int main(int argc, char* argv[])
{
char* myString;
int* suffixArray;
int stringLength;
int i;
ifstream inFile;
inFile.open(argv[1]);
Timing timehere;
if (strcmp(argv[1], "test.dat") != 0)
{
timehere.markbeg();
if (strstr(argv[1], ".fas")[0] == '.')
{
read_fasta(inFile, myString, stringLength);
}
else
{
read_input(inFile, myString, stringLength);
}
timehere.markend();
inFile.close();
cout << "finish read "
<< stringLength << " characters."<< endl;
timehere.outtime();
}
else
{
read_input(inFile, myString, stringLength);
inFile.close();
cout << "finish read "
<< stringLength << " characters."<< endl;
}
timehere.markbeg();
suffixArray = LinearSuffixSort(myString, stringLength);
timehere.markend();
timehere.outtime("finish suffix sort,");
if (strcmp(argv[1], "test.dat") == 0)
{
int result;
bool pass = true;
ifstream resultF;
resultF.open("result.test.dat");
cout << "Testing the Suffix Array" << endl;
for (i = 0; i < stringLength; i++)
{
resultF >> result;
if (result != suffixArray[i])
{
pass = false;
}
}
if (pass == false)
{
cout << endl;
cout << "***************" << endl;
cout << "test has failed" << endl;
cout << "***************" << endl;
}
else
{
cout << endl;
cout << "******************" << endl;
cout << "test is successful" << endl;
cout << "******************" << endl;
}
}
int main (int argc, char *argv[])
{
char *myString;
int stringLength;
int i;
string namePattern = "";
char *version = "1.06";
bool display_usage = false;
char c;
CPSSCAN ps_scan;
PBS pbs;
PPT ppt(PPT_window, 4);
char *tRNA_file = NULL;
char *ps_dir = NULL;
bool ishtml = false;
struct re_pattern_buffer id_filter;
id_filter.allocated = 0;
id_filter.buffer = 0;
id_filter.fastmap = 0;
id_filter.translate = 0;
Timing timehere;
Timing timeall;
timeall.markbeg();
while ((c = getopt (argc, argv, "ho:t:e:m:u:D:d:L:l:p:s:cw:S:a:P:g:F:B:b:J:j:O:r:M:f:xEiCG:T:")) > 0)
{
switch (c)
{
case 'h':
display_usage = true;
break;
case 'o':
gap_open = atoi (optarg);
break;
case 't':
gap_ext = atoi (optarg);
break;
case 'e':
gap_end = atoi (optarg);
break;
case 'm':
score_match = atoi (optarg);
break;
case 'u':
score_mismatch = atoi (optarg);
break;
case 'D':
Dmax = atoi (optarg);
break;
case 'd':
Dmin = atoi (optarg);
break;
case 'L':
Lmax = atoi (optarg);
break;
case 'l':
Lmin = atoi (optarg);
break;
case 'p':
Lex = atoi (optarg);
break;
case 'c':
CHECK_PAIRS = 1;
break;
case 'x':
ishtml = true;
break;
case 'E':
edge_signal = true;
break;
case 'i':
showPairNum = true;
break;
case 'w':
wrought = atoi(optarg);
break;
case 'P':
namePattern = optarg;
break;
case 'F':
Filter = optarg;
break;
case 'g':
MaxGap = atoi(optarg);
break;
case 'O':
outAlignLen = atoi(optarg);
break;
case 'G':
max_sub_rt_gap = atoi(optarg);
break;
case 'T':
min_sub_rt_count = atoi(optarg);
break;
case 'r':
PBS_minLen = atoi(optarg);
break;
case 'j':
JoinThreshold = atof(optarg);
break;
case 'J':
SplitThreshold = atof(optarg);
break;
case 'S':
minOutScore = atof(optarg);
break;
case 'B':
HigherSharpness = atof(optarg);
break;
case 'b':
LowerSharpness = atof(optarg);
break;
case 'M':
minMatchSim = atof(optarg);
break;
case 's':
tRNA_file = optarg;
break;
case 'a':
ps_dir = optarg;
break;
case 'f':
fig_file = optarg;
break;
case 'C':
checkCentriole = true;
break;
default:
display_usage = true;
// case 'm': show_score_matrix(); break;
}
}
if (optind >= argc || display_usage)
{
fprintf (stderr, "ltr_finder v%s\n", version);
fprintf (stderr, "Usage : [options] <INPUT_FASTA_FILE>\n");
// fprintf (stderr, "Options: -b NUM bandwidth, default is %d\n",
// band_width);
fprintf (stderr,
" -o NUM gap open penalty, default is %d\n",
gap_open);
fprintf (stderr,
" -t NUM gap extension penalty, default is %d\n",
gap_ext);
fprintf (stderr,
" -e NUM gap end penalty, default is %d\n",
gap_end);
fprintf (stderr, " -m NUM match score, default is %d\n",
score_match);
fprintf (stderr, " -u NUM unmatch score, default is %d\n",
score_mismatch);
