Alphabet.cpp

Go to the documentation of this file.
00001 /*
00002  * This program is free software; you can redistribute it and/or modify
00003  * it under the terms of the GNU General Public License as published by
00004  * the Free Software Foundation; either version 3 of the License, or
00005  * (at your option) any later version.
00006  *
00007  * Written (W) 2006-2009 Soeren Sonnenburg
00008  * Copyright (C) 2006-2009 Fraunhofer Institute FIRST and Max-Planck-Society
00009  */
00010 
00011 #include <string.h>
00012 #include <math.h>
00013 
00014 #include "features/Alphabet.h"
00015 #include "lib/io.h"
00016 
00017 //define numbers for the bases 
00018 const uint8_t CAlphabet::B_A=0;
00019 const uint8_t CAlphabet::B_C=1;
00020 const uint8_t CAlphabet::B_G=2;
00021 const uint8_t CAlphabet::B_T=3;
00022 const uint8_t CAlphabet::MAPTABLE_UNDEF=0xff;
00023 const char* CAlphabet::alphabet_names[11]={"DNA", "RAWDNA", "RNA", "PROTEIN", "ALPHANUM", "CUBE", "RAW", "IUPAC_NUCLEIC_ACID", "IUPAC_AMINO_ACID", "NONE", "UNKNOWN"};
00024 
00025 CAlphabet::CAlphabet(char* al, int32_t len)
00026 : CSGObject()
00027 {
00028     EAlphabet alpha=NONE;
00029 
00030     if (len>=(int32_t) strlen("DNA") && !strncmp(al, "DNA", strlen("DNA")))
00031         alpha = DNA;
00032     else if (len>=(int32_t) strlen("RAWDNA") && !strncmp(al, "RAWDNA", strlen("RAWDNA")))
00033         alpha = RAWDNA;
00034     else if (len>=(int32_t) strlen("RNA") && !strncmp(al, "RNA", strlen("RNA")))
00035         alpha = RNA;
00036     else if (len>=(int32_t) strlen("PROTEIN") && !strncmp(al, "PROTEIN", strlen("PROTEIN")))
00037         alpha = PROTEIN;
00038     else if (len>=(int32_t) strlen("ALPHANUM") && !strncmp(al, "ALPHANUM", strlen("ALPHANUM")))
00039         alpha = ALPHANUM;
00040     else if (len>=(int32_t) strlen("CUBE") && !strncmp(al, "CUBE", strlen("CUBE")))
00041         alpha = CUBE;
00042     else if ((len>=(int32_t) strlen("BYTE") && !strncmp(al, "BYTE", strlen("BYTE"))) || 
00043             (len>=(int32_t) strlen("RAW") && !strncmp(al, "RAW", strlen("RAW"))))
00044         alpha = RAWBYTE;
00045     else if (len>=(int32_t) strlen("IUPAC_NUCLEIC_ACID") && !strncmp(al, "IUPAC_NUCLEIC_ACID", strlen("IUPAC_NUCLEIC_ACID")))
00046         alpha = IUPAC_NUCLEIC_ACID;
00047     else if (len>=(int32_t) strlen("IUPAC_AMINO_ACID") && !strncmp(al, "IUPAC_AMINO_ACID", strlen("IUPAC_AMINO_ACID")))
00048         alpha = IUPAC_AMINO_ACID;
00049     else {
00050       SG_ERROR( "unknown alphabet %s\n", al);
00051    }
00052     
00053     set_alphabet(alpha);
00054 }
00055 
00056 CAlphabet::CAlphabet(EAlphabet alpha)
00057 : CSGObject()
00058 {
00059     set_alphabet(alpha);
00060 }
00061 
00062 CAlphabet::CAlphabet(CAlphabet* a)
00063 : CSGObject()
00064 {
00065     ASSERT(a);
00066     set_alphabet(a->get_alphabet());
00067     copy_histogram(a);
00068 }
00069 
00070 CAlphabet::~CAlphabet()
00071 {
00072 }
00073 
00074 bool CAlphabet::set_alphabet(EAlphabet alpha)
00075 {
00076     bool result=true;
00077     alphabet=alpha;
00078 
00079     switch (alphabet)
00080     {
00081         case DNA:
00082         case RAWDNA:
00083             num_symbols = 4;
00084             break;
00085         case RNA:
00086             num_symbols = 4;
00087             break;
00088         case PROTEIN:
00089             num_symbols = 26;
00090             break;
00091         case ALPHANUM:
00092             num_symbols = 36;
00093             break;
00094         case CUBE:
00095             num_symbols = 6;
00096             break;
00097         case RAWBYTE:
00098             num_symbols = 256;
