'''
Created on Jun 27, 2014
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/AND_TCR/Analysis/Chips1_2_3/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
datasets = []
for peptide in ('K99A', 'NoPep', 'PCC'):
pep_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(peptide, ab))
filename = yzer.get_filename(
pep_dirpath, '{}_{}_enhancers.txt'.format(peptide, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
# Only in all three reps
#data = data[(data['p1_id'] > 0) & (data['p2_id'] > 0) & (data['p3_id'] > 0)]
datasets.append(data)
# ONLY this pep:
only = sum((data['id(2)'] == 0) & (data['id(3)'] == 0))
'''
Created on Mar 28, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/TReg_enhancers/2013_03_19'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs')
for antibody in ('me2', 'ac'):
treg = yzer.import_file(
yzer.get_filename(
dirpath, 'treg_with_naive_{0}.txt'.format(antibody))).fillna(0)
naive = yzer.import_file(
yzer.get_filename(
dirpath, 'naive_with_treg_{0}.txt'.format(antibody))).fillna(0)
# Filter out promoters
treg = treg[treg['tss_id'] == 0]
naive = naive[naive['tss_id'] == 0]
# Get venn-diagram sets for foxp3/me2
only_treg = treg[treg['naive_id'] == 0]
only_naive = naive[naive['treg_id'] == 0]
shared = treg[treg['naive_id'] > 0]
print len(only_treg), len(only_naive), len(shared)
'''
Created on Sep 7, 2012
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland_2012/GR_Project/'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs', 'from_peaks')
all_data = yzer.import_file(
yzer.get_filename(motif_dirpath, 'p65_kla_vectors.txt'))
size = 100
if True:
for ratio in (3, 2, 1.5):
enhancers = yzer.import_file(
yzer.get_filename(
dirpath, 'boxplots_non_refseq_by_p65',
'enhancer_like_lose_p65_{0}x_change_dsg_only.txt'.format(
ratio)))
enhancers['glass_transcript_id'] = enhancers['id']
# Limit to peaks and touching transcripts, then pull out peaks that intersect our enhancer set
data = all_data[all_data['touches'] == 't']
data = data.merge(enhancers,
how='right',
on='glass_transcript_id',
grapher.get_filename(
scatter_dirpath,
'two_color_kla_vs_kla_dex_group_{0}_runs_{1}.png'.
format(x, slug_label)))
grapher.show_plot()
if False:
# Gene names
print grapher.get_gene_names(refseq[(refseq['kla_1_lfc'] >= 1)],
add_quotes=True)
print grapher.get_gene_names(
refseq[(refseq['kla_1_lfc'] >= 1)
& (refseq['dex_over_kla_1_lfc'] < -.58)])
if False:
yzer = MotifAnalyzer()
motif_dirpath = yzer.get_filename(dirpath, 'motifs/size_200')
distal = data[data['distal'] == 't']
dataset = distal[(distal['kla_lfc'] >= 1)]
yzer.prep_files_for_homer(dataset,
'distal_up_in_kla_all',
motif_dirpath,
center=False,
reverse=False,
preceding=True,
size=200)
dataset = distal[(distal['kla_1_lfc'] >= 1)]
yzer.prep_files_for_homer(dataset,
'distal_up_in_kla_1',
'''
Created on Nov 7, 2012
@author: karmel
'''
from __future__ import division
from glasslab.dataanalysis.misc.cd4tcell_finland_2012.resources import replicate_sets
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells_Finland_2012/Analysis_2013_02'
dirpath = yzer.get_path(dirpath)
go_path = yzer.get_and_create_path(dirpath, 'with_me3', 'go_analysis',
'0_8_min_lfc')
data = yzer.import_file(
yzer.get_filename(dirpath, 'transcript_vectors.txt'))
data = data.fillna(0)
data = data[data['naive_me3_tag_count'] + data['act_me3_tag_count'] > 0]
if False:
curr_path = yzer.get_and_create_path(dirpath, 'with_me3',
'motif_analysis')
yzer.run_homer(data,
'all_refseq_preceding',
curr_path,
center=False,
reverse=False,
preceding=True,
'''
Created on Oct 5, 2014
@author: karmel
'''
from glasslab.dataanalysis.misc.rodrigo.samples import get_threshold,\
get_breed_sets
