static CLParserPtr initParser()
{
CLParserPtr optionsParser = hdf5CLParserInstance(true);
optionsParser->addArgument("inFile", "existing tree");
optionsParser->addArgument("deleteNode", "(leaf) genome to delete");
optionsParser->addOptionFlag("noMarkAncestors", "don't mark ancestors for"
" update", false);
return optionsParser;
}
static CLParserPtr initParser()
{
CLParserPtr optionsParser = hdf5CLParserInstance(true);
optionsParser->addArgument("inFile", "existing tree");
optionsParser->addOption("bottomAlignmentFile", "hal file containing an "
"alignment of the genome and its children. "
"Required for non-leaf genomes.", "\"\"");
optionsParser->addOption("topAlignmentFile", "hal file containing an "
"alignment of the genome, its parent, and "
"its siblings. Required if the genome to be "
"replaced is not the root.", "\"\"");
optionsParser->addArgument("genomeName", "name of genome to be replaced");
optionsParser->addOptionFlag("noMarkAncestors", "don't mark ancestors for"
" update", false);
return optionsParser;
}
static CLParserPtr initParser()
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->addArgument("halFile", "input hal file");
optionsParser->addArgument("srcGenome", "source genome name");
optionsParser->addArgument("srcBed", "path of input bed file. set as stdin "
"to stream from standard input");
optionsParser->addArgument("tgtGenome", "target genome name");
optionsParser->addArgument("tgtBed", "path of output bed file. set as stdout"
" to stream to standard output.");
optionsParser->addOptionFlag("noDupes", "do not map between duplications in"
" graph.", false);
optionsParser->addOptionFlag("append", "append results to tgtBed", false);
optionsParser->addOption("inBedVersion", "bed version of input file "
"as integer between 3 and 9 or 12 reflecting "
"the number of columns (see bed "
"format specification for more details). Will "
"be autodetected by default.", 0);
optionsParser->addOption("outBedVersion", "bed version of output file "
"as integer between 3 and 9 or 12 reflecting "
"the number of columns (see bed "
"format specification for more details). Will "
"be same as input by default.", 0);
optionsParser->addOption("coalescenceLimit", "coalescence limit genome:"
" the genome at or above the MRCA of source"
" and target at which we stop looking for"
" homologies (default: MRCA)",
"");
optionsParser->addOptionFlag("outPSL", "write output in PSL instead of "
"bed format. overrides --outBedVersion when "
"specified.", false);
optionsParser->addOptionFlag("outPSLWithName", "write output as input BED name followed by PSL line instead of "
"bed format. overrides --outBedVersion when "
"specified.", false);
optionsParser->addOptionFlag("keepExtra", "keep extra columns. these are "
"columns in the input beyond the specified or "
"detected bed version, and which are cut by "
"default.", false);
optionsParser->addOptionFlag("tab", "input is tab-separated. this allows"
" column entries to contain spaces. if this"
" flag is not set, both spaces and tabs are"
" used to separate input columns.", false);
optionsParser->setDescription("Map BED genome interval coordinates between "
"two genomes.");
return optionsParser;
}
static CLParserPtr initParser()
{
CLParserPtr optionsParser = hdf5CLParserInstance(true);
optionsParser->addArgument("inFile", "existing tree");
optionsParser->addArgument("botAlignmentFile", "tree containing insert, its "
"proper bottom segments, and the new leaf genome");
optionsParser->addArgument("topAlignmentFile", "tree containing insert, its "
"parent, and its proper top segments");
optionsParser->addArgument("parentName", "insert's future parent");
optionsParser->addArgument("insertName", "insert name");
optionsParser->addArgument("childName", "insert's future child");
optionsParser->addArgument("leafName", "name of new leaf genome");
