Skip to content
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

Jump to bottom

Ancestor lengths in genome and time bins #68

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Ancestor lengths in genome and time bins #68

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
fb483e2
Added child_left, child_right columns to nodes_df and tests for these.
savitakartik Aug 25, 2023
e86158c
int slider removed, edge case fixed, bin count logged, actual node co…
savitakartik Sep 2, 2023
47536db
naively handling case for heatmap where time units are uncalibrated.
savitakartik Oct 2, 2023
cb74ebc
incremented nodes cache version
savitakartik Oct 2, 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
67 changes: 66 additions & 1 deletion model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -436,7 +436,7 @@ def edges_df(self):
)

@cached_property
@disk_cache("v1")
@disk_cache("v2")
def nodes_df(self):
ts = self.ts
child_left, child_right = self.child_bounds(
Expand All @@ -449,6 +449,10 @@ def nodes_df(self):
"time": ts.nodes_time,
"num_mutations": self.nodes_num_mutations,
"ancestors_span": child_right - child_left,
"child_left": child_left, # FIXME add test for this
"child_right": child_right, # FIXME add test for this
"child_left": child_left, # FIXME add test for this
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Duplication here?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also, looks like it's tested now?

"child_right": child_right, # FIXME add test for this
"is_sample": is_sample,
}
)
Expand All @@ -458,6 +462,8 @@ def nodes_df(self):
"time": "float64",
"num_mutations": "int",
"ancestors_span": "float64",
"child_left": "float64",
"child_right": "float64",
"is_sample": "bool",
}
)
Expand Down Expand Up @@ -584,3 +590,62 @@ def calc_mutations_per_tree(self):
mutations_per_tree = np.zeros(self.ts.num_trees, dtype=np.int64)
mutations_per_tree[unique_values] = counts
return mutations_per_tree

def compute_ancestor_spans_heatmap_data(self, num_x_bins, num_y_bins):
"""
Calculates the average ancestor span in a genomic-time window
"""
if self.ts.time_units == tskit.TIME_UNITS_UNCALIBRATED:
logger.warning(
"Cannot compute ancestor spans for uncalibrated tree sequence"
)
return pd.DataFrame(
{
"position": [],
"time": [],
"overlapping_node_count_log10": [],
"overlapping_node_count": [],
}
)
else:
nodes_df = self.nodes_df[self.nodes_df.ancestors_span != -np.inf]
nodes_df = nodes_df.reset_index(drop=True)
nodes_left = nodes_df.child_left
nodes_right = nodes_df.child_right
nodes_time = nodes_df.time

x_bins = np.linspace(nodes_left.min(), nodes_right.max(), num_x_bins + 1)
y_bins = np.linspace(0, nodes_time.max(), num_y_bins + 1)
heatmap_counts = np.zeros((num_x_bins, num_y_bins))

x_starts = np.digitize(nodes_left, x_bins, right=True)
x_ends = np.digitize(nodes_right, x_bins, right=True)
y_starts = np.digitize(nodes_time, y_bins, right=True)

for u in range(len(nodes_left)):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is probably slow - should we do it with numba?

x_start = max(0, x_starts[u] - 1)
x_end = max(0, x_ends[u] - 1)
y_bin = max(0, y_starts[u] - 1)
heatmap_counts[x_start : x_end + 1, y_bin] += 1

x_coords = np.repeat(x_bins[:-1], num_y_bins)
y_coords = np.tile(y_bins[:-1], num_x_bins)
overlapping_node_count = heatmap_counts.flatten()
overlapping_node_count[overlapping_node_count == 0] = 1
# FIXME - better way to avoid log 0 above?
df = pd.DataFrame(
{
"position": x_coords.flatten(),
"time": y_coords.flatten(),
"overlapping_node_count_log10": np.log10(overlapping_node_count),
"overlapping_node_count": overlapping_node_count,
}
)
return df.astype(
{
"position": "int",
"time": "int",
"overlapping_node_count_log10": "int",
"overlapping_node_count": "int",
}
)
57 changes: 53 additions & 4 deletions pages/nodes.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
import hvplot.pandas # noqa
import numpy as np
import panel as pn
from bokeh.models import HoverTool