fprintf (stderr,
" -D NUM Max distance between 5'&3'LTR, default is %d\n",
Dmax);
fprintf (stderr,
" -d NUM Min distance between 5'&3'LTR, default is %d\n",
Dmin);
fprintf (stderr,
" -L NUM Max length of 5'&3'LTR, default is %d\n",
Lmax);
fprintf (stderr,
" -l NUM Min length of 5'&3'LTR, default is %d\n",
Lmin);
fprintf (stderr,
" -p NUM min length of exact match pair, default is %d\n",
Lex);
fprintf (stderr,
" -g NUM Max gap between joined pairs, default is %d\n",
MaxGap);
fprintf (stderr,
" -G NUM Max gap between RT sub-domains, default is %d\n",
max_sub_rt_gap);
fprintf (stderr,
" -T NUM Min sub-domains found in a RT domain, default is %d\n",
min_sub_rt_count);
fprintf (stderr, " -j NUM Threshold for join new sequence in existed alignment\n");
fprintf (stderr, " new alignment similarity higher than this will be joined,\n");
fprintf (stderr, " default is %0.2f\n",
JoinThreshold);
fprintf (stderr, " -J NUM Threshold for split existed alignment to two part\n");
fprintf (stderr, " new alignment similarity lower than this will be split,\n");
fprintf (stderr, " set this threshold lower than -j, means turn it off,\n");
fprintf (stderr, " default is %0.2f\n",
SplitThreshold);
fprintf (stderr,
" -S NUM output Score limit, default is %0.2f, [0,10]\n",
minOutScore);
fprintf (stderr,
" -M NUM min LTR similarity threshold, default is %0.2f, [0,1]\n",
minMatchSim);
fprintf (stderr, " -B NUM Boundary alignment sharpness threshold, higher one.\n");
fprintf (stderr, " one of the two edge's sharpness must higher than\n");
fprintf (stderr, " this threshold, default is %0.3f, [0,1]\n",
HigherSharpness);
fprintf (stderr, " -b NUM Boundary alignment sharpness threshold, lower one.\n");
fprintf (stderr, " both of the two edge's sharpness must higher than\n");
fprintf (stderr, " this threshold, default is %0.3f, [0,1]\n",
LowerSharpness);
fprintf (stderr,
" -r NUM PBS detecting threshold, min tRNA match length: %d, [1,18]\n",
PBS_minLen);
fprintf (stderr,
" -w NUM output format: [0]-full, 1-summary, 2-table.\n");
fprintf (stderr,
" -O NUM output alignment length(only affect -w0), default is %d\n",
outAlignLen);
fprintf (stderr, " -P STR SeqIDs, will only calculate matched SeqID\n");
fprintf (stderr, " POSIX style regular express is supported.\n");
fprintf (stderr,
" -s filename tRNA sequence file(FASTA format)\n");
fprintf (stderr,
" -f filename data file used to draw figure\n");
fprintf (stderr,
" -a ps_scan_dir Use ps_scan to predict protein domain\n");
fprintf (stderr, " -x Output in html format\n");
fprintf (stderr, " -E LTR must have edge signal\n");
fprintf (stderr, " (at least two of PBS,PPT,TSR)\n");
fprintf (stderr, " -C detect Centriole, delete highly repeat regions\n");
fprintf (stderr,
" -F 01string Filter to choose desired result,default is 0\n");
fprintf (stderr, " 10000000000 5'-LTR must have TG\n");
fprintf (stderr, " 01000000000 5'-LTR must have CA\n");
fprintf (stderr, " 00100000000 3'-LTR must have TG\n");
fprintf (stderr, " 00010000000 3'-LTR must have CA\n");
fprintf (stderr, " 00001000000 TSR must be found\n");
fprintf (stderr, " 00000100000 PBS must be found\n");
fprintf (stderr, " 00000010000 PPT must be found\n");
fprintf (stderr, " 00000001000 RT domain muse be found\n");
fprintf (stderr, " 00000000100 Integrase core must be found\n");
fprintf (stderr, " 00000000010 Integrase c-term must be found\n");
fprintf (stderr, " 00000000001 RNase H must be found\n");
// fprintf(stderr, " -m show score matrix\n");
fprintf (stderr, " -h help\n");
exit (1);
}
// if(argc<2)
// {
// cerr<<"ltr_finder INPUT_FASTA_FILE"<<endl;
// exit(1);
// }
//Program : BGF
//Version : 2.1.2
//Time : Wed Nov 22 20:53:23 2006
if (ishtml)
printf("<html>\n<head>\n<title>LTR_FINDER Result</title>\n</head>\n<body>\n<pre>\n");
printf("Program : LTR_FINDER\n");
printf("Version : %s\n\n", version);
const char *id_filter_stat;
re_syntax_options = RE_SYNTAX_POSIX_EGREP |
RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL;
id_filter_stat = re_compile_pattern(namePattern.c_str(),
namePattern.length(), &id_filter);
if (id_filter_stat != NULL)
{
printf("not a vaild POSIX regex after -P, code = %s\n", id_filter_stat);
}
if (tRNA_file != NULL)
{
timehere.markbeg();
string tmp_out = "Load tRNA db [";
tmp_out += tRNA_file;
tmp_out += "] ";
pbs.LoadSeq(tRNA_file);
timehere.markend ();
timehere.outtime(tmp_out.c_str());
}
ps_scan.init(ps_dir);
set_score_matrix(gap_open, gap_ext, gap_end, score_match, score_mismatch);
FILE * inFASTA = fopen(argv[optind], "r");
if (inFASTA == NULL)
{
cerr << "open " << argv[optind] << " error!" << endl;
exit(1);
}
//init transDNA
for(int i=0;i<128;++i)
transDNA[i]='N';
transDNA['a']='A';
transDNA['A']='A';
transDNA['c']='C';
transDNA['C']='C';
transDNA['g']='G';
transDNA['G']='G';
transDNA['t']='T';
transDNA['T']='T';
transDNA['u']='T';
transDNA['U']='T';
seq_t sequence;
sequence.s = NULL;
sequence.m = 0;
char name[1024];
ofstream fig;
if (fig_file != NULL)
fig.open(fig_file);
int total_sequence = 0;
int total_img = 0;
while ( -1 != read_fasta(inFASTA, &sequence, name, 0))
{
total_sequence++;
if (namePattern.length() > 0 && id_filter_stat == NULL &&
re_search( &id_filter, name, strlen(name),
0, strlen(name), 0) < 0 ) //name Pattern not found
//namePattern.find(name)==string::npos)
continue;
for (int i = 0;i < sequence.l;++i)
sequence.s[i] = transDNA[sequence.s[i]];
char* myString = (char*)sequence.s;
int stringLength = sequence.l;
if (showPairNum)
cout << "Sequence:" << name << " Len:" << stringLength << " ";
//cout<<"begin GetPairs\n";
vector < candidate > Pair;
GetPairs (myString, stringLength, Lex, Lmax, Dmin, Dmax, Pair);
if (showPairNum)
continue;
//output pairs for checking
for (int i = 0; i < Pair.size () && CHECK_PAIRS; ++i)
{
cout << setw (10);
cout << Pair[i].pos1 << "-" << Pair[i].pos1 + Pair[i].len - 1 << " * "
<< Pair[i].pos2 << "-" << Pair[i].pos2 + Pair[i].len - 1 << endl;
}
//constrcut sticks
//cout<<used[2]<<endl;
vector < stick > sticks;
JoinPairs(myString, stringLength, Pair, sticks);
//cerr<<"pairs after join:"<<sticks.size()<<endl;
int count_do = 0;
ps_scan.reset();
for (int i = 0; i < sticks.size (); ++i)
{
sticks[i].score = 0;
//cout<<"begin extend pairs\n";
if (!ExtendPairs(myString, stringLength, sticks[i]))
continue;
count_do++;
//cout<<"begin findsignal\n";
FindSignal(myString, stringLength, pbs, ppt, sticks[i]);
ps_scan.AddRegion(sticks[i].end1 + 1, sticks[i].pos2 - 1);
}
timehere.markbeg();
ps_scan.Predict(myString, stringLength);
timehere.markend ();
timehere.outtime ("Predict protein Domains");
for (int i = 0;i < sticks.size (); ++i)
{
if (sticks[i].score < 0)
continue;
//cout<<"begin finddomain\n";
ps_scan.Find(sticks[i].end1 + 1, sticks[i].pos2 - 1, sticks[i].motif);
//cout<<"count score\n";
CountScore(sticks[i]);
//cerr<<"len:"<<sticks[i].match_len<<" score:"<<sticks[i].match_score<<endl;
sticks[i].match_score = sticks[i].match_score / sticks[i].match_len; //score/average_len
if (sticks[i].match_score > 1)
{
//cerr << "similarity bigger than1:" << sticks[i].match_score << endl;
sticks[i].match_score=1;//why???
}
}
//cerr<<"after minSharpness filter:"<<count_do<<endl;
//may be this function is needn't
//EraseOverlap(sticks);
stable_sort (sticks.begin (), sticks.end ());
//cout<<"out result\n";
int result_count = OutPutResult(name, myString, sequence.l, sticks, fig);
if (CHECK_PAIRS)
ps_scan.PrintNoUsed();
cout << endl;
if (result_count && ishtml && fig.good())
{
total_img++;
cout << "</pre>\n";
cout << "<img src=\"" << total_img << ".png\">\n";
cout << "<pre>\n";
}
}
if (total_sequence == 0)
{
cout << "No sequence found, please input FASTA format sequence and try again\n";
}
fig.close();
fclose(inFASTA);
free (sequence.s);
timeall.markend ();
timeall.outtime("Total consume");
if (ishtml)
cout << "</pre>\n</body>\n</html>\n";
return (0);
}