optionsParser->addArgument("upperBranchLength", "length of branch from parent"
" to insert");
optionsParser->addArgument("leafBranchLength", "leaf branch length");
optionsParser->addOptionFlag("noMarkAncestors", "don't mark ancestors for"
" update", false);
return optionsParser;
}
int main(int argc, char** argv)
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->addArgument("halFile", "path to hal file to validate");
optionsParser->setDescription("Check if hal database is valid");
string path;
try
{
optionsParser->parseOptions(argc, argv);
path = optionsParser->getArgument<string>("halFile");
}
catch(exception& e)
{
cerr << e.what() << endl;
optionsParser->printUsage(cerr);
exit(1);
}
try
{
AlignmentConstPtr alignment = openHalAlignmentReadOnly(path, optionsParser);
validateAlignment(alignment);
}
catch(hal_exception& e)
{
cerr << "hal exception caught: " << e.what() << endl;
return 1;
}
catch(exception& e)
{
cerr << "Exception caught: " << e.what() << endl;
return 1;
}
cout << "\nFile valid" << endl;
return 0;
}
int main(int argc, char** argv)
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->setDescription("Rertrieve chain (pairwise alignment) "
"information from a hal database.\n"
"WARNING: THIS TOOL WAS NEVER FINISHED OR"
" TESTED. USE AT OWN RISK. PLEASE "
"CONSIDER halLiftover --outPSL INSTEAD.");
optionsParser->addArgument("halFile", "path to hal file to analyze");
optionsParser->addArgument("genome", "(query) genome to process");
optionsParser->addOption("sequence", "sequence name in query genome ("
"all sequences if not specified)", "\"\"");
optionsParser->addOption("start", "start position in query genome", 0);
optionsParser->addOption("length", "maximum length of chain to output.", 0);
optionsParser->addOption("chainFile", "path for output file. stdout if not"
" specified", "\"\"");
optionsParser->addOption("maxGap",
"maximum indel length to be considered a gap within"
" a chain.",
20);
string halPath;
string chainPath;
string genomeName;
string sequenceName;
hal_size_t start;
hal_size_t length;
hal_size_t maxGap;
try
{
optionsParser->parseOptions(argc, argv);
halPath = optionsParser->getArgument<string>("halFile");
genomeName = optionsParser->getArgument<string>("genome");
sequenceName = optionsParser->getOption<string>("sequence");
start = optionsParser->getOption<hal_size_t>("start");
length = optionsParser->getOption<hal_size_t>("length");
chainPath = optionsParser->getOption<string>("chainFile");
maxGap = optionsParser->getOption<hal_size_t>("maxGap");
}
catch(exception& e)
{
cerr << e.what() << endl;
optionsParser->printUsage(cerr);
exit(1);
}
try
{
cerr << "WARNING: THIS TOOL WAS NEVER FINISHED OR TESTED. USE AT OWN RISK."
<< " PLEASE CONSIDER halLiftover --outPSL INSTEAD." <<endl;
AlignmentConstPtr alignment = openHalAlignmentReadOnly(halPath,
optionsParser);
const Genome* genome = alignment->openGenome(genomeName);
if (genome == NULL)
{
throw hal_exception(string("Genome not found: ") + genomeName);
}
hal_index_t endPosition =
length > 0 ? start + length : genome->getSequenceLength();
const Sequence* sequence = NULL;
if (sequenceName != "\"\"")
{
sequence = genome->getSequence(sequenceName);
if (sequence == NULL)
{
throw hal_exception(string("Sequence not found: ") + sequenceName);
}
start += sequence->getStartPosition();
endPosition =
length > 0 ? start + length : sequence->getSequenceLength();
}
ofstream ofile;
ostream& outStream = chainPath == "\"\"" ? cout : ofile;
if (chainPath != "\"\"")
{
ofile.open(chainPath.c_str());
if (!ofile)
{
throw hal_exception(string("Error opening output file ") +
chainPath);
}
}
TopSegmentIteratorConstPtr top = genome->getTopSegmentIterator();
top->toSite(start, false);
// do slicing here;
GappedTopSegmentIteratorConstPtr gtop =
genome->getGappedTopSegmentIterator(top->getArrayIndex(), maxGap);
// need to review!
