void SequenceCleaner::read_sequences (istream* pios) {
Sequence seq;
string retstring;
int ft = test_seq_filetype_stream(*pios, retstring);
int num_current_char = 0;
bool first = true;
while (read_next_seq_from_stream(*pios, ft, retstring, seq)) {
sequences_[seq.get_id()] = seq.get_sequence();
num_current_char = seq.get_sequence().size();
if (first) {
num_char_ = num_current_char; // just getting this from an arbitrary (first) sequence for now
if (is_dna_) {
string alpha_name = seq.get_alpha_name();
if (alpha_name == "AA") {
is_dna_ = false;
//cout << "I believe this is a protein!" << endl;
}
}
first = false;
continue;
} else {
if (num_current_char != num_char_) {
cout << "Error: sequences are not all of the same length. Exiting."
<< endl;
exit(0);
}
}
}
if (ft == 2) {
sequences_[seq.get_id()] = seq.get_sequence();
num_current_char = seq.get_sequence().size();
if (num_current_char != num_char_) {
cout << "Error: sequences are not all of the same length. Exiting."
<< endl;
exit(0);
}
}
num_taxa_ = sequences_.size();
}
NJOI::NJOI (istream* pios, int & threads):ntax_(0), nchar_(0), nthreads_(threads) {
Sequence seq;
string retstring;
int ft = test_seq_filetype_stream(*pios, retstring);
int seqcount = 0;
// some error checking. should be in general seq reader class
bool first = true;
while (read_next_seq_from_stream(*pios, ft, retstring, seq)) {
sequences_[seq.get_id()] = seq.get_sequence();
if (!first) {
if ((int)seq.get_length() != nchar_) {
cout << "Error: sequence " << seq.get_id() << " has "
<< seq.get_length() << " characters, was expecting "
<< nchar_ << "." << endl << "Exiting." << endl;
exit(1);
}
} else {
nchar_ = seq.get_length();
first = false;
}
seqcount++;
}
//fasta has a trailing one
if (ft == 2) {
sequences_[seq.get_id()] = seq.get_sequence();
if ((int)seq.get_length() != nchar_) {
cout << "Error: sequence " << seq.get_id() << " has "
<< seq.get_length() << " characters, was expecting "
<< nchar_ << "." << endl << "Exiting." << endl;
exit(1);
};
seqcount++;
}
ntax_ = seqcount;
set_name_key ();
Matrix = BuildMatrix(sequences_);
TREEMAKE(names_, name_key_, Matrix);
}
int main(int argc, char * argv[]) {
log_call(argc, argv);
bool cfileset = false;
bool tfileset = false;
bool outfileset = false;
char * treef = NULL;
char * charf = NULL;
char * outf = NULL;
int analysis = 0;
while (1) {
int oi = -1;
int c = getopt_long(argc, argv, "c:t:o:hV", long_options, &oi);
if (c == -1) {
break;
}
switch(c) {
case 'c':
cfileset = true;
charf = strdup(optarg);
check_file_exists(charf);
break;
case 't':
tfileset = true;
treef = strdup(optarg);
check_file_exists(treef);
break;
case 'o':
outfileset = true;
outf = strdup(optarg);
break;
case 'h':
print_help();
exit(0);
case 'V':
cout << versionline << endl;
exit(0);
default:
print_error(argv[0], (char)c);
exit(0);
}
}
istream * pios = NULL;
istream * poos = NULL;
ifstream * cfstr = NULL;
ifstream * tfstr = NULL;
ostream * poouts = NULL;
ofstream * ofstr = NULL;
if (tfileset == true) {
tfstr = new ifstream(treef);
poos = tfstr;
} else {
poos = &cin;
}
if (cfileset == true) {
cfstr = new ifstream(charf);
pios = cfstr;
} else {
cout << "you have to set a character file. Only a tree file can be read in through the stream;" << endl;
}
//out file
//
if (outfileset == true){
ofstr = new ofstream(outf);
poouts = ofstr;
} else{
poouts = &cout;
}
//
string retstring;
int ft = test_char_filetype_stream(*pios, retstring);
if (ft != 1 && ft != 2) {
cout << "only fasta and phylip (with spaces) supported so far" << endl;
exit(0);
}
Sequence seq;
vector <Sequence> seqs;
map <string, int> seq_map;
int y = 0;
int nchars = 0 ;
while (read_next_seq_char_from_stream(*pios, ft, retstring, seq)) {
seqs.push_back(seq);
nchars = seq.get_num_cont_char();
seq_map[seq.get_id()] = y;
seq.clear_cont_char();
y++;
}
cout << "nchars: " << nchars << endl;
if (ft == 2) {
seqs.push_back(seq);
seq_map[seq.get_id()] = y;