import config
from plot_helpers import filter_points
Expand Down Expand Up @@ -40,8 +41,15 @@ def make_node_hist_panel(tsm, log_y):
points = df_nodes.hvplot.scatter(
x="ancestors_span",
y="time",
hover_cols=["ancestors_span", "time"],
).opts(width=config.PLOT_WIDTH, height=config.PLOT_HEIGHT)
hover_cols=["ancestors_span", "time"], # add node ID
).opts(
width=config.PLOT_WIDTH,
height=config.PLOT_HEIGHT,
title="Node span by time",
xlabel="width of genome spanned by node ancestors",
ylabel="node time",
axiswise=True,
)

range_stream = hv.streams.RangeXY(source=points)
streams = [range_stream]
Expand All @@ -54,7 +62,48 @@ def make_node_hist_panel(tsm, log_y):
)

plot_options = pn.Column(
pn.pane.Markdown("# Plot Options"),
log_y_checkbox,
)
return pn.Column(main, hist_panel, plot_options)

def make_heatmap(num_x_bins, num_y_bins):
anc_span_data = tsm.compute_ancestor_spans_heatmap_data(num_x_bins, num_y_bins)
tooltips = [
("position", "@position"),
("time", "@time"),
("overlapping_nodes", "@overlapping_node_count"),
]
hover = HoverTool(tooltips=tooltips)
heatmap = hv.HeatMap(anc_span_data).opts(
width=config.PLOT_WIDTH,
height=config.PLOT_HEIGHT,
tools=[hover],
colorbar=True,
title="Average ancestor length in time and genome bins",
axiswise=True,
)
return heatmap

max_x_bins = int(np.sqrt(df_nodes.child_right.max()))
x_bin_input = pn.widgets.IntInput(
name="genome bins",
value=min(50, max_x_bins),
start=1,
end=max_x_bins,
)
max_y_bins = int(np.sqrt(df_nodes.time.max()))
y_bin_input = pn.widgets.IntInput(
name="time bins", value=min(50, int(max_y_bins)), start=1, end=max_y_bins
)
hm_options = pn.Column(x_bin_input, y_bin_input)

hm_panel = pn.bind(
make_heatmap,
num_x_bins=x_bin_input,
num_y_bins=y_bin_input,
)

return pn.Column(
pn.Column(main),
pn.Column(hist_panel, plot_options),
pn.Column(hm_panel, hm_options),
)
4 changes: 4 additions & 0 deletions tests/test_data_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -168,6 +168,8 @@ def test_single_tree_example(self):
nt.assert_array_equal(df.time, [0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 2.0])
nt.assert_array_equal(df.num_mutations, [1, 1, 1, 1, 1, 1, 1])
nt.assert_array_equal(df.ancestors_span, [10, 10, 10, 10, 10, 10, -np.inf])
nt.assert_array_equal(df.child_left, [0, 0, 0, 0, 0, 0, np.inf])
nt.assert_array_equal(df.child_right, [10, 10, 10, 10, 10, 10, 0])
nt.assert_array_equal(df.is_sample, [1, 1, 1, 1, 0, 0, 0])

def test_multiple_tree_example(self):
Expand All @@ -178,6 +180,8 @@ def test_multiple_tree_example(self):
nt.assert_array_equal(df.time, [0.0, 0.0, 0.0, 1.0, 2.0])
nt.assert_array_equal(df.num_mutations, [0, 0, 0, 0, 0])
nt.assert_array_equal(df.ancestors_span, [10, 10, 10, 10, -np.inf])
nt.assert_array_equal(df.child_left, [0, 0, 0, 0, np.inf])
nt.assert_array_equal(df.child_right, [10, 10, 10, 10, 0])
nt.assert_array_equal(df.is_sample, [1, 1, 1, 0, 0])


Expand Down
Loading