BPNet tf-Modisco report

In [1]:
modisco_dir = "/users/avsec/workspace/basepair/data/processed/chipnexus/exp/models/oct-sox-nanog-klf/models/n_dil_layers=9/modisco/valid/new-hparams"
In [2]:
# Parameters
modisco_dir = "."
In [3]:
from basepair.modisco.results import ModiscoResult
from basepair.config import get_data_dir
from basepair.utils import read_json
from basepair.plot.vdom import vdom_modisco
from kipoi.readers import HDF5Reader
import matplotlib.pyplot as plt
import numpy as np
import os
import pandas as pd
from plotnine import *

Using TensorFlow backend.
In [4]:
mr = ModiscoResult(f"{modisco_dir}/modisco.h5")
mr.open()
In [5]:
# load the data
modisco_kwargs = read_json(os.path.join(modisco_dir, "kwargs.json"))
d = HDF5Reader(modisco_kwargs['imp_scores'])
d.open()
In [6]:
strand_dist_file = f"{modisco_dir}/strand_distances.h5"
if modisco_kwargs.get("ignore_strand_dist", False) and os.path.exists(strand_dist_file):
    included_samples = HDF5Reader.load(strand_dist_file)['included_samples']
else:
    included_samples = np.ones(d.f['inputs'].shape[:1], dtype=bool)
    
if modisco_kwargs.get("filter_npy", None) is not None:
    included_samples = np.load(modisco_kwargs['filter_npy']) * included_samples
    
    
id_hash = pd.DataFrame({"peak_id": d.f['/metadata/interval_from_task'][:][included_samples], 
                        "example_idx": np.arange(d.f['/metadata/interval_from_task'][included_samples].shape[0])})
tasks = list(d.f["targets"]["profile"].keys())
In [7]:
# get all seqlet instances
dfp = mr.seqlet_df_instances().rename(columns=dict(seqname="example_idx"))
dfp = pd.merge(dfp, id_hash, on="example_idx")

TF-MoDISco is using the TensorFlow backend.
In [8]:
# row = example_idx
total_counts = pd.DataFrame({task: d.f[f"/targets/profile/{task}"][:][included_samples].sum(axis=-1).sum(axis=-1)
          for task in tasks
         })
In [9]:
len(mr.patterns())
Out[9]:
122
In [10]:
# total number of seqlets
len(dfp)
Out[10]:
69360
In [11]:
# Number of metaclusters
len(mr.metaclusters())
Out[11]:
9

Number of seqlets per pattern

In [12]:
mc_stat = mr.metacluster_stats()
In [13]:
ggplot(aes(x="pattern", y='n'), mc_stat) + geom_bar(stat='identity') + \
    facet_wrap("~metacluster", ncol=4, labeller='label_both') + \
    ylab("Number of seqlets") + theme_classic()
Out[13]:
<ggplot: (8747734615535)>

Zoom-into the 500 seqlet range

In [14]:
ggplot(aes(x="pattern", y='n'), mc_stat) + geom_bar(stat='identity') + \
    facet_wrap("~metacluster", ncol=4, labeller='label_both') + \
    ylab("Number of seqlets") + theme_classic() + coord_cartesian(ylim=[0, 500])
Out[14]:
<ggplot: (-9223363289119726115)>

Important tasks per metacluster

In [15]:
mcs_grouped = mc_stat.groupby("metacluster").n.agg(["count", "sum"]).reset_index()
fig, ax = plt.subplots(2, 1, sharex=False, figsize=(18,6), 
                       gridspec_kw={'height_ratios': [2,1]})
mcs_grouped.plot("metacluster", "count", 
                 label="# patterns per metacluster", style="o--", 
                 ax=ax[0], 
                 yticks=range(mcs_grouped['count'].max()+1),
                 xticks=range(38),
                 fontsize='large',
                 xlim=(-.5, len(mr.metaclusters()) - .5 ))
mcs_grouped.plot("metacluster", "sum", 
                 label="# seqlets per metacluster", 
                 style="o--", ax=ax[0], secondary_y=True)
ax[0].grid(linewidth=0.2)
mr.plot_metacluster_activity(ax[1], cbar=False)
ax[1].set_title("Importance score activity: Red = positive, Blue = negative");
In [16]:
vdom_modisco(mr, "plots", total_counts, dfp, is_open=True, trim_frac=0.08, letter_width=0.15, height=0.5)
Out[16]:
metacluster_0, # patterns: 30, # seqlets: 23441, important for: Klf4,Nanog,Oct4,Sox2
  • pattern_0: # seqlets: 13278
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 12588 / 98428 regions (12.8%)

