Skip to content

markerCorrelation

Short Description

The sm.pl.markerCorrelation function computes and visualizes the correlation among selected markers (genes, proteins, etc.) within an AnnData object.

Function

markerCorrelation(adata, layer='log', subsetMarkers=None, orderRow=None, orderColumn=None, clusterRows=True, clusterColumns=True, cmap='vlag', figsize=None, overlayValues=False, fontSize=10, fontColor='black', fileName='markerCorrelation.pdf', saveDir=None, **kwargs)

Parameters:

Name Type Description Default
adata AnnData or str

An AnnData object containing the dataset, or a string path to an AnnData file to be loaded.

 required
layer str

Specifies the layer of adata to use for the heatmap. If None, the .X attribute is used. If you want to plot the raw data use raw

 'log'
subsetMarkers list of str

A list of marker names to include in the correlation analysis. If None, all markers are used.

 None
orderRow list of str

Specifies a custom order for the rows (markers) based on their names.

 None
orderColumn list of str

Specifies a custom order for the columns (markers) based on their names.

 None
clusterRows bool

Whether to apply hierarchical clustering to rows.

 True
clusterColumns bool

Whether to apply hierarchical clustering to columns.

 True
cmap str

The colormap for the heatmap.

 'vlag'
figsize tuple of float

The size of the figure to create. If None, the size is inferred based on the data.

 None
overlayValues bool

If True, overlays the actual correlation values on the heatmap.

 False
fontSize int

Font size for the overlay values.

 10
fontColor str

Color of the font used for overlay values.

 'black'
fileName str

Name of the file to save the heatmap. Relevant only if saveDir is not None.

 'markerCorrelation.pdf'
saveDir str

Directory to save the generated heatmap. If None, the heatmap is not saved.

 None

Returns:

Name Type Description
plot matplotlib

Displays or saves a heatmap visualizing the correlation between specified markers.
Example 
1
2
3
4
5
6
7
8
    # Example 1: Basic usage with all markers and default parameters
    sm.pl.markerCorrelation(adata)

    # Example 2: With subset of markers, custom clustering, and overlaying correlation values
    sm.pl.markerCorrelation(adata, subsetMarkers=['Marker1', 'Marker2', 'Marker3'], clusterRows=False, overlayValues=True, fontSize=12)

    # Example 3: Saving the heatmap to a specific directory
    sm.pl.markerCorrelation(adata, fileName='myHeatmap.pdf', saveDir='/path/to/save')
Source code in scimap/plotting/markerCorrelation.py 
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
def markerCorrelation(
    adata,
    layer='log',
    subsetMarkers=None,
    orderRow=None,
    orderColumn=None,
    clusterRows=True,
    clusterColumns=True,
    cmap='vlag',
    figsize=None,
    overlayValues=False,
    fontSize=10,
    fontColor='black',
    fileName='markerCorrelation.pdf',
    saveDir=None,
    **kwargs,
):
    """
    Parameters:
            adata (AnnData or str):
                An AnnData object containing the dataset, or a string path to an AnnData file to be loaded.

            layer (str, optional):
                Specifies the layer of `adata` to use for the heatmap. If None, the `.X` attribute is used. If you want to plot the raw data use `raw`

            subsetMarkers (list of str, optional):
                A list of marker names to include in the correlation analysis. If None, all markers are used.

            orderRow (list of str, optional):
                Specifies a custom order for the rows (markers) based on their names.

            orderColumn (list of str, optional):
                Specifies a custom order for the columns (markers) based on their names.

            clusterRows (bool, optional):
                Whether to apply hierarchical clustering to rows.

            clusterColumns (bool, optional):
                Whether to apply hierarchical clustering to columns.

            cmap (str, optional):
                The colormap for the heatmap.

            figsize (tuple of float, optional):
                The size of the figure to create. If None, the size is inferred based on the data.

            overlayValues (bool, optional):
                If True, overlays the actual correlation values on the heatmap.

            fontSize (int, optional):
                Font size for the overlay values.

            fontColor (str, optional):
                Color of the font used for overlay values.

            fileName (str, optional):
                Name of the file to save the heatmap. Relevant only if `saveDir` is not None.

            saveDir (str, optional):
                Directory to save the generated heatmap. If None, the heatmap is not saved.

