New upstream version 1.0.1 (56f3e8c3) · Commits · Debian Med / e-mem

Makefile

+2 −2

Original line number	Diff line number	Diff line
		@@ -5,8 +5,8 @@ endif
		CC = g++
		EXEC = e-mem
		CFLAGS = -Wall -Wextra -Wunused -mpopcnt -std=gnu++0x -fopenmp
		CDEBUG = -g -ggdb -DDEBUG
		CPROF = -g -ggdb -DDEBUG -pg
		CDEBUG = -g -ggdb -gdwarf-3 -DDEBUG
		CPROF = -g -ggdb -gdwarf-3 -DDEBUG -pg
		COPTIMIZE = -Wuninitialized -O3 -fomit-frame-pointer
		CLIBS = -lm

README.md

+25 −3

Original line number	Diff line number	Diff line
		@@ -15,7 +15,29 @@ E-MEM is an efficient MEM computation program for large genomes which can be use

		OUTPUT:

		stdout a list of exact matches
		The E-MEM program prints maximal exact matches on standard output. The output format varies depending on the command line options used. In the default mode 3-column output is printed. The option -b allows printing of matches in both forward and reverse complement directions. Below is the description of the output format with respect to the example included with e-mem program. For each query sequence, the sequence name or ID is reported on the first line followed by a '>' character. The sequence name will be reported even if there are no MEMs found for this sequence. For example, the query sequence in 'Reverse' reports sequence name with no matching MEMs. Note that, for each query sequence, the reverse complemented MEMs immediately follow the forward MEMs. For each match, the 3-columns list the start position in the reference sequence, the start position in the query sequence, and the length of the match respectively. For reverse matches, the positions are reported relative to the reverse query sequence. The option -c is used to report reverse match positions with respect to forward query sequence.

		> gi\|5524211\|gb\|AAD44166.1\|
		4 1 51
		8 1 51
		12 1 51
		16 1 51
		20 1 51
		24 1 51
		28 1 51
		1 2 50
		> gi\|5524211\|gb\|AAD44166.1\| Reverse

		The 3-column output is sufficient for reporting matches between one reference sequence and one/more query sequences. For multiple reference sequences, 4-column output is desired. E-MEM provides an option -F to print 4-column output which also prints the reference sequence name/ID for each match. Below is a dummy example of 4-column output. The first column now indicates the name/ID of the reference sequence followed by the 3-column output format as described above.

		> gi\|5524211\|gb\|AAD44166.1\|
		ref1 4 1 51
		ref1 8 1 51
		ref2 12 1 51
		ref2 16 1 51
		ref3 20 1 51
		ref3 24 1 51


		##-- INSTALLATION --

		@@ -69,7 +91,7 @@ For example, in order to use NUCMER (all-vs-all comparison of nucleotide sequenc

		The other important script in MUmmer3 is run-mummer3 (the alignment program). To use this script with E-MEM, simply replace "$bindir/mummer" with "< path >/e-mem" where < path > is e-mem installation directory.

		###CITE
		## CITE
		If you find E-MEM program useful, please cite the E-MEM paper:

		N. Khiste, L. Ilie [E-MEM: efficient computation of maximal exact matches for very large genomes](http://bioinformatics.oxfordjournals.org/content/31/4/509.short) Bioinformatics, 2015

e-mem.cpp

+10 −1

Original line number	Diff line number	Diff line
		@@ -234,6 +234,14 @@ void helperReportMem(uint64_t &currRPos, uint64_t &currQPos, uint64_t totalRBits
		rQue-=2;
		break;
		}
		/* if current rRef/rQue plus matchSize smaller than minMEMLength, then simply return.
		* Note that one less character is compared due to a mismatch
		*/
		if (rRef+matchSize-lRef < static_cast<uint64_t>(commonData::minMemLen))
		{
		return;
		}

		mismatch=1;
		matchSize/=2;
		if (matchSize%2)
		@@ -488,7 +496,7 @@ void checkCommandLineOptions(uint32_t &options)
		void print_help_msg()
		{
		cout << endl;
		cout << "E-MEM Version 1.0.0, Sep. 25, 2014" << endl;
		cout << "E-MEM Version 1.0.1, Dec. 12, 2017" << endl;
		cout << "© 2014 Nilesh Khiste, Lucian Ilie" << endl;
		cout << endl;
		cout << "E-MEM finds and outputs the position and length of all maximal" << endl;
		@@ -739,6 +747,7 @@ int main (int argc, char *argv[])
		QueryFile.closeFile();

		arrayTmpFile.removeDuplicates(refSeqInfo, querySeqInfo, revComplement);
		fflush(0);
		return 0;
		}

file.h

+16 −16

Original line number	Diff line number	Diff line
		@@ -278,7 +278,7 @@ class seqFileReadInfo {
		{
		size = size/commonData::d;
		binReadSize = floor((size+numSequences*RANDOM_SEQ_SIZE+commonData::d)/32+4);
		binReads = new uint64_t[binReadSize];
		binReads = new uint64_t[binReadSize+1];
		return size;
		}

