def chunk_iterator(itr, fn, n=100, show_progress=True, label=None):
'''Breaks an iterable into chunks and applies a function to each chunk.
Arguments:
- itr the iterable to be chunked
- fn the function to be applied to each chunks
- n the size of each chunk
- show_progress show a progress bar
- label the label to show on the progress bar
'''
if len(itr) == 0:
return
label = "" if label is None else label
if len(itr)/n <= 1:
show_progress = False
swga.message(label)
chunked_itr = chunks(itr, n)
if show_progress:
chunked = progress.bar(
chunked_itr,
label=label,
expected_size=max(len(itr)/n, 1))
else:
chunked = chunked_itr
for chunk in chunked:
fn(chunk)
def main(argv, cfg_file):
cmd = Command('find_sets', cfg_file=cfg_file)
score_cmd = Command('score', cfg_file=cfg_file)
cmd.parse_args(argv)
score_cmd.parse_args(argv)
init_db(cmd.primer_db)
# We need to clear all the previously-used sets each time due to uniqueness
# constraints
allsets = Set.select()
if allsets.count() > 0:
if not cmd.force:
click.confirm("Remove all previously-found sets?", abort=True)
for s in progress.bar(allsets, expected_size=allsets.count()):
s.primers.clear()
s.delete_instance()
make_graph(cmd.max_dimer_bp, graph_fname)
swga.message("Now finding sets. If nothing appears, try relaxing your parameters.")
if cmd.workers <= 1:
setlines = setfinder.find_sets(
cmd.min_bg_bind_dist,
cmd.min_size,
cmd.max_size,
cmd.bg_genome_len,
graph_fp=graph_fname)
else:
setlines = setfinder.mp_find_sets(
nprocesses=cmd.workers,
graph_fp=graph_fname,
min_bg_bind_dist=cmd.min_bg_bind_dist,
min_size=cmd.min_size,
max_size=cmd.max_size,
bg_genome_len=cmd.bg_genome_len)
score_sets(
setlines,
cmd.fg_genome_fp,
score_cmd.score_expression,
cmd.max_fg_bind_dist,
cmd.max_sets)