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Enhancers_set2'
dirpath = yzer.get_path(dirpath)
datasets = {}
breed_sets = get_breed_sets()
# Working with only WT for now
sample_names, short_names = breed_sets[0]
# Set up our samples
datasets = {}
for sample_prefix in short_names:
sample_dirpath = yzer.get_filename(dirpath, sample_prefix)
filename = yzer.get_filename(sample_dirpath,
sample_prefix + '_enhancers.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
'''
Created on Sep 27, 2014
@author: karmel
'''
from glasslab.dataanalysis.misc.rodrigo.samples import SAMPLES, sample_name,\
get_threshold
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Enhancers'
dirpath = yzer.get_path(dirpath)
for cond, seq, breed in SAMPLES:
sample_prefix = sample_name(cond, seq, breed)
sample_dirpath = yzer.get_filename(dirpath, sample_prefix)
filename = yzer.get_filename(sample_dirpath,
sample_prefix + '_enhancers.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
if True:
min_thresh = get_threshold(seq)
subdata = data[data['tag_count'] >= min_thresh]
yzer.run_homer(subdata,
'all',
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/AND_TCR/Analysis/IDR/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', ): #'ac'):
for peptide in ('K99A', 'NoPep', 'PCC'):
pep_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(peptide, ab))
if True:
filename = yzer.get_filename(
pep_dirpath, '{}_{}_enhancers.txt'.format(peptide, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
if True:
yzer.run_homer(data,
'all',
pep_dirpath,
cpus=8,
center=True,
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/H3K4me2/Analysis'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs')
filename = yzer.get_filename(dirpath, 'thio_peak_vectors.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
# me2
if True:
if False:
yzer.run_homer(data,
'thio_all',
motif_dirpath,
cpus=6,
center=True,
reverse=False,
preceding=False,
size=200,
length=[8, 10, 12, 15])
data = data[data['tss_id'] == 0]
'''
Created on Feb 12, 2013
@author: karmel
Note that set 1 of Rudensky's Foxp3 chip has 2x as many peaks,
but he seems to use set 2 in the paper (?). Here we use
summed tags and peaks found in that.
'''
from __future__ import division
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.dataanalysis.graphing.seq_grapher import SeqGrapher
if __name__ == '__main__':
yzer = MotifAnalyzer()
grapher = SeqGrapher()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/TReg_enhancers/2013_04_01'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs')
graph_dirpath = yzer.get_filename(dirpath, 'piecharts')
min_score = 10
for antibody in ('me2', ):
data = {}
celltypes = ['treg', 'naive', 'th1', 'th2']
for celltype in ('treg', 'th1'):
data[celltype] = yzer.import_file(
yzer.get_filename(
dirpath, '{0}_{1}_versus_others_with_foxp3.txt'.format(
celltype, antibody))).fillna(0)
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/iTreg_enhancers/2014_04_11/'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
condition = 'treg'
motif_dirpath = yzer.get_filename(dirpath, 'Motifs')
filename = yzer.get_filename(
dirpath, '{}_{}_enhancers.txt'.format(condition, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
min_thresh = 20
if False:
subdata = data[data['tag_count'] > min_thresh]
subdata = subdata[subdata['tag_count(2)'] > min_thresh]
subdata = subdata[subdata['tag_count(3)'] <= 0]
subdata = subdata[subdata['tag_count(4)'] <= 0]
subdir = 'treg_shared_' + ab
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/TReg_enhancers/2013_03_19'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs')
for antibody in ('me2', ):
data = {}
celltypes = ['treg', 'naive', 'th1', 'th2']
