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 // ---------- flp.hh -----------
15 // ----------------------------------------
16 #include <boost/filesystem/path.hpp>
23 //! FLP = Fixed Length Pairs (Data)
25 * vector of linked lists of the match type struct
32 //! Setup a FLP and reserve space for the match lists
34 * Initialize the all_matches structure with a list of matches
35 * for enough windows for the size of sequence 1
43 friend bool operator==(const match&, const match& );
44 friend std::ostream &operator<<(std::ostream&, const match&);
47 void setup(int win_size, int hard_thres);
48 //! compare two sequences and store the list of results in all_matches
49 void seqcomp(std::string seq1, std::string seq2, bool is_RC);
50 //bool FLPs::match_less(match *match1, match *match2);
52 //! Return all the matches for a particular window?
53 std::list<match> matches(int index) const;
54 //! Return all the match indexes for a particular window?
55 std::list<int> match_locations(int index) const;
56 //! Return all the match indexes for a particular window above a threshold
57 std::list<int> thres_matches(int index, int thres) const;
58 //! Return the number of windows stored in this FLPs
59 /*! (this is should be the same as seq1_len-window_size+1
62 //! Save all the FLPs to save_file_path
63 void save(boost::filesystem::path save_file_path);
64 //! Load a vector of a lists of FLPs
65 void load(boost::filesystem::path file_path);
67 //! how far we are through current seqcomp
68 float progress() const;
70 const static int seqcomp_not_running = -1;
72 //! the number of base pairs in the sliding window
74 //! the minimum tnumber of base pairs need for this window to be saved.
77 //! add a match from location seq1_i to seq2_i with a threshold of a_score
78 /*! i2_offset is used to shift the window start when doing a reverse
79 * compliment. (This is called by seqcomp() )
81 inline void add(int seq1_i, int seq2_i, int a_score, int i2_offset);
82 //! this list of matches between the two sequences
83 /*! All of the matches are stored here, it appears each window position
84 * in sequence 1 gets an entry in the vector, the list contains matches
85 * between sequence 1 and sequence 2
87 std::vector<std::list<match> > *all_matches;
88 //! reserve space for the appropriate number of windows.
89 /*! this is mostly so seqcomp can use operator[]
91 void alloc_matches(std::string::size_type len1=0);
93 //! current loop index
95 //! end seqcomp index (when terminating, seqcomp_i == seqcomp_end.
96 //! when not running seqcomp_i == seqcomp_end == seqcomp_not_running