Skip to content

Use all CP kernel windowsizes #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
julianhess wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Use all CP kernel windowsizes #10

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
bf6618e
Use all window sizes for all segment lengths
julianhess Mar 24, 2023
1ddfd05
Merge branch 'master' into cp_kernel_update
julianhess May 8, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 6 additions & 21 deletions hapaseg/coverage_MCMC.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -220,7 +220,6 @@ def _run_kernel(self, residuals, window):
returns a list of peaks above the 98th quantile that are seperated by at least max(25, window/2)
"""
def _change_kernel(self, ind, residuals, window):

difs = self._run_kernel(residuals, window)
if window > 10:
x = np.r_[:len(difs)]
Expand Down Expand Up @@ -253,34 +252,20 @@ def _find_difs(self, ind):
residuals = np.exp(np.log(self.r[ind[0]:ind[1]].flatten()) - (self.mu.flatten()) - (
self.C[ind[0]:ind[1]] @ self.beta).flatten())

minimal=False
len_ind = ind[1]-ind[0]
if len_ind > 50000:
windows=np.array([1000])
elif len_ind > 10000:
windows=np.array([500])
elif len_ind > 5000:
windows=np.array([500,100,25])
elif len_ind > 1000:
windows = np.array([100, 50, 25])
else:
minimal = True
windows = np.array([50, 10])
windows = windows[2*windows < len_ind]
windows = np.logspace(1, 3, 5).astype(int)
windows = windows[2*windows < (ind[1] - ind[0])]
difs = []
for window in windows:
difs.append(self._change_kernel(ind, residuals, window))
difs_idxs = set().union(*difs)
difs_idxs = np.r_[list(difs_idxs)]
# if there are no that pass the threshold return them all
if len(difs_idxs) ==0:
# if there are no that pass the threshold return them all (should not happen?)
if len(difs_idxs) == 0:
print('no significant bins', flush=True)
return list(np.r_[ind[0] + 2:ind[1] - 1])

# add on first and last windows edges if were in a small interval
# since these are not considered by the kernel
if minimal:
difs_idxs = np.r_[difs_idxs, ind[0] + 10, ind[1]-10]
# add on first and last windows edges since these are not considered by the kernel
difs_idxs = np.r_[difs_idxs, ind[0] + 10, ind[1]-10]
return list(difs_idxs)

"""
Expand Down