//-------------------------------------------------------------------------
void CWinMaskApplication::Init(void)
{
HideStdArgs(fHideLogfile | fHideConffile | fHideVersion | fHideDryRun);
auto_ptr< CArgDescriptions > arg_desc( new CArgDescriptions );
// Set the program description
arg_desc->SetUsageContext( GetArguments().GetProgramBasename(),
USAGE_LINE );
CWinMaskConfig::AddWinMaskArgs(*arg_desc);
// Parse the arguments according to descriptions.
SetupArgDescriptions(arg_desc.release());
}
void CSeqDBPerfApp::Init()
{
HideStdArgs(fHideConffile | fHideFullVersion | fHideXmlHelp | fHideDryRun);
auto_ptr<CArgDescriptions> arg_desc(new CArgDescriptions);
// Specify USAGE context
arg_desc->SetUsageContext(GetArguments().GetProgramBasename(),
"CSeqDB performance testing client");
arg_desc->SetCurrentGroup("BLAST database options");
arg_desc->AddDefaultKey("db", "dbname", "BLAST database name",
CArgDescriptions::eString, "nr");
arg_desc->AddDefaultKey("dbtype", "molecule_type",
"Molecule type stored in BLAST database",
CArgDescriptions::eString, "guess");
arg_desc->SetConstraint("dbtype", &(*new CArgAllow_Strings,
"nucl", "prot", "guess"));
arg_desc->SetCurrentGroup("Retrieval options");
arg_desc->AddFlag("scan_uncompressed",
"Do a full database scan of uncompressed sequence data", true);
arg_desc->AddFlag("scan_compressed",
"Do a full database scan of compressed sequence data", true);
arg_desc->AddFlag("get_metadata",
"Retrieve BLAST database metadata", true);
arg_desc->SetDependency("scan_compressed", CArgDescriptions::eExcludes,
"scan_uncompressed");
arg_desc->SetDependency("scan_compressed", CArgDescriptions::eExcludes,
"get_metadata");
arg_desc->SetDependency("scan_uncompressed", CArgDescriptions::eExcludes,
"get_metadata");
arg_desc->AddDefaultKey("num_threads", "number",
"Number of threads to use (requires OpenMP)",
CArgDescriptions::eInteger, "1");
arg_desc->SetConstraint("num_threads", new CArgAllow_Integers(0, kMax_Int));
//arg_desc->AddFlag("one_db_handle", "Build only 1 CSeqDB object?", true);
//arg_desc->SetDependency("one_db_handle", CArgDescriptions::eRequires, "num_threads");
arg_desc->SetCurrentGroup("Output configuration options");
arg_desc->AddDefaultKey("out", "output_file", "Output file name",
CArgDescriptions::eOutputFile, "-");
SetupArgDescriptions(arg_desc.release());
}
void CVecScreenApp::Init()
{
HideStdArgs(fHideLogfile | fHideConffile | fHideFullVersion | fHideXmlHelp | fHideDryRun);
auto_ptr<CArgDescriptions> arg_desc(new CArgDescriptions);
arg_desc->SetUsageContext(GetArguments().GetProgramBasename(),
"Vector screening tool, version " +
CVecScreenVersion().Print());
arg_desc->SetCurrentGroup("Input query options");
arg_desc->AddDefaultKey(kArgQuery, "input_file",
"Input file name",
CArgDescriptions::eInputFile, kDfltArgQuery);
arg_desc->SetCurrentGroup("BLAST database options");
arg_desc->AddDefaultKey(kArgDb, "dbname", "BLAST database name",
CArgDescriptions::eString, kDefaultVectorDb);
arg_desc->SetCurrentGroup("Output configuration options");
arg_desc->AddDefaultKey(kArgOutput, "output_file", "Output file name",
CArgDescriptions::eOutputFile, "-");
//arg_desc->AddDefaultKey("outfmt", "format",
arg_desc->AddDefaultKey(kArgOutputFormat, "format",
"VecScreen results options:\n"
" 0 = Show alignments pairwise,\n"
" 1 = Do not show alignments, just contaminated range offsets\n",
CArgDescriptions::eInteger,
NStr::IntToString(kDfltArgOutputFormat));
arg_desc->SetConstraint(kArgOutputFormat,
new CArgAllowValuesBetween(0, CVecscreenRun::CFormatter::eEndValue-1, true));
// Produce Text output?