00099             break;
00100         case IUPAC_NUCLEIC_ACID:
00101             num_symbols = 16;
00102             break;
00103         case IUPAC_AMINO_ACID:
00104             num_symbols = 23;
00105             break;
00106         case NONE:
00107             num_symbols = 0;
00108             break;
00109         default:
00110             num_symbols = 0;
00111             result=false;
00112             break;
00113     }
00114 
00115     num_bits=(int32_t) ceil(log((float64_t) num_symbols)/log((float64_t) 2));
00116     init_map_table();
00117     clear_histogram();
00118 
00119     SG_DEBUG( "initialised alphabet %s\n", get_alphabet_name(alphabet));
00120 
00121     return result;
00122 }
00123 
00124 void CAlphabet::init_map_table()
00125 {
00126     int32_t i;
00127     for (i=0; i<(1<<(8*sizeof(uint8_t))); i++)
00128     {
00129         maptable_to_bin[i] = MAPTABLE_UNDEF;
00130         maptable_to_char[i] = MAPTABLE_UNDEF;
00131         valid_chars[i] = false;
00132     }
00133 
00134     switch (alphabet)
00135     {
00136         case CUBE:
00137             valid_chars[(uint8_t) '1']=true;
00138             valid_chars[(uint8_t) '2']=true;
00139             valid_chars[(uint8_t) '3']=true;
00140             valid_chars[(uint8_t) '4']=true;    
00141             valid_chars[(uint8_t) '5']=true;    
00142             valid_chars[(uint8_t) '6']=true;    //Translation '123456' -> 012345
00143 
00144             maptable_to_bin[(uint8_t) '1']=0;
00145             maptable_to_bin[(uint8_t) '2']=1;
00146             maptable_to_bin[(uint8_t) '3']=2;
00147             maptable_to_bin[(uint8_t) '4']=3;   
00148             maptable_to_bin[(uint8_t) '5']=4;   
00149             maptable_to_bin[(uint8_t) '6']=5;   //Translation '123456' -> 012345
00150 
00151             maptable_to_char[(uint8_t) 0]='1';
00152             maptable_to_char[(uint8_t) 1]='2';
00153             maptable_to_char[(uint8_t) 2]='3';
00154             maptable_to_char[(uint8_t) 3]='4';
00155             maptable_to_char[(uint8_t) 4]='5';
00156             maptable_to_char[(uint8_t) 5]='6';  //Translation 012345->'123456'
00157             break;
00158 
00159         case PROTEIN:
00160             {
00161                 int32_t skip=0 ;
00162                 for (i=0; i<21; i++)
00163                 {
00164                     if (i==1) skip++ ;
00165                     if (i==8) skip++ ;
00166                     if (i==12) skip++ ;
00167                     if (i==17) skip++ ;
00168                     valid_chars['A'+i+skip]=true;
00169                     maptable_to_bin['A'+i+skip]=i ;
00170                     maptable_to_char[i]='A'+i+skip ;
00171                 } ;                   //Translation 012345->acde...xy -- the protein code
00172             } ;
00173             break;
00174 
00175         case ALPHANUM:
00176             {
00177                 for (i=0; i<26; i++)
00178                 {
00179                     valid_chars['A'+i]=true;
00180                     maptable_to_bin['A'+i]=i ;
00181                     maptable_to_char[i]='A'+i ;
00182                 } ;
00183                 for (i=0; i<10; i++)
00184                 {
00185                     valid_chars['0'+i]=true;
00186                     maptable_to_bin['0'+i]=26+i ;
00187                     maptable_to_char[26+i]='0'+i ;
00188                 } ;        //Translation 012345->acde...xy0123456789
00189             } ;
00190             break;
00191 
00192         case RAWBYTE:
00193             {
00194                 //identity
00195                 for (i=0; i<256; i++)
00196                 {
