"""
======
line
======
Draw line plots of site-level properties.
"""
import matplotlib.pyplot as plt
import matplotlib.ticker
import numpy
import dmslogo.colorschemes
import dmslogo.utils
[docs]
def data_units_from_linewidth(linewidth, ax, reference):
"""Convert linewidth in points to data units.
Args:
`linewidth` (float)
Linewidth in points.
`ax` (matplotlib axis)
The axis used to extract the relevant transformation
(data limits and size must not change afterwards).
`reference` (str)
The axis that is taken as a reference for the data width.
Possible values: 'x' and 'y'.
Returns
Linewidth in data units.
Function is the inverse of the one defined here:
https://stackoverflow.com/a/35501485
"""
fig = ax.get_figure()
if reference == "x":
length = fig.bbox_inches.width * ax.get_position().width
value_range = numpy.diff(ax.get_xlim())[0]
elif reference == "y":
length = fig.bbox_inches.height * ax.get_position().height
value_range = numpy.diff(ax.get_ylim())[0]
else:
raise ValueError(f"invalid `ax` of {ax}")
# Convert length to points
length *= 72
# Scale linewidth to value range
return linewidth / (length / value_range)
[docs]
def draw_line(
data,
*,
x_col,
height_col,
height_col2=None,
xtick_col=None,
show_col=None,
xlabel=None,
ylabel=None,
title=None,
color="black",
color2="gray",
show_color=dmslogo.colorschemes.CBPALETTE[1],
linewidth=1,
widthscale=1,
heightscale=1,
axisfontscale=1,
hide_axis=False,
ax=None,
ylim_setter=None,
fixed_ymin=None,
fixed_ymax=None,
):
"""Draw line plot.
Args:
`data` (pandas DataFrame)
Holds data to plot. If there are duplicate rows for
the columns of interest, removes duplicates.
`height_col` (str)
Column in `data` with line height.
`height_col2` (str or `None`)
Optional second column in `data` giving second line height. This is
typically useful when `height_col` has positive values and you also
want to plot negative values: those can be in `height_col2`.
`x_col` (str)
Column in `data` with integer site numbers. Must be full
set of sequential numbers, gaps in numbering not allowed.
`xtick_col` (`None` or str)
Column in `data` used to label sites if not using `x_col`.
`show_col` (`None` or str)
Underline sites where this column is True. Useful for
marking selected sites that are zoomed in logo plots.
`xlabel` (`None` or str)
Label for x-axis if not using `xtick_col` or `x_col`.
`ylabel` (`None` or str)
Label for y-axis if not using `height_col`.
`title` (`None` or str)
Title to place above plot.
`color` (str)
Color of line plotting data in `height_col`.
`color2` (str)
Color of line plotting any data in `height_col2`.
`show_color` (str or `None`)
Color of underlines specified by `show_col`, or `None` if
you don't want to show underlines.
`linewidth` (float)
Width of line.
`widthscale` (float)
Scale width by this much.
`heightscale` (float)
Scale height by this much.
`axisfontscale` (float)
Scale size of font for axis ticks and labels by this much.
`hide_axis` (bool)
Do we hide the axis and tick labels?
`ax` (`None` or matplotlib axes.Axes object)
Use to plot on an existing axis.
`ylim_setter` (`None` or :class:`dmslogo.utils.AxLimSetter`)
Object used to set y-limits. If `None`, a
:class:`dmslogo.utils.AxLimSetter` is created using
default parameters). If `fixed_ymin` and/or `fixed_ymax`
are set, they override the limits from this setter.
`fixed_ymin` (`None` or float)
If not `None`, then fixed y-axis minimum.
`fixed_ymax` (`None` or float)
If not `None`, then fixed y-axis maximum.
Returns:
The 2-tuple `(fig, ax)` giving the figure and axis.