arg_desc->AddFlag("text_output", "Produce text output?", true);
arg_desc->SetCurrentGroup("");
SetupArgDescriptions(arg_desc.release());
}
BEGIN_NCBI_SCOPE
void CCompartApp::Init()
{
HideStdArgs(fHideLogfile | fHideConffile | fHideVersion);
auto_ptr<CArgDescriptions> argdescr(new CArgDescriptions);
argdescr->SetUsageContext(GetArguments().GetProgramName(),
"Compart v.1.35. Unless -qdb and -sdb are specified, "
"the tool expects tabular blast hits at stdin collated "
"by query and subject, e.g. with 'sort -k 1,1 -k 2,2'");
argdescr->AddOptionalKey ("qdb", "qdb", "cDNA BLAST database",
CArgDescriptions::eString);
argdescr->AddOptionalKey ("sdb", "sdb", "Genomic BLAST database",
CArgDescriptions::eString);
argdescr->AddFlag ("ho", "Print raw hits only - no compartments");
argdescr->AddDefaultKey("penalty", "penalty", "Per-compartment penalty",
CArgDescriptions::eDouble, "0.55");
argdescr->AddDefaultKey("min_idty", "min_idty", "Minimal overall identity. Note: in current implementation there is no sense to set different 'min_idty' and 'min_singleton_idty' (minimum is used anyway).",
CArgDescriptions::eDouble, "0.70");
argdescr->AddDefaultKey("min_singleton_idty", "min_singleton_idty",
"Minimal identity for singleton compartments. "
"The actual parameter passed to the compartmentization "
"procedure is least of this parameter multipled "
"by the seq length, and min_singleton_idty_bps. Note: in current implementation there is no sense to set different 'min_idty' and 'min_singleton_idty' (minimum is used anyway).",
CArgDescriptions::eDouble, "0.70");
argdescr->AddDefaultKey("min_singleton_idty_bps", "min_singleton_idty_bps",
"Minimal identity for singleton compartments "
"in base pairs. Default = parameter disabled.",
CArgDescriptions::eInteger, "9999999");
argdescr->AddDefaultKey ("max_intron", "max_intron",
"Maximum intron length (in base pairs)",
CArgDescriptions::eInteger,
NStr::IntToString
(CCompartmentFinder<THit>::s_GetDefaultMaxIntron()));
argdescr->AddDefaultKey("dropoff", "dropoff",
"Max score drop-off during hit extension.",
CArgDescriptions::eInteger,
NStr::IntToString(CElementaryMatching::
s_GetDefaultDropOff()));
argdescr->AddDefaultKey("min_query_len", "min_query_len",
"Minimum length for individual cDNA sequences.",
CArgDescriptions::eInteger, "50");
argdescr->AddDefaultKey("min_hit_len", "min_hit_len",
"Minimum length for reported hits in hits-only mode. "
"No effect in compartments mode.",
CArgDescriptions::eInteger, "16");
argdescr->AddDefaultKey ("maxvol", "maxvol",
"Maximum index volume size in MB (approximate)",
CArgDescriptions::eInteger,
"512");
argdescr->AddFlag("noxf", "[With external hits] Suppress overlap x-filtering: "
"print all compartment hits intact.");
argdescr->AddOptionalKey("seqlens", "seqlens",
"[With external hits] Two-column file with sequence IDs "
"and their lengths. If none supplied, the program will "
"attempt fetching the lengths from GenBank. "
"Cannot be used with -qdb.",
CArgDescriptions::eInputFile);
argdescr->AddDefaultKey("N", "N",
"[With external hits] Max number of compartments "
"per query (0 = All).",
CArgDescriptions::eInteger, "0");
CArgAllow* constrain01 (new CArgAllow_Doubles(0.0, 1.0));
argdescr->SetConstraint("penalty", constrain01);
argdescr->SetConstraint("min_idty", constrain01);
argdescr->SetConstraint("min_singleton_idty", constrain01);
CArgAllow_Integers* constrain_maxvol (new CArgAllow_Integers(128,1024));
argdescr->SetConstraint("maxvol", constrain_maxvol);
CArgAllow_Integers* constrain_minqlen (new CArgAllow_Integers(21,99999));
argdescr->SetConstraint("min_query_len", constrain_minqlen);
CArgAllow_Integers* constrain_minhitlen (new CArgAllow_Integers(1,99999));