  • pattern_1: # seqlets: 1862
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1840 / 98428 regions (1.9%)

  • pattern_2: # seqlets: 1647
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1613 / 98428 regions (1.6%)

  • pattern_3: # seqlets: 847
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 843 / 98428 regions (0.9%)

  • pattern_4: # seqlets: 669
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 658 / 98428 regions (0.7%)

  • pattern_5: # seqlets: 583
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 582 / 98428 regions (0.6%)

  • pattern_6: # seqlets: 530
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 530 / 98428 regions (0.5%)

  • pattern_7: # seqlets: 379
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 379 / 98428 regions (0.4%)

  • pattern_8: # seqlets: 323
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 322 / 98428 regions (0.3%)

  • pattern_9: # seqlets: 315
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 315 / 98428 regions (0.3%)

  • pattern_10: # seqlets: 272
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 272 / 98428 regions (0.3%)

  • pattern_11: # seqlets: 233
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 233 / 98428 regions (0.2%)

  • pattern_12: # seqlets: 246
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 246 / 98428 regions (0.2%)

  • pattern_13: # seqlets: 211
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 211 / 98428 regions (0.2%)

  • pattern_14: # seqlets: 208
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 208 / 98428 regions (0.2%)

  • pattern_15: # seqlets: 230
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 230 / 98428 regions (0.2%)

  • pattern_16: # seqlets: 191
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 191 / 98428 regions (0.2%)

  • pattern_17: # seqlets: 170
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 169 / 98428 regions (0.2%)

  • pattern_18: # seqlets: 156
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 156 / 98428 regions (0.2%)

  • pattern_19: # seqlets: 131
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 127 / 98428 regions (0.1%)

  • pattern_20: # seqlets: 112
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 112 / 98428 regions (0.1%)

  • pattern_21: # seqlets: 111
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 111 / 98428 regions (0.1%)

  • pattern_22: # seqlets: 97
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 97 / 98428 regions (0.1%)

  • pattern_23: # seqlets: 95
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 95 / 98428 regions (0.1%)

  • pattern_24: # seqlets: 92
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 92 / 98428 regions (0.1%)

  • pattern_25: # seqlets: 87
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 87 / 98428 regions (0.1%)

  • pattern_26: # seqlets: 85
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 85 / 98428 regions (0.1%)

  • pattern_27: # seqlets: 135
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 133 / 98428 regions (0.1%)

  • pattern_28: # seqlets: 83
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 83 / 98428 regions (0.1%)

  • pattern_29: # seqlets: 63
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 63 / 98428 regions (0.1%)

metacluster_1, # patterns: 18, # seqlets: 18267, important for: Klf4
  • pattern_0: # seqlets: 12735
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 12113 / 98428 regions (12.3%)

  • pattern_1: # seqlets: 2064
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1985 / 98428 regions (2.0%)

  • pattern_2: # seqlets: 935
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 927 / 98428 regions (0.9%)

  • pattern_3: # seqlets: 400
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 390 / 98428 regions (0.4%)

  • pattern_4: # seqlets: 382
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 380 / 98428 regions (0.4%)

  • pattern_5: # seqlets: 246
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 245 / 98428 regions (0.2%)

  • pattern_6: # seqlets: 179
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 174 / 98428 regions (0.2%)

  • pattern_7: # seqlets: 157
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 157 / 98428 regions (0.2%)

  • pattern_8: # seqlets: 162
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 159 / 98428 regions (0.2%)

  • pattern_9: # seqlets: 146
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 146 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 139
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 138 / 98428 regions (0.1%)