seq.clear_cont_char();
}
//read trees
TreeReader tr;
vector<Tree *> trees;
while (getline(*poos,retstring)) {
if (retstring.size()<4){
continue;
}
trees.push_back(tr.readTree(retstring));
}
int x = 0;
//conduct analyses for each character
for (int i=0; i < trees[x]->getExternalNodeCount(); i++) {
vector<Superdouble> tv (nchars);
for (int c=0; c < nchars; c++) {
tv[c] = seqs[seq_map[trees[x]->getExternalNode(i)->getName()]].get_cont_char(c);
}
trees[x]->getExternalNode(i)->assocDoubleVector("val",tv);
}
for (int i=0; i < trees[x]->getInternalNodeCount(); i++) {
vector<Superdouble> tv (nchars);
for (int c=0; c < nchars; c++) {
tv[c] = 0;
}
trees[x]->getInternalNode(i)->assocDoubleVector("val",tv);
}
float sigma = 1;
cout << calc_bm_prune(trees[x], sigma) << endl;
optimize_single_rate_bm_bl(trees[x]);
(*poouts) << trees[x]->getRoot()->getNewick(true) << ";" << endl;
cout << calc_bm_prune(trees[x], sigma) << endl;
if (cfileset) {
cfstr->close();
delete pios;
}
if (tfileset) {
tfstr->close();
delete poos;
}
if (outfileset) {
ofstr->close();
delete poouts;
}
return EXIT_SUCCESS;
}
void SequenceConcatenater::read_sequences (string & seqf) {
filename_ = seqf;
string retstring;
istream * pios = new ifstream(filename_);
ft_ = test_seq_filetype_stream(*pios, retstring);
Sequence seq;
int counter = 0;
int length = 0;
// phylip (1) NEXUS (0)
if (ft_ == 1 || ft_ == 0) {
if (ft_ == 1) {
vector <string> fileDim = tokenize(retstring);
num_taxa_ = stoi(fileDim[0]);
num_char_ = stoi(fileDim[1]);
} else {
get_nexus_dimensions_file(seqf, num_taxa_, num_char_, interleave_);
}
if (!interleave_) {
while (read_next_seq_from_stream(*pios, ft_, retstring, seq)) {
length = (int)seq.get_sequence().size();
if (length != num_char_) {
cout << "Sequence '" << seq.get_id() << "' has " << length << " characters, but the file '"
<< filename_ << "' specified " << num_char_ << " characters. Exiting." << endl;
delete pios;
exit(1);
}
if (toupcase_) {
seq.set_sequence(seq.seq_to_upper());
}
seqs_.push_back(seq);
counter++;
}
if (counter != num_taxa_) {
cout << "Read " << counter << " taxa, but the file '" << filename_ << "' specified "
<< num_taxa_ << " taxa. Exiting." << endl;
delete pios;
exit(1);
}
} else {
seqs_ = read_interleaved_nexus_file(seqf, num_taxa_, num_char_);
if (toupcase_) {
for (int i = 0; i < num_taxa_; i++) {
seqs_[i].set_sequence(seqs_[i].seq_to_upper());
}
}
}
} else if (ft_ == 2) { // fasta
bool first = true;
while (read_next_seq_from_stream(*pios, ft_, retstring, seq)) {
int curr = (int)seq.get_sequence().size();
if (!first) {
if (curr != length) {
cout << "Error: current sequence has " << curr << " characters, but previous sequence had "
<< length << " characters. Exiting." << endl;
delete pios;
exit(1);
}
} else {
length = curr;
first = false;
}
if (toupcase_) {
seq.set_sequence(seq.seq_to_upper());
}
seqs_.push_back(seq);
counter++;
}
// fasta has a trailing one
if (toupcase_) {
seq.set_sequence(seq.seq_to_upper());
}
seqs_.push_back(seq);
counter++;
num_taxa_ = counter;
num_char_ = length;
} else {
cout << "I don't know what that alignment file format is! Exiting." << endl;
exit(0);
}
num_partitions_ = 1;
partition_sizes_.push_back(num_char_);
delete pios;
}
int main(int argc, char * argv[]) {
log_call(argc, argv);
bool outfileset = false;
bool sfileset = false;
bool cfileset = false;
bool nfileset = false;
bool verbose = false;
char * outf = NULL;
char * seqf = NULL;
string cnamef = "";
string nnamef = "";
while (1) {
int oi = -1;
int c = getopt_long(argc, argv, "s:c:n:o:vhV", long_options, &oi);
if (c == -1) {
break;
}
switch(c) {
case 's':
sfileset = true;
seqf = strdup(optarg);
check_file_exists(seqf);
break;
case 'c':
cfileset = true;
cnamef = strdup(optarg);
check_file_exists(cnamef.c_str());
break;
case 'n':
nfileset = true;
nnamef = strdup(optarg);
check_file_exists(nnamef.c_str());
break;
case 'o':
outfileset = true;