00197                     valid_chars[i]=true;
00198                     maptable_to_bin[i]=i;
00199                     maptable_to_char[i]=i;
00200                 }
00201             }
00202             break;
00203 
00204         case DNA:
00205             valid_chars[(uint8_t) 'A']=true;
00206             valid_chars[(uint8_t) 'C']=true;
00207             valid_chars[(uint8_t) 'G']=true;
00208             valid_chars[(uint8_t) 'T']=true;    
00209 
00210             maptable_to_bin[(uint8_t) 'A']=B_A;
00211             maptable_to_bin[(uint8_t) 'C']=B_C;
00212             maptable_to_bin[(uint8_t) 'G']=B_G;
00213             maptable_to_bin[(uint8_t) 'T']=B_T; 
00214 
00215             maptable_to_char[B_A]='A';
00216             maptable_to_char[B_C]='C';
00217             maptable_to_char[B_G]='G';
00218             maptable_to_char[B_T]='T';
00219             break;
00220         case RAWDNA:
00221             {
00222                 //identity
00223                 for (i=0; i<4; i++)
00224                 {
00225                     valid_chars[i]=true;
00226                     maptable_to_bin[i]=i;
00227                     maptable_to_char[i]=i;
00228                 }
00229             }
00230             break;
00231 
00232         case RNA:
00233             valid_chars[(uint8_t) 'A']=true;
00234             valid_chars[(uint8_t) 'C']=true;
00235             valid_chars[(uint8_t) 'G']=true;
00236             valid_chars[(uint8_t) 'U']=true;    
00237 
00238             maptable_to_bin[(uint8_t) 'A']=B_A;
00239             maptable_to_bin[(uint8_t) 'C']=B_C;
00240             maptable_to_bin[(uint8_t) 'G']=B_G;
00241             maptable_to_bin[(uint8_t) 'U']=B_T; 
00242 
00243             maptable_to_char[B_A]='A';
00244             maptable_to_char[B_C]='C';
00245             maptable_to_char[B_G]='G';
00246             maptable_to_char[B_T]='U';
00247             break;
00248 
00249         case IUPAC_NUCLEIC_ACID:
00250             valid_chars[(uint8_t) 'A']=true; // A   Adenine
00251             valid_chars[(uint8_t) 'C']=true; // C   Cytosine
00252             valid_chars[(uint8_t) 'G']=true; // G   Guanine
00253             valid_chars[(uint8_t) 'T']=true; // T   Thymine
00254             valid_chars[(uint8_t) 'U']=true; // U   Uracil
00255             valid_chars[(uint8_t) 'R']=true; // R   Purine (A or G)
00256             valid_chars[(uint8_t) 'Y']=true; // Y   Pyrimidine (C, T, or U)
00257             valid_chars[(uint8_t) 'M']=true; // M   C or A
00258             valid_chars[(uint8_t) 'K']=true; // K   T, U, or G
00259             valid_chars[(uint8_t) 'W']=true; // W   T, U, or A
00260             valid_chars[(uint8_t) 'S']=true; // S   C or G
00261             valid_chars[(uint8_t) 'B']=true; // B   C, T, U, or G (not A)
00262             valid_chars[(uint8_t) 'D']=true; // D   A, T, U, or G (not C)
00263             valid_chars[(uint8_t) 'H']=true; // H   A, T, U, or C (not G)
00264             valid_chars[(uint8_t) 'V']=true; // V   A, C, or G (not T, not U)
00265             valid_chars[(uint8_t) 'N']=true; // N   Any base (A, C, G, T, or U)
00266 
00267             maptable_to_bin[(uint8_t) 'A']=0; // A  Adenine
00268             maptable_to_bin[(uint8_t) 'C']=1; // C  Cytosine
00269             maptable_to_bin[(uint8_t) 'G']=2; // G  Guanine
00270             maptable_to_bin[(uint8_t) 'T']=3; // T  Thymine
00271             maptable_to_bin[(uint8_t) 'U']=4; // U  Uracil
00272             maptable_to_bin[(uint8_t) 'R']=5; // R  Purine (A or G)
00273             maptable_to_bin[(uint8_t) 'Y']=6; // Y  Pyrimidine (C, T, or U)
