static std::pair< std::string,std::vector<bool> > circularBwt(std::string const s, uint64_t const beg, uint64_t const len, int64_t const term, uint64_t * zrank = 0)
{
assert ( beg + len <= s.size() );
std::string const s2 = s+s;
// std::cerr << "[V] Block sorting...";
uint8_t const * utext = reinterpret_cast<uint8_t const *>(s2.c_str()) + beg;
typedef ::libmaus::suffixsort::DivSufSort<32,uint8_t *,uint8_t const *,int32_t *,int32_t const *> sort_type;
typedef sort_type::saidx_t saidx_t;
::libmaus::autoarray::AutoArray<saidx_t> SA(s2.size()-beg);
sort_type::divsufsort ( utext, SA.begin() , SA.size() );
// std::cerr << "done." << std::endl;
std::vector<bool> gt(len+1,false);
std::string bwt(len,' ');
uint64_t j = 0;
bool gtf = false;
for ( uint64_t i = 0; i < SA.size(); ++i )
if ( SA[i] < static_cast<saidx_t>(len) )
{
gt [ SA[i] ] = gtf;
#if defined(DEBUG)
assert ( gtf == (s2.substr(beg+SA[i]) > s2.substr(beg)) );
#endif
if ( SA[i] == 0 )
{
if ( zrank )
*zrank = j;
bwt[j++] = term; // 0;
gtf = true;
}
else
{
bwt[j++] = utext [ SA[i]-1 ];
}
}
#if defined(DEBUG)
std::cerr << "-----\n\n";
for ( uint64_t i = 0; i < SA.size(); ++i )
if ( SA[i] < len )
std::cerr << "[" << std::setw(2) << std::setfill('0') << SA[i] << std::setw(0) << "] = " << s2.substr(beg+SA[i]) << std::endl;
#endif
std::string const gtbackleft = s2.substr(beg+len);
std::string const gtbackright = s2.substr(beg);
#if defined(DEBUG)
std::cerr << "gt[len] = " << gtbackleft << " > " << gtbackright << " = " << (gtbackleft > gtbackright) << std::endl;
#endif
gt[len] = gtbackleft > gtbackright;
assert ( j == len );
return std::make_pair(bwt,gt);
}
TEST bwt_wikipedia_example1()
{
size_t index;
char *input = "^BANANA|";
char output[9];
bwt(input, output, &index, 8);
output[8] = '\0';
ASSERT_STR_EQ("BNN^AA|A", output);
PASS();
}
TEST bwt_drdobbs()
{
size_t index;
char *input = "DRDOBBS";
char output[8];
bwt(input, output, &index, 7);
output[7] = '\0';
ASSERT_EQ(3, index);
ASSERT_STR_EQ("OBRSDDB", output);
PASS();
}
TEST bwt_abracadabra()
{
size_t index;
char *input = "abracadabra";
char output[12];
bwt(input, output, &index, 11);
output[11] = '\0';
ASSERT_EQ(2, index);
ASSERT_STR_EQ("rdarcaaaabb", output);
PASS();
}
TEST bwt_bacabba()
{
size_t index;
char *input = "bacabba";
char output[8];
bwt(input, output, &index, 7);
output[7] = '\0';
ASSERT_EQ(4, index);
ASSERT_STR_EQ("bcbbaaa", output);
PASS();
}
TEST ibwt_wikipedia_example2()
{
size_t index;
char *input = "SIX.MIXED.PIXIES.SIFT.SIXTY.PIXIE.DUST.BOXES";
char encoded[45];
char decoded[45];
bwt(input, encoded, &index, 44);
ibwt(encoded, decoded, index, 44);
decoded[44] = '\0';
ASSERT_STR_EQ(input, decoded);
PASS();
}
TEST bwt_wikipedia_example2()
{
/* Not passing, possibly wrong answer */
SKIP();
size_t index;
char *input = "SIX.MIXED.PIXIES.SIFT.SIXTY.PIXIE.DUST.BOXES";
char output[45];
bwt(input, output, &index, 44);
output[44] = '\0';
ASSERT_STR_EQ("TEXYDST.E.IXXIIXXSSMPPS.B..E.S.UESFXDIIOIIIT", output);
PASS();
}
void CompTool::search(int argc, char** argv){
if(argc < 3) {cout << "File not enough" << endl; exit(1);}
const string seq1_file = argv[1];
const string seq2_file = argv[2];
const string seq1_name = basename(seq1_file);
const string seq2_name = basename(seq2_file);
// Options
int kmer_size = 15;
int slide_letters = 1;
int bwt_interval = 1;
int max_num_matches = 1000000000;
bool search_forward = true;
bool search_reverse = true;
if(argc > 3){
for(int i = 3; i < argc; i++){
if (argv[i][1] == 'k') kmer_size = atoi(argv[++i]);
else if(argv[i][1] == 'l') slide_letters = atoi(argv[++i]);
else if(argv[i][1] == 'i') bwt_interval = atoi(argv[++i]);
else if(argv[i][1] == 'm') max_num_matches = atoi(argv[++i]);
else if(argv[i][1] == 'f') search_reverse = false;
else if(argv[i][1] == 'r') search_forward = false;
}
}
seq1_size_ = get_seq_length(seq1_file) + 1;
seq2_size_ = get_seq_length(seq2_file) + 1;
seq1_ = read_fasta_and_create_int8_t_array(seq1_file, seq1_size_);
seq2_ = read_fasta_and_create_int8_t_array(seq2_file, seq2_size_);
int* SA = create_SA(seq1_file, seq1_size_);
BWT bwt(seq1_, SA, seq1_size_, num_char_, bwt_interval);
if(search_forward) search_forward_matches(seq1_name, seq2_name, SA, bwt, kmer_size, slide_letters, max_num_matches);
if(search_reverse) search_reverse_matches(seq1_name, seq2_name, SA, bwt, kmer_size, slide_letters, max_num_matches);
delete SA;
}