Chain chain;
chain._id = 0;
while (gtop->getRightArrayIndex() <
(hal_index_t)genome->getNumTopSegments() &&
gtop->getLeft()->getStartPosition() < endPosition)
{
if (gtop->hasParent() == true)
{
hal_offset_t leftOffset = 0;
if ((hal_index_t)start > gtop->getStartPosition()
&& (hal_index_t)start < gtop->getEndPosition())
{
leftOffset = start - gtop->getStartPosition() ;
}
hal_offset_t rightOffset = 0;
if (endPosition - 1 > gtop->getStartPosition()
&& endPosition - 1 < gtop->getEndPosition())
{
rightOffset = gtop->getEndPosition() + 1 - endPosition;
}
// need to do offsets for edge cases
gtIteratorToChain(gtop, chain, leftOffset, rightOffset);
outStream << chain;
++chain._id;
}
gtop->toRight();
}
}
catch(hal_exception& e)
{
cerr << "hal exception caught: " << e.what() << endl;
return 1;
}
catch(exception& e)
{
cerr << "Exception caught: " << e.what() << endl;
return 1;
}
return 0;
}
int main(int argc, char** argv)
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->setDescription("Retrieve basic statistics from a hal database");
optionsParser->addArgument("halFile", "path to hal file to analyze");
optionsParser->addOptionFlag("genomes", "print only a list of genomes "
"in alignment", false);
optionsParser->addOption("sequences", "print list of sequences in given "
"genome", "\"\"");
optionsParser->addOption("sequenceStats", "print stats for each sequence in "
"given genome", "\"\"");
optionsParser->addOption("bedSequences", "print sequences of given genome "
"in bed format",
"\"\"");
optionsParser->addOptionFlag("tree", "print only the NEWICK tree", false);
optionsParser->addOptionFlag("branches", "print list of branches. "
"Each branch is specified by the child genome",
false);
optionsParser->addOption("span", "print branches on path (or spanning tree) "
"between comma "
"separated list of genomes", "\"\"");
optionsParser->addOption("spanRoot", "print genomes on path"
"(or spanning tree) between comma "
"separated list of genomes. Different from --span"
"only in that the spanning tree root is also "
"given", "\"\"");
optionsParser->addOption("children", "print names of children of given "
"genome", "\"\"");
optionsParser->addOptionFlag("root", "print root genome name", false);
optionsParser->addOption("parent", "print name of parent of given genome",
"\"\"");
optionsParser->addOption("branchLength", "print branch length between "
"given genome and its parent in the tree",
"\"\"");
optionsParser->addOption("numSegments", "print numTopSegments "
"numBottomSegments for given genome.",
"\"\"");
optionsParser->addOption("baseComp", "print base composition for given "
"genome by sampling every step bases. Parameter "
"value is of the form genome,step. Ex: "
"--baseComp human,1000. The ouptut is of the form "
"fraction_of_As fraction_of_Gs fraction_of_Cs "
"fraction_of_Ts.",
"\"\"");
optionsParser->addOption("genomeMetaData", "print metadata for given genome, "
"one entry per line, tab-seperated.", "\"\"");
optionsParser->addOption("chromSizes", "print the name and length of each"
" sequence in a given genome. This is a subset"
" of the"
" information returned by --sequenceStats but is"
" useful because it is in the format used by"
" wigToBigWig",
"\"\"");
optionsParser->addOption("percentID",
"print % ID of a genome with all other genomes."