"""
# set default values of arguments that can be None
if xtick_col is None:
xtick_col = x_col
if xlabel is None:
xlabel = xtick_col
if ylabel is None:
ylabel = height_col
cols = list({x_col, xtick_col, height_col})
if height_col2 is not None:
cols.append(height_col2)
if show_col:
cols.append(show_col)
if not data[show_col].dtype == bool:
raise ValueError("`show_col` is not bool")
for col in cols:
if col not in data.columns:
raise ValueError(f"`data` lacks column {col}")
data = data[cols].drop_duplicates().sort_values(x_col)
if any(data[x_col] != data[x_col].astype(int)):
raise ValueError("`x_col` does not have integer values")
xmin = data[x_col].min()
xmax = data[x_col].max()
xlen = xmax - xmin + 1
if (xlen != data[x_col].nunique()) or any(
list(range(xmin, xmax + 1)) != data[x_col].unique()
):
raise ValueError("`x_col` not sequential unbroken integers")
if len(data[x_col]) != len(data[x_col].unique()):
raise ValueError(f"not unique mapping of `x_col` to other cols {cols}")
assert len(data) == xlen
# set y-limits
if ylim_setter is None:
ylim_setter = dmslogo.utils.AxLimSetter()
ymin, ymax = ylim_setter.get_lims(data[height_col])
ydata_min = data[height_col].min()
ydata_max = data[height_col].max()
if ylim_setter.include_zero:
ydata_min = min(0, ydata_min)
ydata_max = max(0, ydata_max)
if height_col2 is not None:
ymin2, ymax2 = ylim_setter.get_lims(data[height_col2])
ymin = min(ymin, ymin2)
ymax = max(ymax, ymax2)
ydata_min = min(ydata_min, data[height_col2].min())
ydata_max = max(ydata_max, data[height_col2].max())
if fixed_ymax is not None:
if fixed_ymax < ydata_max:
raise ValueError("`fixed_ymax` less then max of data")
ymax = fixed_ymax
if fixed_ymin is not None:
if fixed_ymin > ydata_min:
raise ValueError("`fixed_ymin` greater then min of data")
ymin = fixed_ymin
# setup axis for plotting
if not ax:
fig, ax = plt.subplots()
# width per site ranges from 0.02 for xlen <= 100 to
# 0.07 for xlen > 700
xwidth = 0.02 - 0.013 * (min(700, max(100, xlen)) - 100) / (700 - 100)
fig.set_size_inches(
(
widthscale * xwidth * xlen + 0.5 * int(not hide_axis),
heightscale * (2 + 0.5 * int(not hide_axis) + 0.5 * int(bool(title))),
)
)
else:
fig = ax.get_figure()
if title:
ax.set_title(title, fontsize=17 * axisfontscale)
ax.set_xlim(xmin - 0.5 - 0.02 * xlen, xmax + 0.5 + 0.02 * xlen)
ax.set_ylim(ymin, ymax)
if not hide_axis:
xbreaks, xlabels = dmslogo.utils.breaksAndLabels(
data[x_col].tolist(), data[xtick_col].tolist(), max(4, xlen // 50)
)
ax.set_xticks(xbreaks)
ax.tick_params(length=5, width=1)
ax.set_xticklabels(xlabels, rotation=90, ha="center", va="top")
ax.yaxis.set_major_locator(matplotlib.ticker.MaxNLocator(4))
ax.tick_params("both", labelsize=12 * axisfontscale)
ax.set_xlabel(xlabel, fontsize=17 * axisfontscale)
ax.set_ylabel(ylabel, fontsize=17 * axisfontscale)
dmslogo.utils.despine(ax=ax, trim=False, top=True, right=True)
else:
ax.axis("off")
xdata = data[x_col].tolist()
ydata = data[height_col].tolist()
# plot with 0.5 before / after last points so steps full length
ax.step(
[xmin - 0.5] + xdata + [xmax + 0.5],
[ydata[0]] + ydata + [ydata[-1]],
color=color,
where="mid",
linewidth=linewidth,
)
if height_col2 is not None:
ydata2 = data[height_col2].tolist()
ax.step(
[xmin - 0.5] + xdata + [xmax + 0.5],
[ydata2[0]] + ydata2 + [ydata2[-1]],
color=color2,
where="mid",
linewidth=linewidth,
)
if show_col and show_color is not None:
lw_to_xdata = data_units_from_linewidth(linewidth, ax, "x")
lw_to_ydata = data_units_from_linewidth(linewidth, ax, "y")
for x in data.query(f"{show_col}")[x_col].tolist():
ax.add_patch(
plt.Rectangle(
xy=(x - 0.5 - lw_to_xdata, ymin),
width=2 + 1 * lw_to_xdata,
height=(ydata_min - ymin) - lw_to_ydata,
edgecolor="none",
facecolor=show_color,
)
)
return fig, ax
if __name__ == "__main__":
import doctest
doctest.testmod()
