func init() {
runtime.GOMAXPROCS(runtime.NumCPU())
util.FlagUse("cpu")
util.FlagParse("database-path frag-lib-dir query-pdb-file "+
"[query-pdb-file ...]", "")
util.AssertLeastNArg(2)
}
func init() {
flag.StringVar(&flagChain, "chain", flagChain,
"When set, only this chain will be tested for a correspondence. "+
"Otherwise, all chains will be tested.")
util.FlagParse("pdb-file", "")
util.AssertNArg(1)
}
func init() {
flag.BoolVar(&flagQuiet, "quiet", flagQuiet,
"When set, hhblits/hhmake output will be hidden.")
util.FlagUse("seq-db")
util.FlagParse("in-fasta-file out-hhm-file", "")
util.AssertNArg(2)
}
func init() {
util.FlagParse(
"out-file matrix-file [ matrix-file ... ]",
"Combines the input matrices into a 'best-of-all' matrix by\n"+
"using the best (lowest) SAS score for each pair. All given\n"+
"matrices must be exactly the same size.")
util.AssertLeastNArg(2)
}
func init() {
flag.Float64Var(&flagRmsd, "rmsd", flagRmsd,
"The RMSD cut-off to use to determine true positives.")
util.FlagUse("pdb-dir")
util.FlagParse("fmap-file", "")
util.AssertNArg(1)
}
func init() {
util.FlagUse("cpu")
util.FlagParse("frag-lib-dir chain pdb-file out-bow",
"Computes and outputs a BOW file for the specified chain in the\n"+
"given PDB file. If 'out-bow' is '--', then a human readable\n"+
"version of the BOW will be printed to stdout instead.")
util.AssertNArg(4)
}
func init() {
flag.IntVar(&flagRandoms, "n", flagRandoms,
"The number of random permutations to use.")
util.FlagUse("cpu", "cpuprof")
util.FlagParse(
"out-file vectors-csv 'comma-sep-cath-domains' ...",
"Computes the probability that each cluster of cath domains is\n"+
"significantly different with respect to their Fragbag vectors.")
util.AssertLeastNArg(3)
}
func init() {
flag.IntVar(&flagNum, "n", flagNum, "The number of PDB entries to echo.")
flag.BoolVar(&flagPaths, "paths", flagPaths,
"When set, full file paths will be echoed instead of PDB ids.")
util.FlagUse("pdb-dir")
util.FlagParse("", "")
rand.Seed(time.Now().UnixNano())
}
func init() {
flag.BoolVar(&flagPaths, "paths", flagPaths,
"When set, the full path of each PDB chain identifier will be\n"+
"displayed, based on the value of the PDB_PATH environment\n"+
"variable.")
util.FlagParse("pdb-select-file",
"Given a file in the PDB Select format, output a list of PDB chain "+
"identifiers (one per line).")
util.AssertNArg(1)
}
func main() {
flag.BoolVar(&flagAllFragments, "all-fragments", flagAllFragments,
"When set, all fragments will be shown, even if the best fragment\n"+
"of each residue set is the same.")
util.FlagParse(
"fraglib align.{fasta,ali,a2m,a3m} out-csv",
"Writes a CSV file to out-csv containing the best matching fragment\n"+
"for each pairwise contiguous set of residues between the\n"+
"first two proteins in the alignment.")