		@@ -495,7 +495,7 @@ class seqFileReadInfo {

		void writeReverseComplementString(string &name, string &content, fstream &file)
		{
		file << ">" << name << endl;
		file << ">" << name << "\n";
		flipNswap(content);
		file << content ;
		}
		@@ -747,15 +747,15 @@ class tmpFilesInfo {
		{
		if (revComplement & 0x1){
		if (commonData::lenInHeader) {
		cout << "> " << (itQ).seq << " Reverse" << " Len = " << ((itQ).end-(*itQ).start+2)/2 << endl;
		cout << "> " << (itQ).seq << " Reverse" << " Len = " << ((itQ).end-(*itQ).start+2)/2 << "\n";
		}else{
		cout << "> " << (*itQ).seq << " Reverse" << endl;
		cout << "> " << (*itQ).seq << " Reverse" << "\n";
		}
		}else{
		if (commonData::lenInHeader){
		cout << "> " << (itQ).seq << " Len = " << ((itQ).end-(*itQ).start+2)/2 << endl;
		cout << "> " << (itQ).seq << " Len = " << ((itQ).end-(*itQ).start+2)/2 << "\n";
		}else{
		cout << "> " << (*itQ).seq << endl;
		cout << "> " << (*itQ).seq << "\n";
		}
		}
		}
		@@ -873,14 +873,14 @@ class tmpFilesInfo {
		if (rRef-lRef+2 >= static_cast<uint64_t>(commonData::minMemLen)){
		if (refSeqInfo.size() == 1 && !commonData::fourColOutput) {
		if ((revComplement & 0x1) && commonData::relQueryPos)
		cout << " " << setw(15) << ((lRef+2)/2) << setw(15) << ((itQ).end-(itQ).start-lQue+2)/2 << setw(15) << ((rRef-lRef+2)/2) << endl;
		cout << " " << setw(15) << ((lRef+2)/2) << setw(15) << ((itQ).end-(itQ).start-lQue+2)/2 << setw(15) << ((rRef-lRef+2)/2) << "\n";
		else
		cout << " " << setw(15) << ((lRef+2)/2) << setw(15) << ((lQue+2)/2) << setw(15) << ((rRef-lRef+2)/2) << endl;
		cout << " " << setw(15) << ((lRef+2)/2) << setw(15) << ((lQue+2)/2) << setw(15) << ((rRef-lRef+2)/2) << "\n";
		}else{
		if ((revComplement & 0x1) && commonData::relQueryPos) {
		cout << " " << setw(30) << std::left <<(itR).seq << setw(15) << ((lRef+2)/2) << setw(15) << ((itQ).end-(*itQ).start-lQue+2)/2 << setw(15) << ((rRef-lRef+2)/2) << endl;
		cout << " " << setw(30) << std::left <<(itR).seq << setw(15) << ((lRef+2)/2) << setw(15) << ((itQ).end-(*itQ).start-lQue+2)/2 << setw(15) << ((rRef-lRef+2)/2) << "\n";
		}else{
		cout << " " << setw(30) << std::left <<(*itR).seq << setw(15) << ((lRef+2)/2) << setw(15) << ((lQue+2)/2) << setw(15) << ((rRef-lRef+2)/2) << endl;
		cout << " " << setw(30) << std::left <<(*itR).seq << setw(15) << ((lRef+2)/2) << setw(15) << ((lQue+2)/2) << setw(15) << ((rRef-lRef+2)/2) << "\n";
		}
		}
		}
		@@ -959,31 +959,31 @@ class tmpFilesInfo {

		filePtr = forFile;
		if(getline((*filePtr), line).good())
		cout << line << endl;
		cout << line << "\n";

		while(getline((*filePtr), line).good()) {
		if(line[0] == '>'){
		if (last_line.size())
		cout << last_line << endl;
		cout << last_line << "\n";
		last_line = line;
		if (filePtr == forFile) {
		filePtr = revFile;
		if (first) {
		if(getline((*filePtr), line).good())
		cout << line << endl;
		cout << line << "\n";
		first=0;
		}
		}else
		filePtr = forFile;
		continue;
		}
		cout << line << endl;
		cout << line << "\n";
		}

		cout << last_line << endl;
		cout << last_line << "\n";
		filePtr = revFile;
		while(getline((*filePtr), line).good())
		cout << line << endl;
		cout << line << "\n";

		(*revFile).close();
		(*forFile).close();