scallops.visualize.crosstalk.pairwise_channel_scatter_plot

scallops.visualize.crosstalk.pairwise_channel_scatter_plot(bases, plot_func=None, q=None, col_wrap=3, intensity_column='intensity', **plot_args)

Pairwise channel intensity scatter plot.

This function creates a scatter plot for pairwise combinations of channels in the provided DataArray.

Parameters:

• bases (DataArray | DataFrame) – The input array containing intensity values for different channels across samples and time points.

• plot_func (Callable[[ndarray, ndarray, int, int, Axes, Any], None]) – Function to plot x-y scatter plot that accepts channel i values, channel j values, i, j, axis, and plot keyword arguments (e.g. markersize).

• q (tuple[float, ...]) – Tuple of lower and upper quantiles to include for Li and Speed fit. If None, do not perform fit.

• col_wrap (int) – Wrap the plot at this width, so that the plot spans multiple rows.

• intensity_column (str) – The name of the column in bases containing the intensity when bases is data frame.

• plot_args – Additional arguments to pass to the plot function (e.g. markersize)

Returns:

The figure and axes object with the plot drawn onto it.

Examples:

1. Generate a pairwise channel scatter plot

   import xarray as xr
   import matplotlib.pyplot as plt
   from scallops.visualize import pairwise_channel_scatter_plot
   
   # Create a sample DataArray
   channels = ["ChannelA", "ChannelT", "ChannelG", "ChannelC"]
   data = xr.DataArray(
       np.random.rand(100, 10, len(channels)),
       dims=("read", "t", "c"),
       coords={"c": channels},
   )
   
   # Generate a pairwise channel scatter plot
   fig, axes = pairwise_channel_scatter_plot(
       data, q=(0.1, 0.9), col_wrap=2
   )
   
   # Show the plot
   plt.show()
   

2. Generate a pairwise channel scatter plot with colors by their base calls

   import xarray as xr
   import matplotlib.pyplot as plt
   import matplotlib.patches as mpatches
   from scallops.visualize import pairwise_channel_scatter_plot
   
   # Create a sample DataArray
   channels = ["ChannelA", "ChannelT", "ChannelG", "ChannelC"]
   data = xr.DataArray(
       np.random.rand(100, 10, len(channels)),
       dims=("read", "t", "c"),
       coords={"c": channels},
   )
   
   bases = ["G", "T", "A", "C"]
   base_calls = pd.Series(np.random.choice(bases, size=100))
   base_colors = {"G": "green", "T": "red", "A": "magenta", "C": "cyan"}
   
   
   def plot_func(x, y, i, j, ax, **plot_args):
       ax.scatter(
           x, y, c=base_calls.apply(lambda x: base_colors[x]), **plot_args
       )
   
   
   fig, axes = pairwise_channel_scatter_plot(
       data.isel(t=0), plot_func=plot_func
   )
   patches = [
       mpatches.Patch(color=color, label=label)
       for label, color in base_colors.items()
   ]
   plt.legend(handles=patches)
   plt.show()
   

Return type:

tuple[Figure, Axes]