def __init__(self, name, data, workdir, mapfile=None):
# i/o paths
self.workdir = workdir
self._datafile = data
self._mapfile = mapfile
self._check_files()
self.data = np.zeros()
self.maparr = np.zeros()
# init default param settings
self.params = Params()
self.popdict = {}
self.mindict = {}
self.npops = len(self.popdict)
self.nboots = 100
# results dataframes
self.results = Params()
# pairwise Fst between all populations
npops = len(self.popdict)
arrfst = np.zeros((npops, npops), dtype=np.uint64)
self.results.fst = pd.DataFrame(arrfst)
# individual pi
nsamples = len(list(chain(*self.popdict.values())))
arrpi = np.zeros(nsamples, dtype=np.uint64)
self.results.pi = pd.DataFrame(arrpi)
# population thetas
npops = len(self.popdict)
arrtheta = np.zeros(npops, dtype=np.uint64)
self.results.theta = pd.DataFrame(arrtheta)
# parse samples from the data file
self._check_files()
def __init__(
self,
data=None,
imap=None,
minmap=1,
newick=None,
nboots=1000,
):
# check imports
if not sys.modules.get("toytree"):
raise ImportError(_TOYTREE_IMPORT)
# parse data as (1) path to data file, or (2) ndarray
if isinstance(data, str):
self.data = os.path.realpath(os.path.expanduser(data))
else:
self.data = data
# check dtype of newick/tree entry
self.newick = newick
if isinstance(newick, toytree.Toytree.ToyTree):
self.newick = newick.newick
# store tests
self.imap = imap
self.minmap = minmap
# parameters
self.params = Params()
self.params.nboots = nboots
self.params.quiet = False
self.params.database = None
# results storage
self.results_table = None
self.results_boots = None
# cluster attributes
self.ipcluster = {
"cluster_id": "",
"profile": "default",
"engines": "Local",
"quiet": 0,
"timeout": 60,
"cores": 0,
"threads": 2,
"pids": {},
}
def __init__(
self,
data,
name="test",
workdir="analysis-treemix",
imap=None,
minmap=None,
seed=None,
quiet=False,
raise_root_error=False,
binary=None,
*args,
**kwargs):
# path attributes
self.name = name
self.data = data
# if not imap then it will be set to 1
self.minmap = minmap
self.imap = imap
# others
self.binary = os.path.join(sys.prefix, "bin", "treemix")
self.binary = (binary if binary else self.binary)
self.raise_root_error = raise_root_error
self._find_binary()
# params dict
self.params = Params()
self.params.k = 0
self.params.m = 0
self.params.g = (None, None)
self.params.bootstrap = 0
self.params.cormig = 0
self.params.climb = 0
self.params.noss = 0
self.params.seed = (seed if seed else np.random.randint(0, int(1e9)))
self.params.root = None
self.params.se = 0
self.params.global_ = 0
# get snps and snpmap
ext = SNPsExtracter(self.data, self.imap, self.minmap, quiet=quiet)
ext.parse_genos_from_hdf5()
self.snps = ext.subsample_snps(seed)
self.names = ext.names
self.nsites = self.snps.shape[1]
self.sidxs = {
i: [self.names.index(i) for i in self.imap[i]]
for i in self.imap
}
# make workdir if it does not exist
if workdir:
self.workdir = os.path.abspath(os.path.expanduser(workdir))
else:
self.workdir = os.path.join(
os.path.abspath(os.path.curdir),
"analysis-treemix",
)
if not os.path.exists(self.workdir):
os.makedirs(self.workdir)
## set params
notparams = set(
["workdir", "name", "data", "minmap", "imap", "seed", "quiet"]
)
for key in set(kwargs.keys()) - notparams:
self.params[key] = kwargs[key]
# results files
self.files = Params()
self.files.tree = os.path.join(self.workdir, self.name + ".treeout.gz")
self.files.cov = os.path.join(self.workdir, self.name + ".cov.gz")
self.files.llik = os.path.join(self.workdir, self.name + ".llik")
# results
self.results = Params()
self.results.tree = ""
self.results.admixture = []
self.results.cov = []
self.results.llik = None
def __init__(self, name, data, workdir=None, **kwargs):
## store attributes
self.name = name
self.data = data
self._kwargs = {
"minsnps": 0,
"maxloci": None,
"seed": None,
"mb_mcmc_ngen": int(1e6),
"mb_mcmc_burnin": int(1e5),
"mb_mcmc_sample_freq": int(1e3),
"bucky_alpha": [0.1, 1.0, 10.0],
"bucky_nchains": 4,
"bucky_nreps": 4,
"bucky_niter": int(1e6),
"copied": False,
"samples": None,
}
self._kwargs.update(kwargs)
## check binaries
self.check_binaries()
## check workdir
if workdir:
self.workdir = os.path.realpath(
os.path.abspath(os.path.expanduser(workdir)))
else:
self.workdir = os.path.realpath(
os.path.join(os.path.curdir, "analysis-bucky"))
if not os.path.exists(self.workdir):
os.makedirs(self.workdir)
## set bucky parameters with defaults
self.params = Params()
notparams = set(["workdir", "samples", "copied"])
for key in set(self._kwargs.keys()) - notparams:
self.params[key] = self._kwargs[key]
## set defaults
self.params.seed = np.random.randint(1e9)
if self._kwargs["seed"]:
self.params.seed = self._kwargs["seed"]
self.samples = []
if self._kwargs["samples"]:
self.samples = set(self._kwargs["samples"])
self._alleles = 0
if ".alleles" in self.data:
self._alleles = 1
## results files
self.files = Params()
self.files.data = data
self.files.nexfiles = []
self.files.sumfiles = []
self.files.buckyfiles = []
## accessible results
## get from property functions
self.results = Params()
self.results.concordance_trees = Params()
self.results.population_trees = Params()
self.results.concordance_factors = Params()
def __init__(self,
data,
name="test",
workdir="analysis-raxml",
*args,
**kwargs):
## path attributes
self._kwargs = {
"f": "a",
"T": 4, # <- change to zero !?