outf = strdup(optarg);
break;
case 'v':
verbose = true;
break;
case 'h':
print_help();
exit(0);
case 'V':
cout << versionline << endl;
exit(0);
default:
print_error(argv[0], (char)c);
exit(0);
}
}
if (sfileset && outfileset) {
check_inout_streams_identical(seqf, outf);
}
istream * pios = NULL;
ostream * poos = NULL;
ifstream * fstr = NULL;
ofstream * ofstr = NULL;
if (!nfileset | !cfileset) {
cout << "Must supply both name files (-c for current, -n for new)." << endl;
exit(0);
}
if (sfileset == true) {
fstr = new ifstream(seqf);
pios = fstr;
} else {
pios = &cin;
if (check_for_input_to_stream() == false) {
print_help();
exit(1);
}
}
if (outfileset == true) {
ofstr = new ofstream(outf);
poos = ofstr;
} else {
poos = &cout;
}
Relabel rl (cnamef, nnamef, verbose);
set <string> orig = rl.get_names_to_replace();
Sequence seq;
string retstring;
int ft = test_seq_filetype_stream(*pios, retstring);
bool success = false;
while (read_next_seq_from_stream(*pios, ft, retstring, seq)) {
string terp = seq.get_id();
success = rl.relabel_sequence(seq);
if (success) {
orig.erase(terp);
}
(*poos) << ">" << seq.get_id() << endl;
(*poos) << seq.get_sequence() << endl;
}
// have to deal with last sequence outside while loop. fix this.
if (ft == 2) {
string terp = seq.get_id();
success = rl.relabel_sequence(seq);
if (success) {
orig.erase(terp);
}
(*poos) << ">" << seq.get_id() << endl;
(*poos) << seq.get_sequence() << endl;
}
if (orig.size() > 0) {
if (verbose) {
cerr << "The following names to match were not found in the alignment:" << endl;
for (auto elem : orig) {
cerr << elem << endl;
}
}
}
if (sfileset) {
fstr->close();
delete pios;
}
if (outfileset) {
ofstr->close();
delete poos;
}
return EXIT_SUCCESS;
}
int main(int argc, char * argv[]) {
log_call(argc, argv);
bool outfileset = false;
bool fileset = false;
bool printpost = false;
bool showancs = false;
bool is_dna = true;
float pinvar = 0.0;
double tot;
string yorn = "n";
int seqlen = 1000;
int pos = 0;
int pos2 = 0;
string infreqs;
string inrates;
string holdrates;
string ancseq;
char * outf = NULL;
char * treef = NULL;
vector <double> diag(20, 0.0);
vector <double> basefreq(4, 0.25);
vector <double> aabasefreq(20, 0.05);
vector <double> userrates;
vector <double> multirates;
int nreps = 1; // not implemented at the moment
int seed = -1;
int numpars = 0;
float alpha = -1.0;
vector<vector <double>> dmatrix;
vector< vector <double> > aa_rmatrix(20, vector<double>(20, 1));
for (unsigned int i = 0; i < aa_rmatrix.size(); i++) {
for (unsigned int j = 0; j < aa_rmatrix.size(); j++) {
if (i == j) { // Fill Diagonal
aa_rmatrix[i][j] = -19.0;
}
}
}
vector< vector <double> > rmatrix(4, vector<double>(4, 0.33));
for (unsigned int i = 0; i < rmatrix.size(); i++) {
for (unsigned int j = 0; j < rmatrix.size(); j++) {
if (i == j) { // Fill Diagonal
rmatrix[i][j] = -0.99;
}
}
}
/*dmatrix = aa_rmatrix;
for (unsigned int i = 0; i < dmatrix.size(); i++) {
for (unsigned int j = 0; j < dmatrix.size(); j++) {
cout << dmatrix[i][j] << " ";
}
cout << "\n";
}*/
while (1) {
int oi = -1;
int c = getopt_long(argc, argv, "t:o:l:b:g:i:r:w:q:n:x:apcm:k:hV", long_options, &oi);
if (c == -1) {
break;
}
switch(c) {
case 't':
fileset = true;
treef = strdup(optarg);
check_file_exists(treef);
break;
case 'o':
outfileset = true;
outf = strdup(optarg);
break;
case 'b':
infreqs = strdup(optarg);
parse_comma_list(infreqs, basefreq);
if (basefreq.size() != 4) {
cout << "Error: must provide 4 base frequencies (" << basefreq.size()
<< " provided). Exiting." << endl;
exit(0);
}
if (!essentially_equal(sum(basefreq), 1.0)) {
cout << "Error: base frequencies must sum to 1.0. Exiting." << endl;
exit(0);
}
break;
case 'l':
seqlen = atoi(strdup(optarg));
break;
case 'a':
showancs = true;
break;
case 'r':
inrates = strdup(optarg);
parse_comma_list(inrates, userrates);
// NOTE: will have to alter this check for a.a., non-reversible, etc.