  • pattern_11: # seqlets: 117
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 113 / 98428 regions (0.1%)

  • pattern_12: # seqlets: 100
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 99 / 98428 regions (0.1%)

  • pattern_13: # seqlets: 87
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 86 / 98428 regions (0.1%)

  • pattern_14: # seqlets: 98
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 97 / 98428 regions (0.1%)

  • pattern_15: # seqlets: 97
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 97 / 98428 regions (0.1%)

  • pattern_16: # seqlets: 160
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 154 / 98428 regions (0.2%)

  • pattern_17: # seqlets: 63
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 63 / 98428 regions (0.1%)

metacluster_2, # patterns: 23, # seqlets: 9087, important for: Nanog
  • pattern_0: # seqlets: 4791
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 4668 / 98428 regions (4.7%)

  • pattern_1: # seqlets: 865
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 862 / 98428 regions (0.9%)

  • pattern_2: # seqlets: 401
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 400 / 98428 regions (0.4%)

  • pattern_3: # seqlets: 393
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 392 / 98428 regions (0.4%)

  • pattern_4: # seqlets: 434
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 432 / 98428 regions (0.4%)

  • pattern_5: # seqlets: 230
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 230 / 98428 regions (0.2%)

  • pattern_6: # seqlets: 167
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 167 / 98428 regions (0.2%)

  • pattern_7: # seqlets: 166
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 166 / 98428 regions (0.2%)

  • pattern_8: # seqlets: 153
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 146 / 98428 regions (0.1%)

  • pattern_9: # seqlets: 146
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 146 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 146
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 145 / 98428 regions (0.1%)

  • pattern_11: # seqlets: 126
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 126 / 98428 regions (0.1%)

  • pattern_12: # seqlets: 126
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 126 / 98428 regions (0.1%)

  • pattern_13: # seqlets: 138
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 137 / 98428 regions (0.1%)

  • pattern_14: # seqlets: 106
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 105 / 98428 regions (0.1%)

  • pattern_15: # seqlets: 89
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 89 / 98428 regions (0.1%)

  • pattern_16: # seqlets: 94
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 94 / 98428 regions (0.1%)

  • pattern_17: # seqlets: 89
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 88 / 98428 regions (0.1%)

  • pattern_18: # seqlets: 103
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 103 / 98428 regions (0.1%)

  • pattern_19: # seqlets: 67
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 67 / 98428 regions (0.1%)

  • pattern_20: # seqlets: 75
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 75 / 98428 regions (0.1%)

  • pattern_21: # seqlets: 89
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 89 / 98428 regions (0.1%)

  • pattern_22: # seqlets: 93
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 93 / 98428 regions (0.1%)

metacluster_3, # patterns: 11, # seqlets: 5444, important for: Nanog,Sox2
  • pattern_0: # seqlets: 1661
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1633 / 98428 regions (1.7%)

  • pattern_1: # seqlets: 1038
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1022 / 98428 regions (1.0%)

  • pattern_2: # seqlets: 684
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 680 / 98428 regions (0.7%)

  • pattern_3: # seqlets: 632
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 626 / 98428 regions (0.6%)

  • pattern_4: # seqlets: 636
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 636 / 98428 regions (0.6%)

  • pattern_5: # seqlets: 210
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 210 / 98428 regions (0.2%)

  • pattern_6: # seqlets: 159
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 159 / 98428 regions (0.2%)

  • pattern_7: # seqlets: 90
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 89 / 98428 regions (0.1%)

  • pattern_8: # seqlets: 97
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 97 / 98428 regions (0.1%)

  • pattern_9: # seqlets: 78
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 78 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 159
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 159 / 98428 regions (0.2%)

metacluster_4, # patterns: 13, # seqlets: 4467, important for: Nanog,Oct4,Sox2
  • pattern_0: # seqlets: 1488
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 1471 / 98428 regions (1.5%)

  • pattern_1: # seqlets: 662
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 662 / 98428 regions (0.7%)

  • pattern_2: # seqlets: 457
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 454 / 98428 regions (0.5%)

