BGI/SOAPdenovo2/sparsePregraph/inc/build_edge.h

   1 /*
   2  * inc/sparse_kmer.h
   3  *
   4  * Copyright (c) 2008-2012 BGI-Shenzhen <soap at genomics dot org dot cn>.
   5  *
   6  * This file is part of SOAPdenovo.
   7  *
   8  * SOAPdenovo is free software: you can redistribute it and/or modify
   9  * it under the terms of the GNU General Public License as published by
  10  * the Free Software Foundation, either version 3 of the License, or
  11  * (at your option) any later version.
  12  *
  13  * SOAPdenovo is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License
  19  * along with SOAPdenovo.  If not, see <http://www.gnu.org/licenses/>.
  20  *
  21  */
  22
  23 #ifndef _BUILD_EDGE_H
  24
  25 #define _BUILD_EDGE_H
  26
  27 void removeMinorTips ( struct hashtable2 * ht, int K_size, int cut_len_tip, int & tip_c );
  28 void kmer2edges ( hashtable2 * ht, int K_size, char * outfile );
  29 void convert ( char * sparse_edge_file, int K_size, char * output_prefix );
  30 void RemovingWeakNodesAndEdges2 ( hashtable2 * ht, int K_size, int NodeCovTh, int EdgeCovTh, size_t * bucket_cnt, size_t * edge_cnt );
  31
  32 struct stacked_node2
  33 {
  34         struct bucket2 * node;
  35         bool is_left; // change it to a byte later
  36         struct edge_node * edge;
  37         struct stacked_node2 * next;
  38 };
  39
  40 typedef struct preedge2
  41 {
  42         struct stacked_node2 * from_node;
  43         struct stacked_node2 * to_node;
  44         string * full_edge;
  45         unsigned short cvg;
  46         unsigned short bal_edge: 2;
  47
  48 } preEDGE2;
  49
  50
  51 // below is static methods
  52 //remove minor tips ...
  53 //void removeMinorTips(struct hashtable2 *ht,int K_size,int cut_len_tip,int &tip_c);
  54 static void mask1in1out ( hashtable2 * ht );
  55 static int clipTipFromNode ( hashtable2 * ht, int K_size, bucket2 * node, int cut_len_tip );
  56 static int count_left_edge_num ( bucket2 * bkt );
  57 static int count_right_edge_num ( bucket2 * bkt );
  58
  59 static void dislink ( hashtable2 * ht, int K_size, stacked_node2 * from_node );
  60 static bucket2 * lastKmer ( hashtable2 * ht, int K_size, bucket2 * node, edge_node * edge, int is_left, int & smaller );
  61 //static bucket2* search_kmer(hashtable2 *ht,uint64_t* t_kmer, int Kmer_arr_sz); old
  62 static bucket2 * search_kmer ( hashtable2 * ht, kmer_t2 * t_kmer );
  63
  64 static void removeEdge ( bucket2 * node, edge_node * edge, int is_left );
  65 static void stat_edge_num ( hashtable2 * ht );
  66 static void stat_edge_cvg_len ( hashtable2 * ht );
  67 static bool isSmaller2 ( uint64_t * kmer, int K_size );
  68
  69 //kmer2edges ....
  70 //void kmer2edges(hashtable2* ht,int K_size,char *outfile);
  71 static void make_edge ( hashtable2 * ht, int K_size, FILE * fp );
  72 static int startEdgeFromNode ( hashtable2 * ht, int K_size, bucket2 * node, FILE * fp );
  73 static void stringBeads ( hashtable2 * ht, int K_size, list<stacked_node2 *> &stack, stacked_node2 * from_node, edge_node * from_edge, int * node_c );
  74
  75 static void process_1stack ( hashtable2 * ht, int K_size, list<stacked_node2 *> &stack, FILE * fp, vector<preEDGE2> &loops_edges );
  76
  77 //static void  get_kmer(const char * seq,int len, int K_size,int pos,uint64_t *kmer,int arr_sz );
  78 static void output_1edge ( preEDGE2 * long_edge, int K_size, FILE * fp );
  79 static string stack2string ( hashtable2 * ht, int K_size, list<stacked_node2 *> & stack );
  80 static bool check_palindrome ( string & str );
  81 static string revCompSeq ( const string & str );
  82
  83 //convert the edge fomat ...
  84 //void convert(char * sparse_edge_file,int K_size, char * output_prefix);
  85 static void convert_kmer ( uint64_t * sparse_kmer, int K_size, int arr_sz );
  86 static  uint64_t * fastReverseComp ( uint64_t * seq_arr, int seq_size, int arr_sz );
  87
  88 #endif
  89
  90
  91
  92
  93
  94
  95
  96