from __future__ import print_function
import numpy as np
from glob import glob
from iced import io
from iced import utils
filenames = glob("data/ay2013/*10000_raw.matrix") + \
glob("data/lemieux2013/25kb/*.matrix")
filenames.sort()
for filename in filenames:
lengths = io.load_lengths(filename.replace(".matrix", ".bed"))
counts = io.load_counts(filename, lengths=lengths)
counts = counts.toarray()
counts = counts.T + counts
mask = utils.get_intra_mask(lengths)
# Just making sure there is no interaction counted in teh diag
counts[np.diag_indices_from(counts)] = 0
print(filename)
print("Total number of counts", counts.sum())
print("%% of intra", counts[mask].sum()/counts.sum() * 100)
print("%% of inter", counts[np.invert(mask)].sum()/counts.sum() * 100)
print()
from matplotlib import colors
from iced import datasets
from iced.utils import get_intra_mask
from iced.utils import get_inter_mask
"""
Extracting parts of a contact map.
This examples shows how to use a mask to plot only the inter or the intra
contact map.
"""
# Loading a sample dataset
counts, lengths = datasets.load_sample_yeast()
intra_mask = get_intra_mask(lengths)
inter_mask = get_inter_mask(lengths)
fig, axes = plt.subplots(ncols=2, figsize=(12, 6))
inter_counts = counts.copy()
inter_counts[intra_mask] = np.nan
intra_counts = counts.copy()
intra_counts[inter_mask] = np.nan
m = axes[0].matshow(intra_counts,
cmap="Blues",
norm=colors.SymLogNorm(1),
origin="bottom",
extent=(0, len(counts), 0, len(counts)))
m = axes[1].matshow(inter_counts,
cmap="Blues",
from iced import datasets
from iced.utils import get_intra_mask
from iced.utils import get_inter_mask
"""
Extracting parts of a contact map.
This examples shows how to use a mask to plot only the inter or the intra
contact map.
"""
# Loading a sample dataset
counts, lengths = datasets.load_sample_yeast()
intra_mask = get_intra_mask(lengths)
inter_mask = get_inter_mask(lengths)
fig, axes = plt.subplots(ncols=2, figsize=(12, 6))
inter_counts = counts.copy()
inter_counts[intra_mask] = np.nan
intra_counts = counts.copy()
intra_counts[inter_mask] = np.nan
m = axes[0].matshow(intra_counts, cmap="Blues", norm=colors.SymLogNorm(1),
origin="bottom",
extent=(0, len(counts), 0, len(counts)))
m = axes[1].matshow(inter_counts, cmap="Blues", norm=colors.SymLogNorm(1),
origin="bottom",
extent=(0, len(counts), 0, len(counts)))
def generate_dataset_from_distances(dis,
alpha=-3,
beta=1,
alpha_inter=None,
lengths=None,
distribution="NegativeBinomial",
random_state=None,
dispersion=7):
"""
Generate dataset from distance matrix
Parameters
----------
dis : (n, n) ndarray
alpha : float, optional, default: -3
count-to-distance parameter
beta : float, optional, default: 1
coverage or scaling factor
alpha_inter : float, optional, default: None
count-to-distance parameter for inter-chromosomal count.
When provided, lengths also needs to be provided
lengths : ndarray (L, ), optional, default: None
Vector of lengths of chromosomes.
distribution : string, optional, default: "NegativeBinomial"
The distribution used to draw contact counts from. Can be "Poisson",
"NB", "NegativeBinomial", or "Intensity".
If "Intensity" is provided, returns the intensity of the random
process instead of a random distribution.
random_state : int, optional, default: None
Determines random number generation. Use an int to make the randomness
deterministic.
dispersion : float, optional, default: 7.
Dispersion parameter for the Negative Binomial distribution.
Will be ignored for the Poisson distribution.
Returns
-------
ndarray (n, n)
"""
if distribution not in ["Poisson", "NB", "NegativeBinomial"]:
raise ValueError(
"Unknown distribution %s. Possibile distributions are %s" %
(distribution, ", ".join(_DISTRIBUTIONS)))
if lengths is not None and dis.shape[0] != lengths.sum():
raise ValueError("lengths and dis are of incompatible shapes")
if alpha_inter is not None and lengths is None:
raise ValueError(
"When alpha_inter is provided, lengths also needs to be provided")
random_state = check_random_state(random_state)
with warnings.catch_warnings():
warnings.simplefilter("ignore")
intensity = beta * dis**alpha
if alpha_inter is not None:
inter_mask = ~get_intra_mask(lengths)
with warnings.catch_warnings():
warnings.simplefilter("ignore")
intensity[inter_mask] = beta * dis[inter_mask]**alpha
intensity[np.isinf(intensity)] = 0
if distribution in ["NB", "NegativeBinomial"]:
if hasattr(dispersion, "predict"):
with warnings.catch_warnings():
warnings.simplefilter("ignore")
d = beta * dispersion.predict(dis**alpha)
else:
d = beta * dispersion
p = intensity / (intensity + d)
counts = random_state.negative_binomial(d, 1 - p)
elif distribution == "Poisson":
counts = random_state.poisson(intensity)
elif distribution == "Intensity":
counts = intensity
counts = (counts + counts.T)
return counts