if (userrates.size() != 6) {
cout << "Error: must provide 6 substitution parameters. " <<
"Only " << userrates.size() << " provided. Exiting." << endl;
exit(0);
}
// NOTE: this uses order: A,T,C,G for matrix, but
// A<->C,A<->G,A<->T,C<->G,C<->T,G<->T for subst. params.
rmatrix[0][2] = userrates[0];
rmatrix[2][0] = userrates[0];
rmatrix[0][3] = userrates[1];
rmatrix[3][0] = userrates[1];
rmatrix[0][1] = userrates[2];
rmatrix[1][0] = userrates[2];
rmatrix[2][3] = userrates[3];
rmatrix[3][2] = userrates[3];
rmatrix[1][2] = userrates[4];
rmatrix[2][1] = userrates[4];
rmatrix[1][3] = userrates[5];
rmatrix[3][1] = userrates[5];
rmatrix[0][0] = (userrates[0]+userrates[1]+userrates[2]) * -1;
rmatrix[1][1] = (userrates[2]+userrates[4]+userrates[5]) * -1;
rmatrix[2][2] = (userrates[0]+userrates[3]+userrates[4]) * -1;
rmatrix[3][3] = (userrates[1]+userrates[3]+userrates[5]) * -1;
/*//Turn on to check matrix
for (unsigned int i = 0; i < rmatrix.size(); i++) {
for (unsigned int j = 0; j < rmatrix.size(); j++) {
cout << rmatrix[i][j] << " ";
}
cout << "\n";
}*/
break;
case 'w':
inrates = strdup(optarg);
parse_comma_list(inrates, userrates);
is_dna = false;
// NOTE: will have to alter this check for a.a., non-reversible, etc.
if (userrates.size() != 190) {
cout << "Error: must provide 190 substitution parameters, I know its a stupidly large amount. " <<
"Only " << userrates.size() << " provided. Exiting." << endl;
exit(0);
}
pos = 0;
pos2 = 1;
//Fill the Matrix
for (unsigned int i = 0; i < userrates.size(); i++){
aa_rmatrix[pos][pos2] = userrates[i];
aa_rmatrix[pos2][pos] = userrates[i];
pos2++;
if (pos2 == 20){
pos += 1;
pos2 = (pos + 1);
}
}
//Replace Diagonal
for (unsigned int i = 0; i < aa_rmatrix.size(); i++) {
for (unsigned int j = 0; j < aa_rmatrix.size(); j++) {
if (i != j){
tot += aa_rmatrix[i][j];
}
}
aa_rmatrix[i][i] = (tot*-1);
tot = 0.0;
}
/*
for (unsigned int i = 0; i < aa_rmatrix.size(); i++) {
for (unsigned int j = 0; j < aa_rmatrix.size(); j++) {
cout << aa_rmatrix[i][j] << " ";
}
cout << "\n";
}*/
break;
case 'n':
nreps = atoi(strdup(optarg));
break;
case 'x':
seed = atoi(strdup(optarg));
break;
case 'q':
is_dna = false;
infreqs = strdup(optarg);
parse_comma_list(infreqs, aabasefreq);
if (aabasefreq.size() != 20) {
cout << "Error: must provide 20 base frequencies (" << aabasefreq.size()
<< " provided). Exiting." << endl;
exit(0);
}
if (!essentially_equal(sum(aabasefreq), 1.0)) {
cout << "Error: base frequencies must sum to 1.0. Exiting." << endl;
exit(0);
}
break;
case 'g':
alpha = atof(strdup(optarg));
break;
case 'i':
pinvar = atof(strdup(optarg));
break;
case 'p':
printpost = true;
break;
case 'c':
is_dna = false;
break;
case 'm':
holdrates = strdup(optarg);
parse_comma_list(holdrates, multirates);
numpars = multirates.size();
if ((numpars - 6) % 7 != 0) {
cout << "Error: must provide 6 background substitution "