argdescr->SetConstraint("min_hit_len", constrain_minhitlen);
SetupArgDescriptions(argdescr.release());
}
void CAppNWA::Init()
{
HideStdArgs(fHideLogfile | fHideConffile | fHideVersion);
auto_ptr<CArgDescriptions> argdescr(new CArgDescriptions);
argdescr->SetUsageContext(GetArguments().GetProgramName(),
"Demo application using xalgoalign library");
argdescr->AddDefaultKey
("matrix", "matrix", "scoring matrix",
CArgDescriptions::eString, "nucl");
argdescr->AddKey
("seq1", "seq1",
"the first input sequence in fasta file",
CArgDescriptions::eString);
argdescr->AddKey
("seq2", "seq2",
"the second input sequence in fasta file",
CArgDescriptions::eString);
argdescr->AddDefaultKey
("esf", "esf",
"End-space free alignment. Format: lrLR where each character "
"can be z (free end) or x (regular end) representing "
"left and right ends. First sequence's ends are specified first.",
CArgDescriptions::eString,
"xxxx");
argdescr->AddDefaultKey
("Wm", "match", "match bonus (nucleotide sequences)",
CArgDescriptions::eInteger,
NStr::IntToString(CNWAligner::GetDefaultWm()).c_str());
argdescr->AddDefaultKey
("Wms", "mismatch", "mismatch penalty (nucleotide sequences)",
CArgDescriptions::eInteger,
NStr::IntToString(CNWAligner::GetDefaultWms()).c_str());
argdescr->AddDefaultKey
("Wg", "gap", "gap opening penalty",
CArgDescriptions::eInteger,
NStr::IntToString(CNWAligner::GetDefaultWg()).c_str());
argdescr->AddDefaultKey
("Ws", "space", "gap extension (space) penalty",
CArgDescriptions::eInteger,
NStr::IntToString(CNWAligner::GetDefaultWs()).c_str());
argdescr->AddDefaultKey
("band", "band", "Band width in banded alignment",
CArgDescriptions::eInteger, "-1");
argdescr->AddDefaultKey
("shift", "shift",
"Band shift in banded alignment "
"(specify negative value to indicate second sequence)",
CArgDescriptions::eInteger, "0");
argdescr->AddFlag("mm",
"Use linear-memory alignment algorithm (Myers & Miller)");
argdescr->AddFlag("mt", "Use multiple threads");
// output formats
argdescr->AddOptionalKey
("o1", "o1", "Filename for type 1 output", CArgDescriptions::eString);
argdescr->AddOptionalKey
("o2", "o2", "Filename for type 2 output", CArgDescriptions::eString);
argdescr->AddOptionalKey
("ofasta", "ofasta",
"Generate gapped FastA output for the aligner sequences",
CArgDescriptions::eString);
argdescr->AddOptionalKey
("oasn", "oasn", "ASN.1 output filename", CArgDescriptions::eString);
CArgAllow_Strings* paa_st = new CArgAllow_Strings;
paa_st->Allow("nucl")->Allow("blosum62");
argdescr->SetConstraint("matrix", paa_st);
CArgAllow_Strings* paa_esf = new CArgAllow_Strings;
paa_esf->Allow("xxxx")->Allow("xxxz")->Allow("xxzx")->Allow("xxzz");
paa_esf->Allow("xzxx")->Allow("xzxz")->Allow("xzzx")->Allow("xzzz");
paa_esf->Allow("zxxx")->Allow("zxxz")->Allow("zxzx")->Allow("zxzz");
paa_esf->Allow("zzxx")->Allow("zzxz")->Allow("zzzx")->Allow("zzzz");
argdescr->SetConstraint("esf", paa_esf);
SetupArgDescriptions(argdescr.release());
}
void CAlignmentRefiner::Init(void)
{
// Create command-line argument descriptions class
auto_ptr<CArgDescriptions> argDescr(new CArgDescriptions);
// Specify USAGE context
argDescr->SetUsageContext(GetArguments().GetProgramBasename(),
"Alignment Refinement w/ 'Leave One Out'");
HideStdArgs(fHideLogfile | fHideConffile | fHideDryRun | fHideXmlHelp);
// *Mandatory*
// input CD/Fasta file name
argDescr->AddKey("i", "CdFilenameIn", "full filename of input CD w/ alignment to refine (ascii or binary)", argDescr->eString);
// base name for output CD file
argDescr->AddDefaultKey("o", "CdBasenameOut", "basename of output CD(s) containing refined alignment; output saved to 'basename_<number>.cn3'; ascii text by default", argDescr->eString, "refiner");
// output binary?
