def ll2s(x):
""" converts a list of lists, e.g. [['guid1',2],['guid2',0]] into a set {'guid1','guid2'} """
neighbour_set = set()
for neighbour in x:
neighbour_set.add(neighbour[0])
return neighbour_set
# define directory where the fastas are
fastadir = os.path.join('..', 'demos', 'AA041', 'fasta')
outputdir = os.path.join('..', 'demos', 'AA041', 'output')
# instantiate client
fn3c = fn3Client(
) # expects operation on local host; pass baseurl if somewhere else.
# names of the clustering algorithms
clusters = fn3c.clustering()
existing_guids = set(fn3c.guids())
clustering_created = False
print("There are {0} existing guids".format(len(existing_guids)))
# add control fasta files. The system evaluates the %N in terms of the population existing
# we load 50 randomly selected guids as controls
for i, fastafile in enumerate(
glob.glob(os.path.join(fastadir, 'control', '*.fasta'))):
guid = "ctrl_" + os.path.basename(fastafile).replace('.fasta', '')
seq = fn3c.read_fasta_file(fastafile)['seq']
if not guid in existing_guids:
p.mkdir(parents=True, exist_ok=True)
p = pathlib.Path(inputdir)
p.mkdir(parents=True, exist_ok=True)
# determine input files
inputfiles = glob.glob(os.path.join(inputdir, '*.fasta'))
random.shuffle(inputfiles) # read them in order
if len(inputfiles) < max_sequences:
raise ValueError(
"Asked to add {0} sequences, but only {1} are available in the input directory {2}"
.format(max_sequences, len(inputfiles), inputdir))
else:
inputfiles = inputfiles[0:max_sequences]
print("opening connection to fn3 server")
fn3c = fn3Client(baseurl="http://127.0.0.1:5020")
# determine all masked positions
excluded_positions = fn3c.nucleotides_excluded()
# determine how many samples there are currently in the server.
nSamples = len(fn3c.guids())
print("There are {0} existing samples. Adding more ..".format(nSamples))
# create output file with header line
outputfile = os.path.join(outputdir, 'timings_{0}.tsv'.format(nSamples))
nAdded_this_batch = 0
with open(outputfile, 'w+t') as f:
output_line = "{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\n".format(
'nSamples', 's_insert', 'e_insert', 'd_insert', 's_read', 'e_read',
'd_read')
Example usage:
# first, a server must be running
python findNeighbour3-server.py ../demos/simulation/config/config.json
# then simulations must be generated (e.g. with run_simulation)
python run_simulation.py ../output/simulation_set_1""")
parser.add_argument('inputdir',
type=str,
nargs=1,
help='data will be read from the inputdir')
args = parser.parse_args()
basedir = os.path.abspath(args.inputdir[0])
# connect to server
fn3c = fn3Client("http://localhost:5020")
# iterate over simulated data
for inputdir in glob.glob(os.path.join(basedir, '*')):
print(inputdir)
# define filenames
fasta_filename = os.path.join(inputdir, 'phylogeny.fasta')
sequence_filename = os.path.join(inputdir, 'phylogeny.txt')
observed_filename = os.path.join(inputdir, 'observed.txt')
tree_filename = os.path.join(inputdir, 'phylogeny.nwk')
ref_filename = os.path.join(inputdir, 'reference.fasta')
treepic_filename = os.path.join(inputdir,
'{0}.png'.format('tree_image'))
annotated_treepic_filename = os.path.join(
inputdir, '{0}.png'.format('annotated_tree_image'))
def __init__(self):
self.fn3c = fn3Client() # expect success