  • pattern_3: # seqlets: 383
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 366 / 98428 regions (0.4%)

  • pattern_4: # seqlets: 334
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 332 / 98428 regions (0.3%)

  • pattern_5: # seqlets: 273
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 267 / 98428 regions (0.3%)

  • pattern_6: # seqlets: 251
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 251 / 98428 regions (0.3%)

  • pattern_7: # seqlets: 154
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 151 / 98428 regions (0.2%)

  • pattern_8: # seqlets: 129
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 129 / 98428 regions (0.1%)

  • pattern_9: # seqlets: 91
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 91 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 89
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 88 / 98428 regions (0.1%)

  • pattern_11: # seqlets: 78
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 78 / 98428 regions (0.1%)

  • pattern_12: # seqlets: 78
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 78 / 98428 regions (0.1%)

metacluster_5, # patterns: 12, # seqlets: 3964, important for: Klf4,Nanog,Sox2
  • pattern_0: # seqlets: 850
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 846 / 98428 regions (0.9%)

  • pattern_1: # seqlets: 693
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 682 / 98428 regions (0.7%)

  • pattern_2: # seqlets: 730
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 726 / 98428 regions (0.7%)

  • pattern_3: # seqlets: 598
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 598 / 98428 regions (0.6%)

  • pattern_4: # seqlets: 199
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 199 / 98428 regions (0.2%)

  • pattern_5: # seqlets: 197
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 197 / 98428 regions (0.2%)

  • pattern_6: # seqlets: 157
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 157 / 98428 regions (0.2%)

  • pattern_7: # seqlets: 146
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 142 / 98428 regions (0.1%)

  • pattern_8: # seqlets: 95
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 95 / 98428 regions (0.1%)

  • pattern_9: # seqlets: 105
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 105 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 99
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 99 / 98428 regions (0.1%)

  • pattern_11: # seqlets: 95
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 95 / 98428 regions (0.1%)

metacluster_6, # patterns: 13, # seqlets: 3594, important for: Klf4,Nanog
  • pattern_0: # seqlets: 929
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 920 / 98428 regions (0.9%)

  • pattern_1: # seqlets: 703
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 699 / 98428 regions (0.7%)

  • pattern_2: # seqlets: 355
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 355 / 98428 regions (0.4%)

  • pattern_3: # seqlets: 319
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 317 / 98428 regions (0.3%)

  • pattern_4: # seqlets: 254
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 254 / 98428 regions (0.3%)

  • pattern_5: # seqlets: 216
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 216 / 98428 regions (0.2%)

  • pattern_6: # seqlets: 172
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 172 / 98428 regions (0.2%)

  • pattern_7: # seqlets: 116
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 116 / 98428 regions (0.1%)

  • pattern_8: # seqlets: 100
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 100 / 98428 regions (0.1%)

  • pattern_9: # seqlets: 103
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 101 / 98428 regions (0.1%)

  • pattern_10: # seqlets: 144
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 144 / 98428 regions (0.1%)

  • pattern_11: # seqlets: 89
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 89 / 98428 regions (0.1%)

  • pattern_12: # seqlets: 94
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 94 / 98428 regions (0.1%)

metacluster_7, # patterns: 2, # seqlets: 1096, important for: Oct4,Sox2
  • pattern_0: # seqlets: 896
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 894 / 98428 regions (0.9%)

  • pattern_1: # seqlets: 200
    Aggregated profiles and contribution scores)
    Aggregated hypothetical contribution scores)
    Sequence

    ChIP-nexus counts
    Importance scores (profile)
    Importance scores (counts)
    Positional distribution
    Total count distribution

    Pattern occurs in 197 / 98428 regions (0.2%)

In [17]:
print("Metaclusters heatmap")
import seaborn as sns
activity_patterns = np.array(mr.f.f['metaclustering_results']['attribute_vectors'])[
                    np.array(
        [x[0] for x in sorted(
                enumerate(mr.f.f['metaclustering_results']['metacluster_indices']),
               key=lambda x: x[1])])]
sns.heatmap(activity_patterns, center=0);

Metaclusters heatmap