util.AssertNArg(3)
flib := util.SequenceLibrary(util.Arg(0))
aligned := util.MSA(util.Arg(1))
outcsv := util.CreateFile(util.Arg(2))
csvWriter := csv.NewWriter(outcsv)
csvWriter.Comma = '\t'
defer csvWriter.Flush()
pf := func(record ...string) {
util.Assert(csvWriter.Write(record), "Problem writing to '%s'", outcsv)
}
pf("start1", "end1", "start2", "end2", "frag1", "frag2", "rat1", "rat2")
iter := newContiguous(
flib.FragmentSize(), aligned.GetFasta(0), aligned.GetFasta(1))
for iter.next() {
best1 := flib.BestSequenceFragment(iter.res1)
best2 := flib.BestSequenceFragment(iter.res2)
if !flagAllFragments && best1 == best2 {
continue
}
if best1 == -1 || best2 == -1 {
continue
}
p1 := flib.AlignmentProb(best1, iter.res1)
p2 := flib.AlignmentProb(best2, iter.res2)
if p1.Distance(p2) > 0.14 {
continue
}
pf(
fmt.Sprintf("%d", iter.s1()),
fmt.Sprintf("%d", iter.e1()),
fmt.Sprintf("%d", iter.s2()),
fmt.Sprintf("%d", iter.e2()),
fmt.Sprintf("%d", best1),
fmt.Sprintf("%d", best2),
fmt.Sprintf("%f", p1),
fmt.Sprintf("%f", p2),
)
}
}
func init() {
flag.StringVar(&flagFragbag, "fragbag", "fragbag",
"The old fragbag executable.")
flag.BoolVar(&flagOldStyle, "oldstyle", false,
"When true, PDB chains will be concatenated together as if they were "+
"one chain to compute a BOW vector.")
util.FlagParse(
"library-file brk-file pdb-file [pdb-file ...]",
"Note that if the old library and the new library don't have the\n"+
"same number of fragments and the same fragment size, bad things\n"+
"will happen.\n")
util.AssertLeastNArg(3)
}
func init() {
flag.StringVar(&flagInFmt, "infmt", flagInFmt,
"Force the format of the input file. Legal values are fasta, "+
"stockholm, a2m and a3m.")
flag.StringVar(&flagOutFmt, "outfmt", flagOutFmt,
"Force the format of the output file. Legal values are fasta, "+
"stockholm, a2m and a3m.")
util.FlagParse("in-msa out-msa",
"Convert the format of an MSA file from 'in-msa' to 'out-msa'.\n"+
"The formats are auto detected from the file's extension, but\n"+
"they may be forced with the 'infmt' and 'outfmt' flags.")
util.AssertNArg(2)
}
func init() {
flag.StringVar(&flagChain, "chain", flagChain,
"This may be set to one or more chain identifiers. Only amino acids "+
"belonging to a chain specified will be included.")
flag.StringVar(&flagSplit, "split", flagSplit,
"When set, each FASTA entry produced will be written to a file in the "+
"specified directory with the PDB id code and chain identifier as "+
"the name.")
util.FlagParse("in-pdb-file [out-fasta-file]", "")
if util.NArg() != 1 && util.NArg() != 2 {
util.Usage()
}
}
func main() {
flag.BoolVar(&flagAllFragments, "all-fragments", flagAllFragments,
"When set, all fragments will be shown, even if the best fragment\n"+
"of each ATOM set is the same.")
util.FlagParse(
"fraglib align.{fasta,ali,a2m,a3m} pdb-file out-csv",
"Writes a CSV file to out-csv containing the best matching fragment\n"+
"for each pairwise contiguous set of alpha-carbon atoms of the\n"+
"first two proteins in the alignment and PDB file.")