argDescr->AddFlag("ob", "output binary data");
// File for data from the refinement process
argDescr->AddOptionalKey
("details", "filename",
"create a file to save refinenment process details",
CArgDescriptions::eOutputFile,
CArgDescriptions::fPreOpen);
// quiet reporting of details (only at end of temp steps and trials)
argDescr->AddFlag("q", "shortened report; only Error level messages");
// Number of cycles per trial (a cycle consists of a LOO phase followed by a
// block-editing phase, either of which may be disabled for all cycles)
// In each LOO phase, each row is left out exactly once unless overriden by the 'nr' option.
// The alignment is NOT reset after a cycle; alignment IS reset after a trial.
argDescr->AddDefaultKey("nc", "integer", "number of cycles per trial; a cycle consists of one leave-one-out (LOO) phase, followed by a block editing phase. Either, but not both, of the phases in a cycle may be turned off.\n(Note: alignments do NOT reset between cycles)\n", argDescr->eInteger, "1");
argDescr->SetConstraint("nc", new CArgAllow_Integers(1, N_MAX_CYCLES));
// Convergence criteria
argDescr->AddDefaultKey("convSameScore", "double", "when >= this % of LOO attempts fail to change the score, stop further cycles", argDescr->eDouble, "0.95");
argDescr->SetConstraint("convSameScore", new CArgAllow_Doubles(0, 1.0));
//
// Leave-one-out parameters & options
//
argDescr->SetCurrentGroup(" Leave-one-out phase parameters and options ");
// disable LOO phase
argDescr->AddFlag("no_LOO", "do not perform the LOO phase of each cycle", true);
argDescr->SetDependency("no_LOO", CArgDescriptions::eExcludes, "selection_order");
// *Mandatory*
// Row selection order for LOO: randomly or based on the self-hit to the initial alignment.
argDescr->AddKey("selection_order", "integer", "Method for row selection in LOO phase:\n0 == randomly (use -n option to specify the number of separate trials)\n1 == increasing self-hit row score to input alignment's PSSM\n2 == decreasing self-hit row score to input alignment's PSSM\n", argDescr->eInteger);
argDescr->SetConstraint("selection_order", new CArgAllow_Integers(0, 2));
argDescr->AddAlias("so", "selection_order");
// switch from 'leave-one-out' to 'leave-N-out': recompute PSSM only after
// lno rows refined.
