def load_trjs_or_features(args):
if args.features:
lengths, data = load_features(args.features, stride=args.subsample)
else:
assert args.trajectories
assert len(args.trajectories) == len(args.topologies)
targets = {
os.path.basename(topf): "%s files" % len(trjfs)
for topf, trjfs in zip(args.topologies, args.trajectories)
}
logger.info("Beginning clustering; targets:\n%s",
json.dumps(targets, indent=4))
with timed("Loading trajectories took %.1f s.", logger.info):
lengths, xyz, select_top = load_trajectories(
args.topologies,
args.trajectories,
selections=args.atoms,
stride=args.subsample,
processes=auto_nprocs())
logger.info("Clustering using %s atoms matching '%s'.", xyz.shape[1],
args.atoms)
# md.rmsd requires an md.Trajectory object, so wrap `xyz` in
# the topology.
data = md.Trajectory(xyz=xyz, topology=select_top)
return lengths, data
def load_trajectories(topologies, trajectories, selections, stride, processes):
for top, selection in zip(topologies, selections):
sentinel_trj = md.load(top)
try:
# noop, but causes fast-fail w/bad args.atoms
sentinel_trj.top.select(selection)
except:
raise exception.ImproperlyConfigured(
("The provided selection '{s}' didn't match the topology "
"file, {t}").format(s=selection, t=top))
flat_trjs = []
configs = []
n_inds = None
for topfile, trjset, selection in zip(topologies, trajectories,
selections):
top = md.load(topfile).top
indices = top.select(selection)
if n_inds is not None:
if n_inds != len(indices):
raise exception.ImproperlyConfigured(
("Selection on topology %s selected %s atoms, but "
"other selections selected %s atoms.") %
(topfile, len(indices), n_inds))
n_inds = len(indices)
for trj in trjset:
flat_trjs.append(trj)
configs.append({
'top': top,
'stride': stride,
'atom_indices': indices,
})
logger.info(
"Loading %s trajectories with %s atoms using %s processes "
"(subsampling %s)", len(flat_trjs), len(top.select(selection)),
processes, stride)
assert len(top.select(selection)) > 0, "No atoms selected for clustering"
with timed("Loading took %.1f sec", logger.info):
lengths, xyz = mpi.io.load_trajectory_as_striped(
flat_trjs, args=configs, processes=auto_nprocs())
with timed("Turned over array in %.2f min", logger.info):
tmp_xyz = xyz.copy()
del xyz
xyz = tmp_xyz
logger.info("Loaded %s frames.", len(xyz))
return lengths, xyz, top.subset(top.select(selection))
def process_command_line(argv):
'''Parse the command line and do a first-pass on processing them into a
format appropriate for the rest of the script.'''
parser = argparse.ArgumentParser(formatter_class=argparse.
ArgumentDefaultsHelpFormatter)
parser.add_argument(
"--assignments", required=True,
help="File containing assignments to states.")
parser.add_argument(
"--n-eigenvalues", default=5, type=int,
help="Number of eigenvalues to compute for each lag time.")
parser.add_argument(
"--lag-times", default="5:100:2",
help="List of lagtimes (in frames) to compute eigenspectra for. "
"Format is min:max:step.")
parser.add_argument(
"--symmetrization", default="transpose",
choices=['transpose', 'row_normalize', 'prior_counts'],
help="The method to use to fit transition probabilities from "
"the transition counts matrix.")
parser.add_argument(
"--trj-ids", default=None,
help="Computed the implied timescales for only the given "
"trajectory ids. This is useful for handling assignments "
"for shared state space clusterings.")
parser.add_argument(
"--processes", default=max(1, auto_nprocs()/4), type=int,
help="Number of processes to use. Because eigenvector "
"decompositions are thread-parallelized, this should "
"usually be several times smaller than the number of "
"cores availiable on your machine.")
parser.add_argument(
"--trim", default=False, action="store_true",
help="Turn ergodic trimming on.")
parser.add_argument(
"--timestep", default=None, type=float,
help='A conversion between frames and nanoseconds (i.e. frames '
'per nanosecond) to scale the axes to physical units '
'(rather than frames).')
parser.add_argument(
"--infer-timestep", default=None,
help="An example trajectory from which to infer the conversion "
"from frame to nanoseconds.")
parser.add_argument(
"--plot", default=None,
help="Path for the implied timescales plot.")
parser.add_argument(
"--logscale", action='store_true',
help="Flag to output y-axis log scale plot.")
args = parser.parse_args(argv[1:])
args.lag_times = range(*map(int, args.lag_times.split(':')))
if args.trj_ids is not None:
args.trj_ids = slice(*map(int, args.trj_ids.split(':')))
if args.symmetrization == 'prior_counts':
args.symmetrization = prior_counts
else:
args.symmetrization = getattr(builders, args.symmetrization)
return args
def process_command_line(argv):
'''Parse the command line and do a first-pass on processing them into a
format appropriate for the rest of the script.'''
parser = argparse.ArgumentParser(
formatter_class=argparse.RawDescriptionHelpFormatter,
description="Compute CARDS matricies for a set of trajectories "
"and save all matrices and dihedral mappings.\n \n"
"Please cite the following papers if you use CARDS with enspara:\n"
"[1] Singh, S. and Bowman, G.R.\n"
" Journal of Chemical Theory and Computation\n"
" 2017 13 (4), 1509-1517\n"
" DOI: 10.1021/acs.jctc.6b01181\n"
"\n"
"[2] Porter,J.R., Zimmerman, M.I., and Bowman G.R.\n"
" bioRxiv 431072; doi: https://doi.org/10.1101/431072\n")
# INPUTS
input_args = parser.add_argument_group("Input Settings")
#input_data_group = parser.add_mutually_exclusive_group(required=True)
input_args.add_argument(
'--trajectories',
required=True,
nargs="+",
action='append',
help="List of paths to aligned trajectory files to cluster. "
"All file types that MDTraj supports are supported here.")
input_args.add_argument('--topology',
required=True,
action='append',
help="The topology file for the trajectories.")
# PARAMETERS
cards_args = parser.add_argument_group("CARDS Settings")
cards_args.add_argument(
'--buffer-size',
default=15,
type=int,
help="Size of buffer zone between rotameric states, in degrees.")
cards_args.add_argument("--processes",
default=max(1,
auto_nprocs() / 4),
type=int,
help="Number of processes to use.")
# OUTPUT
output_args = parser.add_argument_group("Output Settings")
output_args.add_argument(
'--matrices',
required=True,
action=readable_dir,
help="The folder location to write the four CARDS matrices (as pickle)."
)
output_args.add_argument(
'--indices',
required=True,
action=readable_dir,
help="The location to write the dihedral indices file (as CSV).")
args = parser.parse_args(argv[1:])
# CARDS FEATURES
if not (0 < args.buffer_size < 360):
raise exception.ImproperlyConfigured(
"The given buffer size (%s) is not possible." % args.buffer_size)
return args
