1 /* 2 This file is part of BioD. 3 Copyright (C) 2012 Artem Tarasov <lomereiter@gmail.com> 4 5 Permission is hereby granted, free of charge, to any person obtaining a 6 copy of this software and associated documentation files (the "Software"), 7 to deal in the Software without restriction, including without limitation 8 the rights to use, copy, modify, merge, publish, distribute, sublicense, 9 and/or sell copies of the Software, and to permit persons to whom the 10 Software is furnished to do so, subject to the following conditions: 11 12 The above copyright notice and this permission notice shall be included in 13 all copies or substantial portions of the Software. 14 15 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 20 FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 21 DEALINGS IN THE SOFTWARE. 22 23 */ 24 module bio.core.base; 25 26 import bio.core.tinymap; 27 import std.traits; 28 29 /// Code common to both Base5 and Base16 30 mixin template CommonBaseOperations() { 31 /// Convert to char 32 char asCharacter() @property const { return _code2char[_code]; } 33 /// 34 alias asCharacter this; 35 36 } 37 38 /// Base representation supporting full set of IUPAC codes 39 struct Base { 40 mixin TinyMapInterface!16; 41 42 private enum ubyte[256] _char2code = [ 43 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 44 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 45 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 46 1, 2, 4, 8, 15,15,15,15, 15,15,15,15, 15, 0,15,15, 47 48 15, 1,14, 2, 13,15,15, 4, 11,15,15,12, 15, 3,15,15, 49 15,15, 5, 6, 8,15, 7, 9, 15,10,15,15, 15,15,15,15, 50 15, 1,14, 2, 13,15,15, 4, 11,15,15,12, 15, 3,15,15, 51 15,15, 5, 6, 8,15, 7, 9, 15,10,15,15, 15,15,15,15, 52 53 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 54 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 55 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 56 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 57 58 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 59 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 60 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15, 61 15,15,15,15, 15,15,15,15, 15,15,15,15, 15,15,15,15 62 ]; 63 64 // = 0000 65 // 66 // A 0001 67 // C 0010 68 // G 0100 69 // T 1000 70 // 71 // W 1001 (A T) Weak 72 // S 0110 (C G) Strong 73 // 74 // M 0011 (A C) aMino 75 // K 1100 (G T) Keto 76 // R 0101 (A G) puRine 77 // Y 1010 (A G) pYrimidine 78 // 79 // B 1110 (not A) 80 // D 1101 (not C) 81 // H 1011 (not G) 82 // V 0111 (not T) 83 // 84 // N 1111 (aNy base) 85 private enum _code2char = "=ACMGRSVTWYHKDBN"; 86 87 private enum ubyte[16] _complement_table = [0x0, 0x8, 0x4, 0xC, 88 0x2, 0xA, 0x6, 0xE, 89 0x1, 0x9, 0x5, 0xD, 90 0x3, 0xB, 0x7, 0xF]; 91 /// Complementary base 92 Base complement() @property const { 93 // take the code, reverse the bits, and return the base 94 return Base.fromInternalCode(_complement_table[_code]); 95 } 96 97 unittest { 98 import std.ascii; 99 100 foreach (i, c; _code2char) { 101 assert(_char2code[c] == i); 102 } 103 104 foreach (c; 0 .. 256) { 105 auto c2 = _code2char[_char2code[c]]; 106 if (c2 != 'N') { 107 if ('0' <= c && c <= '9') { 108 assert(c2 == "ACGT"[c - '0']); 109 } else { 110 assert(c2 == toUpper(c)); 111 } 112 } 113 } 114 } 115 116 mixin CommonBaseOperations; 117 /// Construct from IUPAC code 118 this(char c) { 119 _code = _char2code[cast(ubyte)c]; 120 } 121 122 /// ditto 123 this(dchar c) { 124 _code = _char2code[cast(ubyte)c]; 125 } 126 127 private enum ubyte[5] nt5_to_nt16 = [1, 2, 4, 8, 15]; 128 private static Base fromBase5(Base5 base) { 129 Base b = void; 130 b._code = nt5_to_nt16[base.internal_code]; 131 