"Only non-duplicated and unambiguous sites are"
"considered",
"\"\"");
optionsParser->addOption("coverage",
"print histogram of coverage of a genome with"
" all genomes", "\"\"");
optionsParser->addOption("topSegments",
"print coordinates of all top segments of given"
" genome in BED format.", "\"\"");
optionsParser->addOption("bottomSegments",
"print coordinates of all bottom segments of given"
" genome in BED format.", "\"\"");
optionsParser->addOptionFlag("allCoverage",
"print histogram of coverage from all genomes to"
" all genomes", false);
string path;
bool listGenomes;
string sequencesFromGenome;
string sequenceStatsFromGenome;
string bedSequencesFromGenome;
string spanGenomes;
string spanRootGenomes;
bool tree;
bool branches;
string childrenFromGenome;
string parentFromGenome;
bool printRoot;
string nameForBL;
string numSegmentsGenome;
string baseCompPair;
string genomeMetaData;
string chromSizesFromGenome;
string percentID;
string coverage;
string topSegments;
string bottomSegments;
bool allCoverage;
try
{
optionsParser->parseOptions(argc, argv);
path = optionsParser->getArgument<string>("halFile");
listGenomes = optionsParser->getFlag("genomes");
sequencesFromGenome = optionsParser->getOption<string>("sequences");
sequenceStatsFromGenome = optionsParser->getOption<string>("sequenceStats");
bedSequencesFromGenome = optionsParser->getOption<string>("bedSequences");
tree = optionsParser->getFlag("tree");
spanGenomes = optionsParser->getOption<string>("span");
spanRootGenomes = optionsParser->getOption<string>("spanRoot");
branches = optionsParser->getFlag("branches");
childrenFromGenome = optionsParser->getOption<string>("children");
parentFromGenome = optionsParser->getOption<string>("parent");
printRoot = optionsParser->getFlag("root");
nameForBL = optionsParser->getOption<string>("branchLength");
numSegmentsGenome = optionsParser->getOption<string>("numSegments");
baseCompPair = optionsParser->getOption<string>("baseComp");
genomeMetaData = optionsParser->getOption<string>("genomeMetaData");
chromSizesFromGenome = optionsParser->getOption<string>("chromSizes");
percentID = optionsParser->getOption<string>("percentID");
coverage = optionsParser->getOption<string>("coverage");
topSegments = optionsParser->getOption<string>("topSegments");
bottomSegments = optionsParser->getOption<string>("bottomSegments");
allCoverage = optionsParser->getFlag("allCoverage");
size_t optCount = listGenomes == true ? 1 : 0;
if (sequencesFromGenome != "\"\"") ++optCount;
if (tree == true) ++optCount;
if (sequenceStatsFromGenome != "\"\"") ++optCount;
if (bedSequencesFromGenome != "\"\"") ++optCount;
if (spanGenomes != "\"\"") ++optCount;
if (spanRootGenomes != "\"\"") ++optCount;
if (branches) ++ optCount;
if (childrenFromGenome != "\"\"") ++optCount;
if (parentFromGenome != "\"\"") ++optCount;
if (printRoot) ++optCount;
if (nameForBL != "\"\"") ++optCount;
if (numSegmentsGenome != "\"\"") ++optCount;
if (baseCompPair != "\"\"") ++optCount;
if (genomeMetaData != "\"\"") ++optCount;
if (chromSizesFromGenome != "\"\"") ++optCount;
if (percentID != "\"\"") ++optCount;
if (coverage != "\"\"") ++optCount;
if (topSegments != "\"\"") ++optCount;
if (bottomSegments != "\"\"") ++optCount;
if (allCoverage) ++optCount;
if (optCount > 1)
{
throw hal_exception("--genomes, --sequences, --tree, --span, --spanRoot, "
"--branches, --sequenceStats, --children, --parent, "
"--bedSequences, --root, --numSegments, --baseComp, "
"--genomeMetaData, --chromSizes, --percentID, "