00274             maptable_to_bin[(uint8_t) 'M']=7; // M  C or A
00275             maptable_to_bin[(uint8_t) 'K']=8; // K  T, U, or G
00276             maptable_to_bin[(uint8_t) 'W']=9; // W  T, U, or A
00277             maptable_to_bin[(uint8_t) 'S']=10; // S C or G
00278             maptable_to_bin[(uint8_t) 'B']=11; // B C, T, U, or G (not A)
00279             maptable_to_bin[(uint8_t) 'D']=12; // D A, T, U, or G (not C)
00280             maptable_to_bin[(uint8_t) 'H']=13; // H A, T, U, or C (not G)
00281             maptable_to_bin[(uint8_t) 'V']=14; // V A, C, or G (not T, not U)
00282             maptable_to_bin[(uint8_t) 'N']=15; // N Any base (A, C, G, T, or U)
00283 
00284             maptable_to_char[0]=(uint8_t) 'A'; // A Adenine
00285             maptable_to_char[1]=(uint8_t) 'C'; // C Cytosine
00286             maptable_to_char[2]=(uint8_t) 'G'; // G Guanine
00287             maptable_to_char[3]=(uint8_t) 'T'; // T Thymine
00288             maptable_to_char[4]=(uint8_t) 'U'; // U Uracil
00289             maptable_to_char[5]=(uint8_t) 'R'; // R Purine (A or G)
00290             maptable_to_char[6]=(uint8_t) 'Y'; // Y Pyrimidine (C, T, or U)
00291             maptable_to_char[7]=(uint8_t) 'M'; // M C or A
00292             maptable_to_char[8]=(uint8_t) 'K'; // K T, U, or G
00293             maptable_to_char[9]=(uint8_t) 'W'; // W T, U, or A
00294             maptable_to_char[10]=(uint8_t) 'S'; // S    C or G
00295             maptable_to_char[11]=(uint8_t) 'B'; // B    C, T, U, or G (not A)
00296             maptable_to_char[12]=(uint8_t) 'D'; // D    A, T, U, or G (not C)
00297             maptable_to_char[13]=(uint8_t) 'H'; // H    A, T, U, or C (not G)
00298             maptable_to_char[14]=(uint8_t) 'V'; // V    A, C, or G (not T, not U)
00299             maptable_to_char[15]=(uint8_t) 'N'; // N    Any base (A, C, G, T, or U)
00300             break;
00301 
00302         case IUPAC_AMINO_ACID:
00303             valid_chars[(uint8_t) 'A']=true; //A    Ala Alanine
00304             valid_chars[(uint8_t) 'R']=true; //R    Arg Arginine
00305             valid_chars[(uint8_t) 'N']=true; //N    Asn Asparagine
00306             valid_chars[(uint8_t) 'D']=true; //D    Asp Aspartic acid
00307             valid_chars[(uint8_t) 'C']=true; //C    Cys Cysteine
00308             valid_chars[(uint8_t) 'Q']=true; //Q    Gln Glutamine
00309             valid_chars[(uint8_t) 'E']=true; //E    Glu Glutamic acid
00310             valid_chars[(uint8_t) 'G']=true; //G    Gly Glycine
00311             valid_chars[(uint8_t) 'H']=true; //H    His Histidine
00312             valid_chars[(uint8_t) 'I']=true; //I    Ile Isoleucine
00313             valid_chars[(uint8_t) 'L']=true; //L    Leu Leucine
00314             valid_chars[(uint8_t) 'K']=true; //K    Lys Lysine
00315             valid_chars[(uint8_t) 'M']=true; //M    Met Methionine
00316             valid_chars[(uint8_t) 'F']=true; //F    Phe Phenylalanine
00317             valid_chars[(uint8_t) 'P']=true; //P    Pro Proline
00318             valid_chars[(uint8_t) 'S']=true; //S    Ser Serine
00319             valid_chars[(uint8_t) 'T']=true; //T    Thr Threonine
00320             valid_chars[(uint8_t) 'W']=true; //W    Trp Tryptophan
00321             valid_chars[(uint8_t) 'Y']=true; //Y    Tyr Tyrosine
00322             valid_chars[(uint8_t) 'V']=true; //V    Val Valine
00323             valid_chars[(uint8_t) 'B']=true; //B    Asx Aspartic acid or Asparagine
00324             valid_chars[(uint8_t) 'Z']=true; //Z    Glx Glutamine or Glutamic acid
00325             valid_chars[(uint8_t) 'X']=true; //X    Xaa Any amino acid