"m": "GTRGAMMA",
"N": 100,
"x": 12345,
"p": 54321,
"o": None,
"binary": "",
}
# update kwargs for user args and drop key if value is None
self._kwargs.update(kwargs)
self._kwargs = {
i: j
for (i, j) in self._kwargs.items() if j is not None
}
# check workdir
if workdir:
workdir = os.path.abspath(os.path.expanduser(workdir))
else:
workdir = os.path.abspath(os.path.curdir)
if not os.path.exists(workdir):
os.makedirs(workdir)
## entered args
self.params = Params()
self.params.n = name
self.params.w = workdir
self.params.s = os.path.abspath(os.path.expanduser(data))
## find the binary
if not self._kwargs["binary"]:
self.params.binary = _find_binary()
## set params
notparams = set(["workdir", "name", "data", "binary"])
for key in set(self._kwargs.keys()) - notparams:
self.params[key] = self._kwargs[key]
## check binary
self._get_binary()
## attributesx
self.rasync = None
self.stdout = None
self.stderr = None
## results files
self.trees = Params()
self.trees.bestTree = OPJ(workdir, "RAxML_bestTree." + name)
self.trees.bipartitionsBranchLabels = OPJ(
workdir, "RAxML_bipartitionsBranchLabels." + name)
self.trees.bipartitions = OPJ(workdir, "RAxML_bipartitions." + name)
self.trees.bootstrap = OPJ(workdir, "RAxML_bootstrap." + name)
self.trees.info = OPJ(workdir, "RAxML_info." + name)
def __init__(self,
data,
name="test",
workdir="analysis-fasttree",
*args,
**kwargs):
## path attributes
self._kwargs = {
"T": 1,
"t": "nt",
"m": "gtr",
"binary": "",
"advanced": "",
"gamma": False,
"overwrite": False,
}
# update kwargs for user args and drop key if value is None
self._kwargs.update(kwargs)
self._kwargs = {
i: j
for (i, j) in self._kwargs.items() if j is not None
}
# check workdir
if workdir:
workdir = os.path.abspath(os.path.expanduser(workdir))
else:
workdir = os.path.abspath(os.path.curdir)
if not os.path.exists(workdir):
os.makedirs(workdir)
## entered args
self.params = Params()
self.params.n = name
self.params.w = workdir
self.params.s = os.path.abspath(os.path.expanduser(data))
self.params.f = OPJ(workdir, name + ".fasta")
## find the binary
if not self._kwargs["binary"]:
self.params.binary = _find_binary(self._kwargs["T"])
## set params
notparams = set(["workdir", "name", "data", "binary"])
for key in set(self._kwargs.keys()) - notparams:
self.params[key] = self._kwargs[key]
## check binary
self._get_binary()
## attributesx
self.rasync = None
self.stdout = None
self.stderr = None
## results files
# self.trees = Params()
self.tree = OPJ(workdir, "FastTree_Tree." + name)
self.log = OPJ(workdir, "FastTree." + name + ".log")
# Convert phylip into fasta cause fasttree only read interleaved phylyp (or fasta) and not sequential phylip
if os.path.exists(str(self.params.f)) and not self.params.overwrite:
print("Fasta file already exist in: {}".format(str(self.params.f)),
end='\n')
else:
open(str(self.params.f), 'w').close(
) #clean fasta file in case it exist (useful for overwrite parameter)
print("Converting phylip file to fasta...", end='\r')
with open(str(self.params.s),
'r') as f: #open phy file to convert in fasta file
for nline, line in enumerate(f, start=1): #move line by line
if nline != 1: #skip first line
with open(str(self.params.f), 'a') as fw:
fw.write(
">" + line.split()[0] + "\n" +
line.split()[1] + "\n"
) #add > at the beggining and split line by spaces, first is name and second is sequence (If name contain space can be a problem)
print("Temporal fasta file saved in: {}".format(str(
self.params.f)),
end='\n')
# Set enviromental variable to control OpenMP (OMP_NUM_THREADS)
if self._kwargs["T"] > 1:
os.environ["OMP_NUM_THREADS"] = str(self._kwargs["T"])
print(
"FastTree will use {} threads, be sure have installed OpenMP".
format(self._kwargs["T"]),
end='\n')