    Returns:
            plot (matplotlib):
                Displays or saves a heatmap visualizing the correlation between specified markers.

    Example:
        ```python

            # Example 1: Basic usage with all markers and default parameters
            sm.pl.markerCorrelation(adata)

            # Example 2: With subset of markers, custom clustering, and overlaying correlation values
            sm.pl.markerCorrelation(adata, subsetMarkers=['Marker1', 'Marker2', 'Marker3'], clusterRows=False, overlayValues=True, fontSize=12)

            # Example 3: Saving the heatmap to a specific directory
            sm.pl.markerCorrelation(adata, fileName='myHeatmap.pdf', saveDir='/path/to/save')

        ```
    """

    # load adata
    if isinstance(adata, str):
        adata = ad.read_h5ad(adata)

    # subset the markers if user requests
    if subsetMarkers:
        subsetMarkers = (
            [subsetMarkers] if isinstance(subsetMarkers, str) else subsetMarkers
        )  # convert to list
        # isolate the data
        if layer == 'raw':
            matrix = adata[:, subsetMarkers].raw.X
        elif layer is None:
            matrix = adata[:, subsetMarkers].X
        else:
            matrix = adata[:, subsetMarkers].layers[layer]
    else:
        # take the whole data if the user does not subset anything
        if layer == 'raw':
            matrix = adata.raw.X
        elif layer is None:
            matrix = adata.X
        else:
            matrix = adata.layers[layer]

    # intialize the markers to be plotted
    if subsetMarkers is None:
        var_names = adata.var_names.tolist()
    else:
        var_names = subsetMarkers

    # run correlation
    corr_matrix = np.corrcoef(matrix.T)

    row_order = np.arange(corr_matrix.shape[0])
    col_order = np.arange(corr_matrix.shape[1])

    if orderRow:
        if clusterRows:
            warnings.warn(
                "Both orderRow and clusterRows were provided. Proceeding with orderRow and ignoring clusterRows."
            )
            clusterRows = False
        row_order = [var_names.index(name) for name in orderRow]

    if orderColumn:
        if clusterColumns:
            warnings.warn(
                "Both orderColumn and clusterColumns were provided. Proceeding with orderColumn and ignoring clusterColumns."
            )
            clusterColumns = False
        col_order = [var_names.index(name) for name in orderColumn]

    corr_matrix = corr_matrix[np.ix_(row_order, col_order)]

    if clusterRows:
        linkage_row = linkage(pdist(corr_matrix), method='average')
        row_order = dendrogram(linkage_row, no_plot=True)['leaves']
        corr_matrix = corr_matrix[row_order, :]

    if clusterColumns:
        linkage_col = linkage(pdist(corr_matrix.T), method='average')
        col_order = dendrogram(linkage_col, no_plot=True)['leaves']
        corr_matrix = corr_matrix[:, col_order]

    if figsize is None:
        base_size = 0.5  # Base size for each cell in inches
        figsize_width = max(10, len(corr_matrix) * base_size)
        figsize_height = max(8, len(corr_matrix) * base_size)
        figsize = (figsize_width, figsize_height)

    plt.figure(figsize=figsize)
    im = plt.imshow(corr_matrix, cmap=cmap, aspect='auto', **kwargs)
    plt.colorbar(im)

    if overlayValues:
        for i in range(corr_matrix.shape[0]):
            for j in range(corr_matrix.shape[1]):
                text = plt.text(
                    j,
                    i,
                    f"{corr_matrix[i, j]:.2f}",
                    ha="center",
                    va="center",
                    color=fontColor,
                    fontsize=fontSize,
                )

    plt.xticks(
        ticks=np.arange(len(col_order)),
        labels=np.array(var_names)[col_order],
        rotation=90,
    )
    plt.yticks(ticks=np.arange(len(row_order)), labels=np.array(var_names)[row_order])
    plt.tight_layout()

    # Saving the figure if saveDir and fileName are provided
    if saveDir:
        if not os.path.exists(saveDir):
            os.makedirs(saveDir)
        full_path = os.path.join(saveDir, fileName)
        plt.savefig(full_path, dpi=300)
        plt.close()
        print(f"Saved plot to {full_path}")
    else:
        plt.show()