00326 
00327             maptable_to_bin[(uint8_t) 'A']=0;  //A  Ala Alanine
00328             maptable_to_bin[(uint8_t) 'R']=1;  //R  Arg Arginine
00329             maptable_to_bin[(uint8_t) 'N']=2;  //N  Asn Asparagine
00330             maptable_to_bin[(uint8_t) 'D']=3;  //D  Asp Aspartic acid
00331             maptable_to_bin[(uint8_t) 'C']=4;  //C  Cys Cysteine
00332             maptable_to_bin[(uint8_t) 'Q']=5;  //Q  Gln Glutamine
00333             maptable_to_bin[(uint8_t) 'E']=6;  //E  Glu Glutamic acid
00334             maptable_to_bin[(uint8_t) 'G']=7;  //G  Gly Glycine
00335             maptable_to_bin[(uint8_t) 'H']=8;  //H  His Histidine
00336             maptable_to_bin[(uint8_t) 'I']=9;  //I  Ile Isoleucine
00337             maptable_to_bin[(uint8_t) 'L']=10; //L  Leu Leucine
00338             maptable_to_bin[(uint8_t) 'K']=11; //K  Lys Lysine
00339             maptable_to_bin[(uint8_t) 'M']=12; //M  Met Methionine
00340             maptable_to_bin[(uint8_t) 'F']=13; //F  Phe Phenylalanine
00341             maptable_to_bin[(uint8_t) 'P']=14; //P  Pro Proline
00342             maptable_to_bin[(uint8_t) 'S']=15; //S  Ser Serine
00343             maptable_to_bin[(uint8_t) 'T']=16; //T  Thr Threonine
00344             maptable_to_bin[(uint8_t) 'W']=17; //W  Trp Tryptophan
00345             maptable_to_bin[(uint8_t) 'Y']=18; //Y  Tyr Tyrosine
00346             maptable_to_bin[(uint8_t) 'V']=19; //V  Val Valine
00347             maptable_to_bin[(uint8_t) 'B']=20; //B  Asx Aspartic acid or Asparagine
00348             maptable_to_bin[(uint8_t) 'Z']=21; //Z  Glx Glutamine or Glutamic acid
00349             maptable_to_bin[(uint8_t) 'X']=22; //X  Xaa Any amino acid
00350 
00351             maptable_to_char[0]=(uint8_t) 'A';  //A Ala Alanine
00352             maptable_to_char[1]=(uint8_t) 'R';  //R Arg Arginine
00353             maptable_to_char[2]=(uint8_t) 'N';  //N Asn Asparagine
00354             maptable_to_char[3]=(uint8_t) 'D';  //D Asp Aspartic acid
00355             maptable_to_char[4]=(uint8_t) 'C';  //C Cys Cysteine
00356             maptable_to_char[5]=(uint8_t) 'Q';  //Q Gln Glutamine
00357             maptable_to_char[6]=(uint8_t) 'E';  //E Glu Glutamic acid
00358             maptable_to_char[7]=(uint8_t) 'G';  //G Gly Glycine
00359             maptable_to_char[8]=(uint8_t) 'H';  //H His Histidine
00360             maptable_to_char[9]=(uint8_t) 'I';  //I Ile Isoleucine
00361             maptable_to_char[10]=(uint8_t) 'L'; //L Leu Leucine
00362             maptable_to_char[11]=(uint8_t) 'K'; //K Lys Lysine
00363             maptable_to_char[12]=(uint8_t) 'M'; //M Met Methionine
00364             maptable_to_char[13]=(uint8_t) 'F'; //F Phe Phenylalanine
00365             maptable_to_char[14]=(uint8_t) 'P'; //P Pro Proline
00366             maptable_to_char[15]=(uint8_t) 'S'; //S Ser Serine
00367             maptable_to_char[16]=(uint8_t) 'T'; //T Thr Threonine
00368             maptable_to_char[17]=(uint8_t) 'W'; //W Trp Tryptophan
00369             maptable_to_char[18]=(uint8_t) 'Y'; //Y Tyr Tyrosine
00370             maptable_to_char[19]=(uint8_t) 'V'; //V Val Valine
00371             maptable_to_char[20]=(uint8_t) 'B'; //B Asx Aspartic acid or Asparagine
00372             maptable_to_char[21]=(uint8_t) 'Z'; //Z Glx Glutamine or Glutamic acid
00373             maptable_to_char[22]=(uint8_t) 'X'; //X Xaa Any amino acid
00374         default:
00375             break; //leave uninitialised
00376     };
00377 }
00378 
00379 void CAlphabet::clear_histogram()
00380 {
00381     memset(histogram, 0, sizeof(histogram));