<< "parameters and 7 values (1 node id + 6 subst. par.) "
<< "for each piecewise model. Exiting." << endl;
exit(0);
}
break;
case 'k':
ancseq = strdup(optarg);
break;
case 'h':
print_help();
exit(0);
case 'V':
cout << versionline << endl;
exit(0);
default:
print_error(argv[0], (char)c);
exit(0);
}
}
if (fileset && outfileset) {
check_inout_streams_identical(treef, outf);
}
if (is_dna) {
dmatrix = rmatrix;
} else {
dmatrix = aa_rmatrix;
}
istream * pios = NULL;
ostream * poos = NULL;
ifstream * fstr = NULL;
ofstream * ofstr = NULL;
if (outfileset == true) {
ofstr = new ofstream(outf);
poos = ofstr;
} else {
poos = &cout;
}
if (fileset == true) {
fstr = new ifstream(treef);
pios = fstr;
} else {
pios = &cin;
if (check_for_input_to_stream() == false) {
print_help();
exit(1);
}
}
/*
* Default Base Frequencies and Rate Matrix
*
*/
//vector <double> basefreq(4, 0.0);
//basefreq[0] = .25;
//basefreq[1] = .25;
//basefreq[2] = .25;
//basefreq[3] = 1.0 - basefreq[0] - basefreq[1] - basefreq[2];
/*
vector< vector <double> > rmatrix(4, vector<double>(4, 0.33));
for (unsigned int i = 0; i < rmatrix.size(); i++) {
for (unsigned int j = 0; j < rmatrix.size(); j++) {
if (i == j) {//Fill Diagnol
rmatrix[i][j] = -1.0;
}
}
}
*/
string retstring;
int ft = test_tree_filetype_stream(*pios, retstring);
if (ft != 0 && ft != 1) {
cerr << "this really only works with nexus or newick" << endl;
exit(0);
}
// allow > 1 tree in input. passing but not yet using nreps
int treeCounter = 0;
bool going = true;
if (ft == 1) { // newick. easy
Tree * tree;
while (going) {
tree = read_next_tree_from_stream_newick (*pios, retstring, &going);
if (tree != NULL) {
//cout << "Working on tree #" << treeCounter << endl;
SequenceGenerator SGen(seqlen, basefreq, dmatrix, tree, showancs,
nreps, seed, alpha, pinvar, ancseq, printpost, multirates, aabasefreq, is_dna);
vector <Sequence> seqs = SGen.get_sequences();
for (unsigned int i = 0; i < seqs.size(); i++) {
Sequence seq = seqs[i];
(*poos) << ">" << seq.get_id() << endl;
//cout << "Here" << endl;
(*poos) << seq.get_sequence() << endl;
}
delete tree;
treeCounter++;
}
}
} else if (ft == 0) { // Nexus. need to worry about possible translation tables
map <string, string> translation_table;
bool ttexists;
ttexists = get_nexus_translation_table(*pios, &translation_table, &retstring);
Tree * tree;
while (going) {
tree = read_next_tree_from_stream_nexus(*pios, retstring, ttexists,
&translation_table, &going);
if (going == true) {
cout << "Working on tree #" << treeCounter << endl;
SequenceGenerator SGen(seqlen, basefreq, dmatrix, tree, showancs,
nreps, seed, alpha, pinvar, ancseq, printpost, multirates, aabasefreq, is_dna);
vector <Sequence> seqs = SGen.get_sequences();
for (unsigned int i = 0; i < seqs.size(); i++) {
Sequence seq = seqs[i];
(*poos) << ">" << seq.get_id() << endl;
(*poos) << seq.get_sequence() << endl;
}
delete tree;
treeCounter++;
}
}
}
return EXIT_SUCCESS;
}
