that must match for a window to be saved
73 int get_threshold() const;
74 //! sets the threshold used for computing the nway paths
75 //! must be in range [threshold..window size]
76 void set_soft_threshold(int sft_thres);
77 int get_soft_threshold() const;
79 void set_analysis_mode(enum analysis_modes new_ana_mode);
80 enum analysis_modes get_analysis_mode() const;
81 //! return a string name for an analysis mode
82 std::string get_analysis_mode_name() const;
84 //! return the refined paths found by the nway analysis.
85 const NwayPaths& paths() const;
87 //! given selected_paths, and view_paths, compute per base pair matches
88 //template <class IteratorT>
89 void createLocalAlignment(std::list<ConservedPath>::iterator begin,
90 std::list<ConservedPath>::iterator end,
91 std::list<ConservedPath::path_type>& result,
92 std::list<std::vector<bool> >& reversed);
94 //! run seqcomp and the nway filtering algorithm.
95 /*!analyze will run seqcomp and then the nway algorithm
96 * on whatever sequences have been loaded into this mussa instance.
97 * w & t are for command line override functionality, set to 0 to ignore
98 * \throws mussa_analysis_error
100 void analyze(int w=0, int t=0,
101 enum analysis_modes ana_mode=TransitiveNway,
102 double ent_thres=0.0);
103 /*! Run the nway filtering algorithm,
104 * this might be used when changing the soft threshhold?
108 //! appends a string sequence to the list of the_seqs
109 void add_a_seq(std::string a_seq);
110 //! Load a sequence from a fasta file and any annotations
111 /*! \param[in] seq_file the full path to the fasta file
112 * \param[in] annot_file the full path to an annotation file,
113 * if is an empty string, we won't bother loading anything
114 * \param[in] fasta_index specify which sequence in a multisequence fasta
116 * \param[in] sub_seq_start starting slice index to select a subsequence
117 * use 0 start from the beginning.
118 * \param[in] sub_seq_end ending slice index to select a subsequence
119 * use 0 to go to the end.
121 void load_sequence(boost::filesystem::path seq_file,
122 boost::filesystem::path annot_file,
123 int fasta_index, int sub_seq_start=0, int sub_seq_end=0);
124 //! allow examining the sequences we have loaded
125 const std::vector<Sequence>& sequences() const;
127 // deprecated - support bridge for python version of mussa
128 // these save & load from the old file format
130 void load_old(char * load_file_path, int s_num);
132 // manage motif lists
133 //! add vector of motifs and colors to our motif collection
134 /*! this depends on sets and color maps being unique
135 * (aka if you add the same item more than once it doesn't
136 * increase the size of the data structure
138 void add_motifs(const std::vector<std::string>& motifs,
139 const std::vector<Color>& colors);
140 //! load motifs from an ifstream
141 /*! The file should look something like
142 * <sequence> <red> <green> <blue>
143 * where sequence is a string of IUPAC symbols
144 * and red,green,blue are a white space separated list of floats
145 * in the range [0.0, 1.0]
147 void load_motifs(std::istream &);
148 //! load a list of motifs from a file named filename
149 void load_motifs(boost::filesystem::path filename);
150 //! return our motifs;
151 const std::set<std::string>& motifs() const;
153 //! return color mapper
154 AnnotationColors& colorMapper();
157 //! push motifs to our attached sequences
158 void update_sequences_motifs();
161 // parameters needed for a mussa analysis
162 //! name of this analysis. (will also be used when saving an analysis)
163 std::string analysis_name;
164 //! how many base pairs to include in a sliding window
166 //! how many base pairs need to match order to record a window as conserved
169 //! which nway comparison algorithm to use.
170 enum analysis_modes ana_mode;
172 //! should we append _w<window_size> to the saved analysis
174 //! should we append _t<threshold> to the saved analysis
178 std::vector<Sequence> the_seqs;
180 std::vector<std::vector<FLPs> > all_comps;
181 //! N-way data, ie the mussa results
185 std::set<std::string> motif_sequences;
187 AnnotationColors color_mapper;
190 //! loads sequence and annotations from fasta and annotation file