"--coverage, --topSegments, --bottomSegments, "
"--allCoverage "
"and --branchLength options are exclusive");
}
}
catch(exception& e)
{
cerr << e.what() << endl;
optionsParser->printUsage(cerr);
exit(1);
}
try
{
AlignmentConstPtr alignment = openHalAlignmentReadOnly(path, optionsParser);
if (listGenomes == true && alignment->getNumGenomes() > 0)
{
printGenomes(cout, alignment);
}
else if (sequencesFromGenome != "\"\"")
{
printSequences(cout, alignment, sequencesFromGenome);
}
else if (tree == true)
{
cout << alignment->getNewickTree() << endl;
}
else if (sequenceStatsFromGenome != "\"\"")
{
printSequenceStats(cout, alignment, sequenceStatsFromGenome);
}
else if (bedSequencesFromGenome != "\"\"")
{
printBedSequenceStats(cout, alignment, bedSequencesFromGenome);
}
else if (spanGenomes != "\"\"")
{
printBranchPath(cout, alignment, chopString(spanGenomes, ","), false);
}
else if (spanRootGenomes != "\"\"")
{
printBranchPath(cout, alignment, chopString(spanRootGenomes, ","), true);
}
else if (branches == true)
{
printBranches(cout, alignment);
}
else if (childrenFromGenome != "\"\"")
{
printChildren(cout, alignment, childrenFromGenome);
}
else if (parentFromGenome != "\"\"")
{
printParent(cout, alignment, parentFromGenome);
}
else if (printRoot == true)
{
printRootName(cout, alignment);
}
else if (nameForBL != "\"\"")
{
printBranchLength(cout, alignment, nameForBL);
}
else if (numSegmentsGenome != "\"\"")
{
printNumSegments(cout, alignment, numSegmentsGenome);
}
else if (baseCompPair != "\"\"")
{
printBaseComp(cout, alignment, baseCompPair);
}
else if (genomeMetaData != "\"\"")
{
printGenomeMetaData(cout, alignment, genomeMetaData);
}
else if (chromSizesFromGenome != "\"\"")
{
printChromSizes(cout, alignment, chromSizesFromGenome);
}
else if (percentID != "\"\"")
{
printPercentID(cout, alignment, percentID);
}
else if (coverage != "\"\"") {
printCoverage(cout, alignment, coverage);
}
else if (topSegments != "\"\"") {
printSegments(cout, alignment, topSegments, true);
}
else if (bottomSegments != "\"\"") {
printSegments(cout, alignment, bottomSegments, false);
} else if (allCoverage) {
printAllCoverage(cout, alignment);
}
else
{
HalStats halStats(alignment);
cout << endl << "hal v" << alignment->getVersion() << "\n" << halStats;
}
}
catch(hal_exception& e)
{
cerr << "hal exception caught: " << e.what() << endl;
return 1;
}
catch(exception& e)
{
cerr << "Exception caught: " << e.what() << endl;
return 1;
}
return 0;
}
int main(int argc, char** argv)
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->setDescription("Write masked intervals of genome into bed "
"file");
optionsParser->addArgument("halFile", "path to hal file to analyze");
optionsParser->addArgument("genome", "name of genome to process");
optionsParser->addOption("maskFile", "path to bed file to write to",
"stdout");
optionsParser->addOption("extend", "extend masked regions by given num. "
"of bases.", 0);
optionsParser->addOption("extendPct", "extend masked regions by percentage"
" of their lengths", 0);
string halPath;
string genomeName;
string bedPath;
hal_size_t extend;
double extendPct;
try
{
optionsParser->parseOptions(argc, argv);
halPath = optionsParser->getArgument<string>("halFile");
genomeName = optionsParser->getArgument<string>("genome");
bedPath = optionsParser->getOption<string>("maskFile");
extend = optionsParser->getOption<hal_size_t>("extend");
extendPct = optionsParser->getOption<double>("extendPct");
if (extend != 0 && extendPct != 0.)