argDescr->AddDefaultKey("lno", "integer", "leave-N-out mode: speed up program by recomputing PSSM after 'lno' rows have been left out\n(values exceeding the number of rows interpreted as Nrows - 1)\nFor best results, value should be < 20% of number of rows in input alignment.\n", argDescr->eInteger, "1");
argDescr->SetConstraint("lno", new CArgAllow_Integers(0, kMax_Int));
// Number of row LOO events per cycle (may be equal to, more than or less than
// number of rows in the CD, and it is not guaranteed that each row will be chosen)
// argDescr->AddOptionalKey("nr", "integer", "absolute number of LOO attempts per LOO cycle (defaults to all eligible rows in alignment except the master)\n", argDescr->eInteger);
// argDescr->SetConstraint("nr", new CArgAllow_Integers(1, N_MAX_ROWS));
// declare whether structures are to be among the rows left out
argDescr->AddFlag("fix_structs", "do not perform LOO refinement on structures (i.e., those sequences having a PDB identifier)", true);
//
// Block aligner parameters / constraints on search space
//
// leave-one-out arguments used by block aligner
argDescr->AddDefaultKey("p", "double", "percentile parameter for loop-length cutoff in block aligner", argDescr->eDouble, "1.0");
argDescr->SetConstraint("p", new CArgAllow_Doubles(0, 10.0));
argDescr->AddDefaultKey("x", "integer", "extension parameter for loop-length cutoff in block aligner", argDescr->eInteger, "0");
argDescr->SetConstraint("x", new CArgAllow_Integers(0, kMax_Int));
argDescr->AddDefaultKey("c", "integer", "cutoff parameter for loop-length cutoff in block aligner", argDescr->eInteger, "0");
argDescr->SetConstraint("c", new CArgAllow_Integers(0, kMax_Int));
// Constraints on sequence length used in search for improved alignments
argDescr->AddFlag("fs", "allow refiner to use full sequence (by default, refinement is constrained to initial aligned footprint)", true);
argDescr->AddDefaultKey("ex","integer", "footprint extension size (symmetric for N- and C-termini); positive values extend the footprint; negative values shrink the footprint and can be used with -fs; relevant extension overriden by -nex and/or -cex", argDescr->eInteger, "0");
argDescr->AddOptionalKey("nex","integer", "N-terminal footprint extension size; positive values extend the footprint; negative values shrink the footprint and can be used with -fs; overrides any N-terminal extension from -ex", argDescr->eInteger);
argDescr->AddOptionalKey("cex","integer", "C-terminal footprint extension size; positive values extend the footprint; negative values shrink the footprint and can be used with -fs; overrides any C-terminal extension from -ex\n", argDescr->eInteger);
// Block freezing/un-freezing
argDescr->AddFlag("ab", "realign all blocks; overrides -f and -l options", true);
argDescr->AddOptionalKey("f", "integer", "first block to realign (post-IBM, from 1); overridden if -ab set", argDescr->eInteger);
argDescr->SetConstraint("f", new CArgAllow_Integers(1, kMax_Int));
argDescr->AddOptionalKey("l", "integer", "last block to realign (post-IBM, from 1); overridden if -ab set", argDescr->eInteger);
argDescr->SetConstraint("l", new CArgAllow_Integers(1, kMax_Int));
// argDescr->AddExtra(0, 25, "Block number (post-IBM, from 1) to freeze: can override -ab or -f, -l for a specific block(s) in a range of unfrozen blocks\n", argDescr->eInteger);
argDescr->AddExtra(0, 25, "Row OR block numbers to exclude from LOO (from 1 to # rows/blocks); master == row 1 is always excluded\n(See 'extras_are_blocks' flag.)\n", argDescr->eInteger);
// Flag to tell whether the extra args are row numbers (if not present) or block numbers (if present).
argDescr->AddFlag("extras_are_blocks", "treat extra arguments as block numbers (default is to treat them as row numbers)", true);
argDescr->SetCurrentGroup("");
//
// Random selection order related options
//
argDescr->SetCurrentGroup(" Random selection order specific options ");
// Number of trials
argDescr->AddDefaultKey("n", "integer", "number of independent trials (restarts from original alignment)\nNOTE: only relevant if use random selection order of rows; ignored if deterministic selection order used (see 'selection_order')", argDescr->eInteger, "3");
argDescr->SetConstraint("n", new CArgAllow_Integers(1, N_MAX_TRIALS));
argDescr->AddDefaultKey("nout", "integer", "number of output CDs; save CDs from the top 'nout' trials", argDescr->eInteger, "1");
argDescr->SetConstraint("nout", new CArgAllow_Integers(1, N_MAX_TRIALS));
argDescr->AddDefaultKey("convScoreChange", "double", "when avg. deviation <= this % of mean score for M trials, stop further trials", argDescr->eDouble, "0.001");
argDescr->SetConstraint("convScoreChange", new CArgAllow_Doubles(0, 1.0));
// Specify seed to RNG
argDescr->AddOptionalKey("seed", "positive_integer", "specify the seed for random number generation\n", argDescr->eInteger);
argDescr->SetConstraint("seed", new CArgAllow_Integers(1000, kMax_Int-1));
argDescr->SetCurrentGroup("");
//
// Block editing options
//
argDescr->SetCurrentGroup(" Block-editing phase parameters and options ");
argDescr->AddFlag("be_fix", "do not modify block boundaries", true);
argDescr->AddFlag("be_noShrink", "do not attempt to shrink block boundaries", true);
argDescr->AddFlag("be_shrinkFirst", "if set, try to shrink before extending a block; requires be_alg = both\n\n", true);
argDescr->AddDefaultKey("be_minSize", "integer", "smallest allowed block width (set to 0 to allow block deletion events);\nblocks that start smaller than min size are not truncated.", argDescr->eInteger, "1");
argDescr->SetConstraint("be_minSize", new CArgAllow_Integers(0, 1000));
// Scoring methods and algorithms for block edits.
