def main():
parser = argparse.ArgumentParser(description='Searches for reactivity differences around given motifs')
parser.add_argument('control',type=str,help='control <.react> file')
parser.add_argument('experimental',type=str,help='experimental <.react> file')
parser.add_argument('fasta',type=str,help='<.fasta> to pull sequences from')
parser.add_argument('motif',type=str,help='Input file or motif')
parser.add_argument('-fp',default=5,type=int, help='[default = 5] Bases to include 5\' of the motif')
parser.add_argument('-tp',default=5,type=int, help='[default = 5] Bases to include 3\' of the motif')
parser.add_argument('-restrict',default = None, help = '<.txt > Limit analysis to these specific transcripts')
parser.add_argument('-outdir',type=str,default = 'motif_out',help='[default = motif_out] Out Directory')
parser.add_argument('-fastaout',action='store_true',default=False,help='Write windows to <.fasta> format as well')
parser.add_argument('-reactout',action='store_true',default=False,help='Write accompanying <.react> files as well')
args = parser.parse_args()
#Check if path exists
if not os.path.isdir(args.outdir):
os.mkdir(args.outdir)
#Read in reactivities and fasta
control,experimental = map(structure_io.read_react,[args.control,args.experimental])
sequences = structure_io.read_fasta(args.fasta)
#Apply filter if input
if args.restrict:
covered = structure_io.read_restrict(args.restrict)
sequences = {name:seq for name,seq in sequences.items() if name in covered}
#Read in motifs
motifs = read_motif(args.motif)
#Out Nomenclature
name_block_1 = [zz.replace('.react','') for zz in [args.control,args.experimental]]
name_block_2 = [str(qq)+q for qq,q in zip([args.fp,args.tp],['fp','tp'])]
#Iterate through motif(s)
for motif in motifs:
#Create a full MotifReport
report = MotifReport(motif,sequences,control,experimental,args.fp,args.tp)
#Generate name for each outfile
out_name = '_'.join(name_block_1+[report.motif]+name_block_2)+'.csv'
#Write out motif <.csv>
write_motif_csv(report,os.path.join(args.outdir,out_name))
#Write out motif <.fasta>
if args.fastaout:
out_fasta_name = out_name.replace('.csv','.fasta')
fasta_dict = {r.generate_name():r.generate_seq() for r in report.records.values()}
structure_io.write_fasta(fasta_dict,os.path.join(args.outdir,out_fasta_name))
#Write out motif <.react>
if args.reactout:
control_out = {c.generate_name():c.A_react() for c in report.records.values()}
exp_out = {e.generate_name():e.B_react() for e in report.records.values()}
control_new = '_'.join([args.control.replace('.react',''),motif,str(args.fp)+'fp',str(args.tp)+'tp'])+'.react'
exp_new = '_'.join([args.experimental.replace('.react',''),motif,str(args.fp)+'fp',str(args.tp)+'tp'])+'.react'
structure_io.write_react(control_out,os.path.join(args.outdir,control_new))
structure_io.write_react(exp_out,os.path.join(args.outdir,exp_new))
def main():
parser = argparse.ArgumentParser(description='Downscales <.rtsc> files.')
parser.add_argument('-f',
default=None,
nargs='+',
help='Specific <.rtsc> to operate on')
parser.add_argument('mode',
type=str.upper,
choices=['FRACTIONAL', 'RANDOMREAD', 'RANDOMPOSITION'])
parser.add_argument('-ratio',
type=float,
default=.50,
help='[default = 0.50] Fraction of RT stops to retain')
parser.add_argument(
'-restrict',
default=None,
help='Limit downscaling to these specific transcripts <.txt> ')
parser.add_argument('-sort',
action='store_true',
default=False,
help='Sort output by transcript name')
args = parser.parse_args()
#Files to operate on, dictionary of functions
fyle_lyst = sorted(glob.glob('*.rtsc')) if args.f == None else sorted(
args.f)
downscale_methods = {
'FRACTIONAL': keep_static_percentage,
'RANDOMPOSITION': keep_random_downsample,
'RANDOMREAD': random_per_stop
}
#Iterate through file(s)
for fyle in fyle_lyst:
#Read in the <.rtsc>, generate new name
data = read_rtsc(fyle)
new_fyle = fyle.replace(
'.rtsc',
'_' + args.mode + '_' + str(args.ratio).replace('.', '') + '.rtsc')
#You could be doing random for a long time if you do not filter by coverage.