{
throw hal_exception("--extend and --extendPct options are exclusive.");
}
}
catch(exception& e)
{
cerr << e.what() << endl;
optionsParser->printUsage(cerr);
exit(1);
}
try
{
AlignmentConstPtr alignment = openHalAlignmentReadOnly(halPath,
optionsParser);
const Genome* genome = alignment->openGenome(genomeName);
if (genome == NULL)
{
throw hal_exception(string("Genome ") + genomeName + " not found.");
}
ostream* bedStream = &cout;
bool newBed = false;
if (bedPath != "stdout")
{
bedStream = new ofstream(bedPath.c_str());
newBed = true;
}
if (!bedStream)
{
throw hal_exception(string("Error opening ") + bedPath + " for writing");
}
MaskExtractor mask;
mask.extract(alignment, genome, bedStream, extend, extendPct);
if (newBed)
{
delete bedStream;
}
}
catch(hal_exception& e)
{
cerr << "hal exception caught: " << e.what() << endl;
return 1;
}
catch(exception& e)
{
cerr << "Exception caught: " << e.what() << endl;
return 1;
}
return 0;
}
int main(int argc, char** argv)
{
CLParserPtr optionsParser = hdf5CLParserInstance();
optionsParser->setDescription("Rertrieve basic statics from a hal database");
optionsParser->addArgument("halFile", "path to hal file to analyze");
optionsParser->addOptionFlag("genomes", "print only a list of genomes "
"in alignment", false);
optionsParser->addOption("sequences", "print list of sequences in given "
"genome", "\"\"");
optionsParser->addOption("sequenceStats", "print stats for each sequence in "
"given genome", "\"\"");
optionsParser->addOptionFlag("tree", "print only the NEWICK tree", false);
optionsParser->addOptionFlag("branches", "print list of branches. "
"Each branch is specified by the child genome",
false);
optionsParser->addOption("span", "print branches on path (or spanning tree) "
"between comma "
"separated list of genomes", "\"\"");
optionsParser->addOption("spanRoot", "print genomes on path"
"(or spanning tree) between comma "
"separated list of genomes. Different from --path"
"only in that the spanning tree root is also "
"given", "\"\"");
string path;
bool listGenomes;
string sequencesFromGenome;
string sequenceStatsFromGenome;
string spanGenomes;
string spanRootGenomes;
bool tree;
bool branches;
try
{
optionsParser->parseOptions(argc, argv);
path = optionsParser->getArgument<string>("halFile");
listGenomes = optionsParser->getFlag("genomes");
sequencesFromGenome = optionsParser->getOption<string>("sequences");
sequenceStatsFromGenome = optionsParser->getOption<string>("sequenceStats");
tree = optionsParser->getFlag("tree");
spanGenomes = optionsParser->getOption<string>("span");
spanRootGenomes = optionsParser->getOption<string>("spanRoot");
branches = optionsParser->getFlag("branches");
size_t optCount = listGenomes == true ? 1 : 0;
if (sequencesFromGenome != "\"\"") ++optCount;
if (tree == true) ++optCount;
if (sequenceStatsFromGenome != "\"\"") ++optCount;
if (spanGenomes != "\"\"") ++optCount;
if (spanRootGenomes != "\"\"") ++optCount;
if (branches) ++optCount;
if (optCount > 1)
{
throw hal_exception("--genomes, --sequences, --tree, --span, "
"--spanRoot, --branches "
"and --sequenceStats "
"options are mutually exclusive");
}
}
catch(exception& e)
{
cerr << e.what() << endl;
optionsParser->printUsage(cerr);
exit(1);
}
try
{
AlignmentConstPtr alignment = openHalAlignmentReadOnly(path, optionsParser);
if (listGenomes == true && alignment->getNumGenomes() > 0)
{
printGenomes(cout, alignment);
}
else if (sequencesFromGenome != "\"\"")
{
printSequences(cout, alignment, sequencesFromGenome);
}
else if (tree == true)
{
cout << alignment->getNewickTree() << endl;
}
else if (sequenceStatsFromGenome != "\"\"")
{
printSequenceStats(cout, alignment, sequenceStatsFromGenome);
}
else if (spanGenomes != "\"\"")
{
printBranchPath(cout, alignment, chopString(spanGenomes, ","), false);
}
else if (spanRootGenomes != "\"\"")
{
printBranchPath(cout, alignment, chopString(spanRootGenomes, ","), true);
}
else if (branches == true)
{
printBranches(cout, alignment);
}
else
{
HalStats halStats(alignment);
cout << endl << "hal v" << alignment->getVersion() << "\n" << halStats;
}
}
catch(hal_exception& e)
{
cerr << "hal exception caught: " << e.what() << endl;
return 1;
}
catch(exception& e)
{
cerr << "Exception caught: " << e.what() << endl;
return 1;
}
return 0;
}