argDescr->AddDefaultKey("be_alg", "string", "block editing method", argDescr->eString, "both");
argDescr->SetConstraint("be_alg", &(*new CArgAllow_Strings, "extend", "greedyExt", "shrink", "both"));
argDescr->AddDefaultKey("be_score", "string", "column scoring method", argDescr->eString, "3.3.3");
argDescr->SetConstraint("be_score", &(*new CArgAllow_Strings, "vote", "sumScores", "median", "scoreWeight", "compound", "3.3.3"));
/////////////////////////
// The following three options define the parameters for the 'compound' scoring method.
// '3.3.3' is a special shortcut compound method that uses 3, .3, .3 for these values,
// respectively.
// For future: Use 'compound' to use same three scorers w/ non-default values.
/////////////////////////
argDescr->AddDefaultKey("be_median", "integer", "if be_score = 'median', '3.3.3', value column median must equal or exceed", argDescr->eInteger, "3");
argDescr->SetConstraint("be_median", new CArgAllow_Integers(-10, 20));
argDescr->AddDefaultKey("be_negScore", "double", "negative score fraction: fraction of total column score less than zero\nto be added to a block, column's fraction must have at least this value\n** used for 'scoreWeight', 'compound', '3.3.3' scoring only", argDescr->eDouble, ".3");
argDescr->SetConstraint("be_negScore", new CArgAllow_Doubles(0, 1));
argDescr->AddDefaultKey("be_negRows", "double", "negative rows fraction: largest fraction of rows in a column with a negative value\nfor the column to be added to a block\n** used for 'compound', '3.3.3' scoring only", argDescr->eDouble, ".3");
argDescr->SetConstraint("be_negRows", new CArgAllow_Doubles(0, 1));
argDescr->AddDefaultKey("be_minScore", "integer", "if be_score = 'vote', score at or above which a row votes 'yes'", argDescr->eInteger, "0");
argDescr->SetConstraint("be_minScore", new CArgAllow_Integers(-100, 100));
argDescr->AddOptionalKey("be_eThresh", "double", "extension threshold:\n column score at or above which a column is added to a block\nif not given, assumes same values as 'be_sThresh', or if neither is present an error is reported\n** with 'vote' scoring, must be in [0, 1]", argDescr->eDouble);
// argDescr->SetConstraint("be_eThresh", new CArgAllow_Doubles(-1000, 1000));
argDescr->AddOptionalKey("be_sThresh", "double", "shrinkage threshold (ignored if -be_noShrink not specified):\n column score below which a column is removed from a block\nif not given, assumes same values as 'be_eThresh', or if neither is present an error is reported\n** with 'vote' scoring, must be in [0, 1]\n", argDescr->eDouble);
// argDescr->SetConstraint("be_sThresh", new CArgAllow_Doubles(-1000, 1000));
argDescr->SetCurrentGroup("");
// argDescr->AddFlag("qd", "shortened details report; only report details at end of cycles and trials. Has an affect only if '-details' is provided");
// Setup arg.descriptions for this application
SetupArgDescriptions(argDescr.release());
}