return b; 132 } 133 134 /// Conversion to Base5 135 Base5 opCast(T)() const 136 if (is(T == Base5)) 137 { 138 return Base5.fromBase16(this); 139 } 140 141 T opCast(T)() const 142 if (is(Unqual!T == char) || is(Unqual!T == dchar)) 143 { 144 return asCharacter; 145 } 146 } 147 148 unittest { 149 Base b = 'W'; 150 assert(b == 'W'); 151 152 b = Base.fromInternalCode(0); 153 assert(b == '='); 154 } 155 156 alias Base Base16; 157 158 /// Base representation supporting only 'A', 'C', 'G', 'T', and 'N' 159 /// (internal codes are 0, 1, 2, 3, and 4 correspondingly) 160 struct Base5 { 161 mixin TinyMapInterface!5; 162 163 private enum ubyte[256] _char2code = [ 164 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 165 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 166 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 167 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 168 169 4, 0, 4, 1, 4, 4, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 170 4, 4, 4, 4, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 171 4, 0, 4, 1, 4, 4, 4, 2, 4, 4, 4, 4, 4, 4, 4, 4, 172 4, 4, 4, 4, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 173 174 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 175 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 176 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 177 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 178 179 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 180 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 181 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 182 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4 183 ]; 184 185 private enum _code2char = "ACGTN"; 186 private enum ubyte[16] nt16_to_nt5 = [4, 0, 1, 4, 2, 4, 4, 4, 3, 4, 4, 4, 4, 4, 4, 4]; 187 188 mixin CommonBaseOperations; 189 190 /// Complementary base 191 Base5 complement() @property const { 192 return Base5.fromInternalCode(cast(ubyte)(_code == 4 ? 4 : (3 - _code))); 193 } 194 195 /// Construct base from one of "acgtACGT" symbols. 196 /// Every other character is converted to 'N' 197 this(char c) { 198 _code = _char2code[cast(ubyte)c]; 199 } 200 201 /// ditto 202 this(dchar c) { 203 _code = _char2code[cast(ubyte)c]; 204 } 205 206 private static Base5 fromBase16(Base16 base) { 207 Base5 b = void; 208 b._code = nt16_to_nt5[base.internal_code]; 209 return b; 210 } 211 212 /// Conversion to Base16 213 Base16 opCast(T)() const 214 if(is(T == Base16)) 215 { 216 return Base16.fromBase5(this); 217 } 218 219 T opCast(T)() const 220 if (is(Unqual!T == char) || is(Unqual!T == dchar)) 221 { 222 return asCharacter; 223 } 224 } 225 226 unittest { 227 auto b5 = Base5('C'); 228 assert(b5.internal_code == 1); 229 b5 = Base5.fromInternalCode(3); 230 assert(b5 == 'T'); 231 232 // doesn't work with std.conv.to 233 // 234 //import std.conv; 235 //assert(to!Base16(b5).internal_code == 8); 236 237 assert((cast(Base16)b5).internal_code == 8); 238 } 239 240 /// Complement base, which might be Base5, Base16, char, or dchar. 241 B complementBase(B)(B base) { 242 static if(is(Unqual!B == dchar) || is(Unqual!B == char)) 243 { 244 return cast(B)(Base16(base).complement); 245 } 246 else 247 return base.complement; 248 } 249 250 /// Convert character to base 251 template charToBase(B=Base16) 252 { 253 B charToBase(C)(C c) 254 if(is(Unqual!C == char) || is(Unqual!C == dchar)) 255 { 256 return B(c); 257 } 258 } 259 260 unittest { 261 assert(complementBase('T') == 'A'); 262 assert(complementBase('G') == 'C'); 263 264 assert(complementBase(Base5('A')) == Base5('T')); 265 assert(complementBase(Base16('C')) == Base16('G')); 266 267 assert(charToBase!Base16('A').complement == Base16('T')); 268 }