BGI/BASE/src/2bwt/DNACount.h

   1 /*
   2
   3    DNACount.h        DNA Count
   4
   5    This module contains DNA occurrence counting functions. The DNA must be
   6    in word-packed format.
   7
   8    Copyright (C) 2004, Wong Chi Kwong.
   9
  10    This program is free software; you can redistribute it and/or
  11    modify it under the terms of the GNU General Public License
  12    as published by the Free Software Foundation; either version 2
  13    of the License, or (at your option) any later version.
  14
  15    This program is distributed in the hope that it will be useful,
  16    but WITHOUT ANY WARRANTY; without even the implied warranty of
  17    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18    GNU General Public License for more details.
  19
  20    You should have received a copy of the GNU General Public License
  21    along with this program; if not, write to the Free Software
  22    Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
  23
  24 */
  25
  26 #ifndef __DNA_COUNT_H__
  27 #define __DNA_COUNT_H__
  28
  29 #include "TypeNLimit.h"
  30
  31 // DNA
  32 #define DNA_ALPHABET_SIZE            8
  33 #define DNA_CHAR_PER_WORD            8
  34 #define DNA_BIT_PER_CHAR            4
  35
  36 // DNA occurrence count table
  37 #define DNA_OCC_CNT_TABLE_SIZE_IN_WORD    65536
  38 #define DNA_OCC_SUM_EXCEPTION(sum)            ((sum & 0xeeeeeeeeeeeeeeef) == 0) //fefefeff
  39
  40 // DNA with 'n'
  41 #define DNA_N_ALPHABET_SIZE            5
  42 #define DNA_N_CHAR_PER_WORD            10
  43 #define DNA_N_BIT_PER_CHAR            3
  44
  45 // DNA with 'n' occurrence count table
  46 #define DNA_N_OCC_CNT_TABLE_SIZE_IN_WORD    32786
  47
  48
  49 void GenerateDNAOccCountTable(unsigned long long *dnaDecodeTable);
  50
  51 // The following functions can only count up to 255 characters
  52 unsigned long long ForwardDNAOccCount(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  53 unsigned long long BackwardDNAOccCount(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  54 void ForwardDNAAllOccCount(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict occCount, const unsigned long long* dnaDecodeTable);
  55 void BackwardDNAAllOccCount(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  56 unsigned long long Forward1OccCount(const unsigned int* bitVector, const unsigned long long index, const unsigned long long* dnaDecodeTable);    // Count number of 1 bit
  57 unsigned long long Backward1OccCount(const unsigned int* bitVector, const unsigned long long index, const unsigned long long* dnaDecodeTable); // Count number of 1 bit
  58
  59 // The following functions have no limit on the number of characters
  60 unsigned long long ForwardDNAOccCountNoLimit(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  61 unsigned long long BackwardDNAOccCountNoLimit(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  62 void ForwardDNAAllOccCountNoLimit(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  63 void BackwardDNAAllOccCountNoLimit(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  64
  65
  66 void GenerateDNA_NOccCountTable(unsigned int *dnaDecodeTable);
  67
  68 // The following functions have no limit on the number of characters
  69 unsigned long long ForwardDNA_NOccCount(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  70 unsigned long long BackwardDNA_NOccCount(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  71 void ForwardDNA_NAllOccCount(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  72 void BackwardDNA_NAllOccCount(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  73
  74 // The following functions have no limit on the number of characters
  75 unsigned long long ForwardDNAnOccCountNoLimit(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  76 unsigned long long BackwardDNA_NOccCountNoLimit(const unsigned int* dna, const unsigned long long index, const unsigned int character, const unsigned long long* dnaDecodeTable);
  77 void ForwardDNA_NAllOccCountNoLimit(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  78 void BackwardDNA_NAllOccCountNoLimit(const unsigned int* dna, const unsigned long long index, unsigned long long* __restrict  occCount, const unsigned long long* dnaDecodeTable);
  79
  80 // The first character from startAddr is indexed as 1
  81 // DNA_NAllOccCount only count occurrence from character 0 to 3
  82
  83 // The following functions work for any word packed text
  84 unsigned long long ForwardOccCount(const unsigned int* packed, const unsigned long long index, const unsigned int character, const unsigned int alphabetSize);
  85 unsigned long long BackwardOccCount(const unsigned int* packed, const unsigned long long index, const unsigned int character, const unsigned int alphabetSize);
  86 void ForwardAllOccCount(const unsigned int* packed, const unsigned long long index, const unsigned int alphabetSize, unsigned long long* occCount);
  87 void BackwardAllOccCount(const unsigned int* packed, const unsigned long long index, const unsigned int alphabetSize, unsigned long long* occCount);
  88
  89
  90 #endif
  91