for celltype in celltypes:
data[celltype] = yzer.import_file(
yzer.get_filename(
dirpath,
'{0}_{1}_versus_others.txt'.format(celltype,
antibody))).fillna(0)
# Filter out promoters
data[celltype] = data[celltype][data[celltype]['tss_id'] == 0]
# Get venn-diagram "only" sets
others = celltypes[:]
others.remove(celltype)
'''
Created on Sep 27, 2014
@author: karmel
'''
from glasslab.dataanalysis.misc.rodrigo.samples import sample_name,\
get_threshold
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Promoters'
dirpath = yzer.get_path(dirpath)
cond, seq, breed = ('naive', 'atac', '')
sample_prefix = sample_name(cond, seq, breed)
sample_dirpath = yzer.get_filename(dirpath, sample_prefix)
filename = yzer.get_filename(sample_dirpath,
sample_prefix + '_promoters.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
min_thresh = get_threshold(seq)
data = data[data['tag_count'] >= min_thresh]
fold = 2
if True:
# ATAC peaks that are absent in the FOXO1 KO
foxo1_critical = data[
'''
Created on Sep 27, 2014
@author: karmel
'''
from glasslab.dataanalysis.misc.rodrigo.samples import sample_name,\
get_threshold
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Enhancers'
dirpath = yzer.get_path(dirpath)
cond, seq, breed = ('naive', 'foxo1', '')
sample_prefix = sample_name(cond, seq, breed)
sample_dirpath = yzer.get_filename(dirpath, sample_prefix)
filename = yzer.get_filename(sample_dirpath,
sample_prefix + '_enhancers.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
min_thresh = get_threshold(seq)
data = data[data['tag_count'] >= min_thresh]
fold = 2
if True:
naive_only = data[data['lcmv_d12_foxo1_tag_count'] < min_thresh]
yzer.run_homer(naive_only,
l + '\n(count: {0})'.format(len(v)) for l, v in zip(labels, vals)
]
title = 'Balbc vs NOD Log-fold Change According to SRF Binding'
if True:
ax = yzer.boxplot(
vals,
labels,
title=title,
xlabel='Transcript subset',
ylabel='log2(NOD notx 1h GRO-seq/BALBc notx 1h GRO-seq)',
show_outliers=False,
show_plot=True,
wide=True,
save_dir=img_dirpath)
if False:
yzer = MotifAnalyzer()
motifs_dirpath = yzer.get_filename(dirpath, 'motifs')
# Motif finding for enhancers with SRF
bg = yzer.get_filename(
motifs_dirpath,
'enhancer_at_least_10_tags_15/enhancer_at_least_10_tags_15_regions_for_homer.txt'
)
enh_with_srf = datasets[4]
yzer.run_homer(enh_with_srf,
'enhancer_with_srf',
motifs_dirpath,
cpus=6,
center=False,
reverse=False,
preceding=False,
'''
Created on Feb 20, 2013
@author: karmel
'''
from __future__ import division
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/Rudensky_enhancers'
dirpath = yzer.get_path(dirpath)
motifs_dirpath = yzer.get_and_create_path(dirpath, 'motifs')
peak_pretty = 'Foxp3'
peak = peak_pretty.lower()
foxp3 = yzer.import_file(
yzer.get_filename(dirpath,
'{0}_1_with_naive_me2.txt'.format(peak))).fillna(0)
naive = yzer.import_file(
yzer.get_filename(dirpath,
'naive_me2_with_{0}.txt'.format(peak))).fillna(0)
# Filter out promoters
foxp3 = foxp3[foxp3['tss_id'] == 0]
naive = naive[naive['tss_id'] == 0]
# Get venn-diagram sets for foxp3/me2
only_foxp3 = foxp3[foxp3['naive_id'] == 0]
only_naive = naive[naive['foxp3_1_id'] == 0]
'''
Created on Jul 4, 2012
@author: karmel
The goal here is to see if we can tease out which features are
predictive of characteristics of interest, such as pausing ratio
or transrepression. Worth a stab.
'''
from glasslab.dataanalysis.misc.gr_project_2012.elongation import get_rep_string
import sys
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland_2012/GR_Project/motifs'
dirpath = yzer.get_path(dirpath)
grouped = yzer.import_file(
yzer.get_filename(dirpath, 'feature_vectors.txt'))
paused = []
if True:
# Can we predict pausing ratio?
# Minimal ratio in KLA+Dex vs. KLA pausing
try:
min_ratio = float(sys.argv[1])
except IndexError:
min_ratio = 2
try:
'''
Created on Sep 7, 2012
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland_2012/GR_Project/motifs'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath,'from_peaks')
transcripts = yzer.import_file(yzer.get_filename(dirpath, 'transcript_vectors.txt'))
transcripts['glass_transcript_id'] = transcripts['id']
for peak_type in ('gr_dex', 'gr_kla_dex', 'p65_kla_dex','p65_kla'):
size = 100
if True:
all_data = yzer.import_file(yzer.get_filename(motif_dirpath,
'{0}_vectors.txt'.format(peak_type)))
all_data = all_data.merge(transcripts, how='left', on='glass_transcript_id',suffixes=['','trans'])
all_data = all_data.fillna(0)
for super_name, data in (#('all', all_data,),
#('refseq', all_data[(all_data['score'] > 10) & (all_data['has_refseq'] == 1)
# & (all_data['touches'] == 't') | (all_data['relationship'] == 'is downstream of')],),
#('distal', all_data[(all_data['distal'] == 't')],),
'''
Created on Mar 1, 2014
@author: karmel
'''
from __future__ import division
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/iTreg_enhancers/2014_02_14/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
for condition in ('treg', 'itreg'):
cond_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(condition, ab))
filename = yzer.get_filename(
cond_dirpath, '{}_{}_enhancers.txt'.format(condition, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
print('Total {} {}'.format(condition, ab))
print(len(data))
min_thresh = 20
if True:
data = data[data['tag_count'] > min_thresh]
'''
Created on Feb 8, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
base_dirpath = yzer.get_path(
'karmel/GlassLab/Notes_and_Reports/NOD_BALBc/ThioMacs/Analysis_2013_02/'
)
dirpath = yzer.get_and_create_path(base_dirpath, 'motifs/')
filename = yzer.get_filename(base_dirpath, 'transcript_vectors.txt')
data = yzer.import_file(filename)
data = data.fillna(0)
# Promoters
if False:
refseq = data[data['has_refseq'] == 1]
refseq = refseq[refseq['transcript_score'] >= 4]
if True:
yzer.run_homer(refseq,
'refseq_promoter',
dirpath,
cpus=6,
center=False,
reverse=False,
preceding=True,
size=400,
@author: karmel
Output TSV for all of the WT and Foxo1 KO ATAC seq samples from the LCMV
timecourse.
If running on Mac, make sure psycopg2 is set up:
sudo ln -s /Library/PostgreSQL/9.2/lib/libssl.1.0.0.dylib /usr/lib
sudo ln -s /Library/PostgreSQL/9.2/lib/libcrypto.1.0.0.dylib /usr/lib
'''
from glasslab.dataanalysis.misc.rodrigo.samples import get_breed_sets
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.utils.database import get_engine, dataframe_from_query
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Enhancers_set2'
dirpath = yzer.get_path(dirpath)
# Get DB engine
engine = get_engine(uri='ec2-23-20-125-153.compute-1.amazonaws.com',
password='******')
breed_sets = get_breed_sets()
# Run and save output from sql queries
for k, (samples, short_names) in enumerate(breed_sets):
oth_breed = breed_sets[1 - k]
for i, sample in enumerate(samples):
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/AND_TCR/Analysis/Chips1_2/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
for peptide in ('K99A', 'NoPep', 'PCC'):
pep_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(peptide, ab))
if False:
filename = yzer.get_filename(
pep_dirpath, '{}_{}_enhancers.txt'.format(peptide, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
yzer.run_homer(data,
'all',
pep_dirpath,
cpus=6,
center=True,
reverse=False,
'''
Created on Jul 5, 2012
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.dataanalysis.misc.gr_project_2012.elongation import get_rep_string
import sys
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland_2012/GR_Project/motifs'
dirpath = yzer.get_path(dirpath)
pausing_data = yzer.import_file(
yzer.get_filename(dirpath, 'feature_vectors.txt'))
data = yzer.import_file(
yzer.get_filename(dirpath, 'transcript_vectors.txt'))
try:
min_ratio = float(sys.argv[1])
except IndexError:
min_ratio = 1.5
if False:
yzer.prep_files_for_homer(data,
'all_transcripts_promoter',
dirpath,
center=False,
reverse=False,
preceding=True,
'''
Created on Jul 4, 2012
@author: karmel
'''
from __future__ import division
from glasslab.dataanalysis.misc.gr_project_2012.elongation import get_rep_string
import sys
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.dataanalysis.graphing.seq_grapher import SeqGrapher
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland 2012/KLA skew/'
dirpath = yzer.get_path(dirpath)
grouped = yzer.import_file(
yzer.get_filename(dirpath, 'feature_vectors.txt'))
repressed = []
if True:
# Can we predict pausing ratio?
# Minimal ratio in KLA+Dex vs. KLA pausing
try:
min_ratio = float(sys.argv[1])
except IndexError:
min_ratio = -1
try:
thresh = int(sys.argv[3])
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/iTreg_enhancers/2014_02_14/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
for condition in ('treg', 'itreg', 'activated'):
cond_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(condition, ab))
if True:
filename = yzer.get_filename(
cond_dirpath, '{}_{}_enhancers.txt'.format(condition, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
yzer.run_homer(data,
'all',
cond_dirpath,
cpus=6,
center=True,
reverse=False,
'''
Created on Feb 12, 2013
@author: karmel
'''
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/iTreg_enhancers/2014_02_14/Motifs'
dirpath = yzer.get_path(dirpath)
for ab in ('me2', 'ac'):
for condition in ('treg', 'itreg', 'activated'):
cond_dirpath = yzer.get_filename(dirpath,
'{}_{}'.format(condition, ab))
if False:
filename = yzer.get_filename(
cond_dirpath, '{}_{}_enhancers.txt'.format(condition, ab))
data = yzer.import_file(filename)
data = data.fillna(0)
min_thresh = 20
data = data[data['tag_count'] > min_thresh]
yzer.run_homer(data,
'filtered_{}'.format(min_thresh),
cond_dirpath,
cpus=6,
'''
Created on Jul 4, 2012
@author: karmel
'''
from glasslab.dataanalysis.misc.gr_project_2012.elongation import get_rep_string
import sys
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.dataanalysis.graphing.seq_grapher import SeqGrapher
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/Classes/Rotations/Finland 2012/KLA skew/'
dirpath = yzer.get_path(dirpath)
grouped = yzer.import_file(
yzer.get_filename(dirpath, 'feature_vectors.txt'))
repressed = []
if True:
# Can we predict pausing ratio?
# Minimal ratio in KLA+Dex vs. KLA pausing
try:
min_ratio = float(sys.argv[1])
except IndexError:
min_ratio = -1
try:
thresh = int(sys.argv[3])
except IndexError:
'''
Created on Feb 12, 2013
@author: karmel
Note that set 1 of Rudensky's Foxp3 chip has 2x as many peaks,
but he seems to use set 2 in the paper (?). Here we use
summed tags and peaks found in that.
'''
from __future__ import division
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from collections import OrderedDict
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/TReg_enhancers/2013_04_01'
dirpath = yzer.get_path(dirpath)
motif_dirpath = yzer.get_filename(dirpath, 'motifs')
min_score = 10
stats = OrderedDict()
all_data = yzer.import_file(
yzer.get_filename(dirpath, 'foxp3_with_treg_enhancers.txt')).fillna(0)
stats['all'] = len(all_data)
# Filter out promoters
data = all_data[all_data['tss_id'] == 0]
stats['enhancers'] = len(data)
stats['me2_treg'] = sum(data['me2_id'] > 0)
stats['ac_treg'] = sum(data['ac_id'] > 0)
'''
Created on Apr 12, 2013
@author: karmel
'''
from __future__ import division
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/CD4TCells/TReg_enhancers/2013_04_01'
dirpath = yzer.get_path(dirpath)
data = yzer.import_file(yzer.get_filename(
dirpath, 'th1_with_stat1_ko.txt')).fillna(0)
print len(data)
data = data[data['tss_id'] == 0]
data['ko_ratio'] = data['ko_id'] / data['th1_id']
data['treg_ratio'] = data['treg_id'] / data['th1_id']
enh = len(data)
print enh
print sum(data['ko_ratio'] < .5), sum(data['ko_ratio'] < .5) / enh
print sum(data['treg_ratio'] < .5), sum(data['treg_ratio'] < .5) / enh
print sum((data['treg_ratio'] < .5) & (data['ko_ratio'] < .5))
print sum((data['treg_ratio'] < .5) & (data['ko_ratio'] < .5)) / enh
'''
Created on Oct 2, 2014
@author: karmel
Output TSV for all of the WT and Foxo1 KO ATAC seq samples from the first
dataset.
'''
from glasslab.dataanalysis.misc.rodrigo.samples import SAMPLES, sample_name
from glasslab.dataanalysis.motifs.motif_analyzer import MotifAnalyzer
from glasslab.utils.database import get_engine, dataframe_from_query
if __name__ == '__main__':
yzer = MotifAnalyzer()
dirpath = 'karmel/Desktop/Projects/GlassLab/Notes_and_Reports/' +\
'Miscellaneous_Collaborations/Rodrigo_CD8s_2014_09/Promoters'
dirpath = yzer.get_path(dirpath)
# Get DB engine
engine = get_engine(uri='ec2-23-20-125-153.compute-1.amazonaws.com',
password='******')
for i, (cond, seq, breed) in enumerate(SAMPLES):
others = SAMPLES[:i] + SAMPLES[i + 1:]
sql = '''select distinct on (p1.id)
chr."name" as chr_name, p1."start", p1."end", p1.tag_count,
p1.*,
'''
selects, joins = [], []
for j, (oth_cond, oth_seq, oth_breed) in enumerate(others):
counter = j + 2