00382     print_histogram();
00383 }
00384 
00385 int32_t CAlphabet::get_max_value_in_histogram()
00386 {
00387     int32_t max_sym=-1;
00388     for (int32_t i=(int32_t) (1 <<(sizeof(uint8_t)*8))-1;i>=0; i--)
00389     {
00390         if (histogram[i])
00391         {
00392             max_sym=i;
00393             break;
00394         }
00395     }
00396 
00397     return max_sym;
00398 }
00399 
00400 int32_t CAlphabet::get_num_symbols_in_histogram()
00401 {
00402     int32_t num_sym=0;
00403     for (int32_t i=0; i<(int32_t) (1 <<(sizeof(uint8_t)*8)); i++)
00404     {
00405         if (histogram[i])
00406             num_sym++;
00407     }
00408 
00409     return num_sym;
00410 }
00411 
00412 int32_t CAlphabet::get_num_bits_in_histogram()
00413 {
00414     int32_t num_sym=get_num_symbols_in_histogram();
00415     if (num_sym>0)
00416         return (int32_t) ceil(log((float64_t) num_sym)/log((float64_t) 2));
00417     else
00418         return 0;
00419 }
00420 
00421 void CAlphabet::print_histogram()
00422 {
00423     for (int32_t i=0; i<(int32_t) (1 <<(sizeof(uint8_t)*8)); i++)
00424     {
00425         if (histogram[i])
00426             SG_PRINT( "hist[%d]=%lld\n", i, histogram[i]);
00427     }
00428 }
00429 
00430 bool CAlphabet::check_alphabet(bool print_error)
00431 {
00432     bool result = true;
00433 
00434     for (int32_t i=0; i<(int32_t) (1 <<(sizeof(uint8_t)*8)); i++)
00435     {
00436         if (histogram[i]>0 && valid_chars[i]==0)
00437         {
00438             result=false;
00439             break;
00440         }
00441     }
00442 
00443     if (!result && print_error)
00444     {
00445         print_histogram();
00446         SG_ERROR( "ALPHABET does not contain all symbols in histogram\n");
00447     }
00448 
00449     return result;
00450 }
00451 
00452 bool CAlphabet::check_alphabet_size(bool print_error)
00453 {
00454     if (get_num_bits_in_histogram() > get_num_bits())
00455     {
00456         if (print_error)
00457         {
00458             print_histogram();
00459             fprintf(stderr, "get_num_bits_in_histogram()=%i > get_num_bits()=%i\n", get_num_bits_in_histogram(), get_num_bits()) ;
00460          SG_ERROR( "ALPHABET too small to contain all symbols in histogram\n");
00461         }
00462         return false;
00463     }
00464     else
00465         return true;
00466 
00467 }
00468 
00469 void CAlphabet::copy_histogram(CAlphabet* a)
00470 {
00471     memcpy(histogram, a->get_histogram(), sizeof(histogram));
00472 }
00473 
00474 const char* CAlphabet::get_alphabet_name(EAlphabet alphabet)
00475 {
00476     
00477     int32_t idx;
00478     switch (alphabet)
00479     {
00480         case DNA:
00481             idx=0;
00482             break;
00483         case RAWDNA:
00484             idx=1;
00485             break;
00486         case RNA:
00487             idx=2;
00488             break;
00489         case PROTEIN:
00490             idx=3;
00491             break;
00492         case ALPHANUM:
00493             idx=4;
00494             break;
00495         case CUBE:
00496             idx=5;
00497             break;
00498         case RAWBYTE:
00499             idx=6;
00500             break;
00501         case IUPAC_NUCLEIC_ACID:
00502             idx=7;
00503             break;
00504         case IUPAC_AMINO_ACID:
00505             idx=8;
00506             break;
00507         case NONE:
00508             idx=9;
00509             break;
00510         default:
00511             idx=10;
00512             break;
00513     }
00514     return alphabet_names[idx];
00515 }

SHOGUN Machine Learning Toolbox - Documentation