if args.restrict != None:
covered = read_restrict(args.restrict)
data = {
name: stops
for name, stops in data.items() if name in covered
}
new_data = {
k: downscale_methods[args.mode](v, args.ratio)
for k, v in data.items()
}
write_rtsc(new_data, new_fyle, args.sort)
def main():
parser = argparse.ArgumentParser(
description=
'Finds reactivity differences in windows of a given size and composition'
)
parser.add_argument('control', type=str, help='control <.react> file')
parser.add_argument('experimental',
type=str,
help='experimental <.react> file')
parser.add_argument('fasta',
type=str,
help='<.fasta> to pull sequences from')
parser.add_argument('bases',
type=str.upper,
help="Bases to query (i.e.\'GC\' or \'AT\')")
parser.add_argument('-size',
default=8,
type=int,
help='[default = 8] Size of window')
parser.add_argument('-perc',
default=1.0,
type=float,
help='[default = 1.0] Percent specified bases')
parser.add_argument('-unique',
action='store_false',
default=True,
help='[default = True] Remove overlapping windows')
parser.add_argument('-fp',
default=5,
type=int,
help='[default = 5] Bases to include 5\' of the motif')
parser.add_argument('-tp',
default=5,
type=int,
help='[default = 5] Bases to include 3\' of the motif')
parser.add_argument(
'-restrict',
default=None,
help='<.txt > Limit analysis to these specific transcripts')
parser.add_argument('-fastaout',
action='store_true',
default=False,
help='Write windows to <.fasta> format as well')
parser.add_argument('-reactout',
action='store_true',
default=False,
help='Write accompanying <.react> files as well')
args = parser.parse_args()
#Read in reactivities and fasta
control, experimental = map(structure_io.read_react,
[args.control, args.experimental])
sequences = structure_io.read_fasta(args.fasta)
#Apply filter if input
if args.restrict:
covered = structure_io.read_restrict(args.restrict)
sequences = {
name: seq
for name, seq in sequences.items() if name in covered
}
#Out Nomenclature
name_block_1 = [
zz.replace('.react', '') for zz in [args.control, args.experimental]
]
name_block_2 = [
str(qq) + q for qq, q in zip([args.fp, args.tp], ['fp', 'tp'])
]
#Generate a full CompositionReport
report = CompositionReport(args.size, args.bases, args.perc, args.unique,
sequences, control, experimental, args.fp,
args.tp)
#Generate out name
detail = 'shared' if args.unique else 'unique'
parameters = [str(x) for x in [args.size, args.bases, args.perc, detail]]
out_name = '_'.join(name_block_1 + parameters + name_block_2) + '.csv'
#Write out motif <.csv>
write_comp_csv(report, out_name)
#Write out motif <.fasta>
if args.fastaout:
out_fasta_name = out_name.replace('.csv', '.fasta')
fasta_dict = {
r.generate_name(): r.generate_seq(True)
for r in report.records.values()
}
structure_io.write_fasta(fasta_dict, out_fasta_name)
if args.reactout:
control_out = {
c.generate_name(): c.A_react(True)
for c in report.records.values()
}
exp_out = {
e.generate_name(): e.B_react(True)
for e in report.records.values()
}
control_namae = [args.control.replace('.react', '')] + parameters + [
str(args.fp) + 'fp', str(args.tp) + 'tp'
]
control_new = '_'.join(control_namae) + '.react'
exp_namae = [args.experimental.replace('.react', '')] + parameters + [
str(args.fp) + 'fp', str(args.tp) + 'tp'
]
exp_new = '_'.join(exp_namae) + '.react'
structure_io.write_react(control_out, control_new)
structure_io.write_react(exp_out, exp_new)
def main():
parser = argparse.ArgumentParser(
description=
'Searches for reactivity differences around given multi-motifs')
parser.add_argument('control', type=str, help='control <.react> file')
parser.add_argument('experimental',
type=str,
help='experimental <.react> file')
parser.add_argument('fasta',
type=str,
help='<.fasta> to pull sequences from')
parser.add_argument('motifs', nargs='+', help='multi-motif components')
parser.add_argument(
'-unique',
action='store_false',
default=True,
help='[default = True] Remove overlapping multi-motifs')
parser.add_argument('-mn',
default=3,
type=int,
help='[default = 3] Number of multi-motifs required')
parser.add_argument(
'-mw',
default=50,
type=int,
help='[default = 50] Query window size for multi-motifs')
parser.add_argument('-FP',
default=30,
type=int,
help='[default = 30] Bases 5\' of multi-motifs')
parser.add_argument('-TP',
default=30,
type=int,
help='[default = 30] Bases 3\' of multi-motifs')
parser.add_argument(
'-restrict',
default=None,
help='<.txt > Limit analysis to these specific transcripts')
parser.add_argument('-fastaout',
action='store_true',
default=False,
help='Write windows to <.fasta> format as well')
parser.add_argument('-reactout',
action='store_true',
default=False,
help='Write accompanying <.react> files as well')
args = parser.parse_args()
#Read in reactivities and fasta
control, experimental = map(structure_io.read_react,
[args.control, args.experimental])
sequences = structure_io.read_fasta(args.fasta)
#Apply filter if input
if args.restrict:
covered = structure_io.read_restrict(args.restrict)
sequences = {
name: seq
for name, seq in sequences.items() if name in covered
}
#Out Nomenclature
name_block_1 = [
zz.replace('.react', '') for zz in [args.control, args.experimental]
]
name_block_2 = [
str(qq) + q for qq, q in zip([args.mn, args.mw, args.FP, args.TP],
['mn', 'mw', 'FP', 'TP'])
]
#Generate base out name
detail = 'shared' if args.unique else 'unique'
out_name = '_'.join(name_block_1 + name_block_2 + [detail] +
args.motifs) + '.csv'
#Generate a full MetaReport
report = MetaMotifReport(args.motifs, args.mn, args.mw, args.FP, args.TP,
args.unique, sequences, control, experimental)
#Write out motif <.csv>
write_meta_csv(report, out_name)
#Write out motif <.fasta>
if args.fastaout:
out_fasta_name = out_name.replace('.csv', '.fasta')
fasta_dict = {
r.generate_name(): r.generate_seq()
for r in report.records.values()
}
structure_io.write_fasta(fasta_dict, out_fasta_name)
if args.reactout:
control_out = {
c.generate_name(): c.A_react()
for c in report.records.values()
}
exp_out = {
e.generate_name(): e.B_react()
for e in report.records.values()
}
#Writeout Control
control_namae = [args.control.replace('.react', '')
] + name_block_2 + [detail] + args.motifs
control_new = '_'.join(control_namae) + '.react'
structure_io.write_react(control_out, control_new)
#Writeout Experimental
exp_namae = [args.experimental.replace('.react', '')
] + name_block_2 + [detail] + args.motifs
exp_new = '_'.join(exp_namae) + '.react'
structure_io.write_react(exp_out, exp_new)