util.AssertNArg(4)
flib := util.StructureLibrary(util.Arg(0))
aligned := util.MSA(util.Arg(1))
pentry := util.PDBRead(util.Arg(2))
outcsv := util.CreateFile(util.Arg(3))
csvWriter := csv.NewWriter(outcsv)
csvWriter.Comma = '\t'
defer csvWriter.Flush()
pf := func(record ...string) {
util.Assert(csvWriter.Write(record), "Problem writing to '%s'", outcsv)
}
pf("start1", "end1", "start2", "end2", "frag1", "frag2", "frag_rmsd")
iter := newContiguous(
flib.FragmentSize(),
aligned.GetFasta(0), aligned.GetFasta(1),
pentry.Chains[0], pentry.Chains[1])
for iter.next() {
best1 := flib.BestStructureFragment(iter.atoms1)
best2 := flib.BestStructureFragment(iter.atoms2)
if !flagAllFragments && best1 == best2 {
continue
}
bestRmsd := structure.RMSD(flib.Atoms(best1), flib.Atoms(best2))
pf(
fmt.Sprintf("%d", iter.s1()),
fmt.Sprintf("%d", iter.e1()),
fmt.Sprintf("%d", iter.s2()),
fmt.Sprintf("%d", iter.e2()),
fmt.Sprintf("%d", best1),
fmt.Sprintf("%d", best2),
fmt.Sprintf("%f", bestRmsd),
)
}
}
func init() {
flag.Float64Var(&flagThreshold, "threshold", flagThreshold,
"The threshold at which to cut the tree.")
flag.StringVar(&flagGobIt, "gobit", flagGobIt,
"If set, alignment distances will be cached to the file given, "+
"then mattbench-cluster will quit.")
util.FlagUse("cpu", "cpuprof", "verbose")
util.FlagParse(
"(astral-alignment-dir | alignment-distances-gob) dendrogram-tree "+
"out-clusters.csv",
"Where `dendrogram-tree` is a file in Newick tree format.")
if len(flagGobIt) > 0 {
util.AssertNArg(1)
} else {
util.AssertNArg(3)
}
}
func init() {
flag.Float64Var(&flagThreshold, "threshold", flagThreshold,
"Set the distance threshold to use when computing AUC.")
util.FlagUse("cpu", "cpuprof")
util.FlagParse(
"cath-domain-labels best-of-all-matrix"+
"(bowdb | matrix-file) out-file "+
"[ (bowdb | matrix-file) out-file ... ]",
"Computes the AUC of each aligner matrix (or BOW database) given\n"+
"with respect to the 'best-of-all' matrix given. Each AUC is\n"+
"written to a separate out-file. The sizes of all matrices must\n"+
"be exactly equivalent.\n"+
"Files are interpreted as BOW databases if they have a '.bowdb'\n"+
"file extension.")
util.AssertLeastNArg(4)
if util.NArg()%2 != 0 {
util.Fatalf("There must be an out file for each matrix or bowdb file.")
}
}
func init() {
u := "pdb-file chain-id start stop pdb-file chain-id start stop"
util.FlagParse(u, "")
util.AssertNArg(8)
}
func init() {
util.FlagUse("cpu", "seq-db", "pdb-hhm-db", "blits",
"hhfrag-min", "hhfrag-max", "hhfrag-inc")
util.FlagParse("target-fasta out-fmap", "")
util.AssertNArg(2)
}
func init() {
util.FlagParse("fasta-file",
"Quickly count the number of sequences in a fasta file.")
util.AssertNArg(1)
}
func init() {
util.FlagParse("fmap-file", "")
util.AssertNArg(1)
}
func init() {
u := "fraglib pdb-file [ chain-id [ start stop ] ]"
util.FlagParse(u, "")
util.AssertLeastNArg(2)
}
func init() {
util.FlagParse("bow1 bow2", "")
util.AssertNArg(2)
}
func init() {
util.FlagParse("fasta-file out-dir",
"Split a single FASTA file into a set of files for each sequence.")
util.AssertNArg(2)
}
func init() {
util.FlagParse("frag-lib-path output-file", "")
util.AssertNArg(2)
}
func init() {
util.FlagUse("cpu", "seq-db", "pdb-hhm-db", "blits", "verbose",
"hhfrag-min", "hhfrag-max", "hhfrag-inc")
util.FlagParse("out-dir target-fasta", "")
util.AssertLeastNArg(2)
}
func init() {
util.FlagUse("cpu")
util.FlagParse("frag-lib-dir fmap-file out-bow", "")
util.AssertNArg(3)
}
