marsilea 0.4.2__py3-none-any.whl → 0.4.4__py3-none-any.whl

marsilea-0.4.4.dist-info/METADATA

@@ -0,0 +1,168 @@
+Metadata-Version: 2.3
+Name: marsilea
+Version: 0.4.4
+Dynamic: Summary
+Project-URL: Home, https://github.com/Marsilea-viz/marsilea
+Author: Zhihang Zheng
+Author-email: Mr-Milk <yzheng@cemm.at>
+License: The MIT License (MIT)
+        
+        Copyright (c) 2024 Mr-Milk
+        
+        Permission is hereby granted, free of charge, to any person obtaining a copy
+        of this software and associated documentation files (the "Software"), to deal
+        in the Software without restriction, including without limitation the rights
+        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+        copies of the Software, and to permit persons to whom the Software is
+        furnished to do so, subject to the following conditions:
+        
+        The above copyright notice and this permission notice shall be included in
+        all copies or substantial portions of the Software.
+        
+        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+        THE SOFTWARE.
+License-File: LICENSE
+Classifier: Framework :: Matplotlib
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3
+Requires-Python: >=3.8
+Requires-Dist: legendkit
+Requires-Dist: matplotlib>=3.6
+Requires-Dist: numpy
+Requires-Dist: pandas
+Requires-Dist: platformdirs
+Requires-Dist: scipy
+Requires-Dist: seaborn
+Provides-Extra: dev
+Requires-Dist: icecream; extra == 'dev'
+Requires-Dist: mpl-fontkit; extra == 'dev'
+Requires-Dist: numpydoc; extra == 'dev'
+Requires-Dist: pydata-sphinx-theme; extra == 'dev'
+Requires-Dist: pytest; extra == 'dev'
+Requires-Dist: python-hmr; extra == 'dev'
+Requires-Dist: ruff; extra == 'dev'
+Requires-Dist: scikit-learn; extra == 'dev'
+Requires-Dist: sphinx; extra == 'dev'
+Requires-Dist: sphinx-copybutton; extra == 'dev'
+Requires-Dist: sphinx-design; extra == 'dev'
+Requires-Dist: sphinx-gallery; extra == 'dev'
+Description-Content-Type: text/markdown
+
+<p align="center">
+  <picture align="center">
+    <source media="(prefers-color-scheme: dark)" srcset="https://github.com/Marsilea-viz/marsilea/raw/main/img/banner-dark.jpg">
+    <source media="(prefers-color-scheme: light)" srcset="https://github.com/Marsilea-viz/marsilea/raw/main/img/banner-blue.jpg">
+    <img alt="Shows a bar chart with benchmark results." src="https://github.com/Marsilea-viz/marsilea/raw/main/img/banner-dark.jpg" width="400">
+  </picture>
+</p>
+
+[![Documentation Status](https://img.shields.io/readthedocs/marsilea?color=57B77E&logo=readthedocs&logoColor=white&style=flat-square)](https://marsilea.readthedocs.io/en/stable)
+![pypi version](https://img.shields.io/pypi/v/marsilea?color=0098FF&logo=python&logoColor=white&style=flat-square)
+![PyPI - License](https://img.shields.io/pypi/l/marsilea?color=FFD43B&style=flat-square)
+
+# Marsilea: Declarative creation of composable visualization!
+
+---
+
+## Documentation
+
+You can read the documentation on Read the Docs.
+
+[Read Documentation](https://marsilea.readthedocs.io/)
+
+## Installation
+
+```shell
+pip install marsilea
+```
+
+## What is Composable Visualization?
+
+<p align="center">
+  <picture align="center">
+    <img alt="Shows a bar chart with benchmark results." src="https://github.com/Marsilea-viz/marsilea/raw/main/img/showcase.gif" width="300">
+  </picture>
+</p>
+
+When we do visualization, we often need to combine multiple plots to show different aspects of the data.
+For example, we may need to create a heatmap to show the expression of genes in different cells,
+and then create a bar chart to show the expression of genes in different cell types.
+A visualization contains multiple plots is called a composable visualization.
+In Marsilea, we employ a declarative approach for user to create composable visualization incrementally.
+
+## Examples
+
+<table>
+    <thead>
+        <tr>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_tiobe_index.html">
+                    Bar Chart With Image
+                </a>
+            </th>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_oil_well.html">
+                    Stacked Bar
+                </a>
+            </th>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_arc_diagram.html">
+                    Arc Diagram
+                </a>
+            </th>
+        </tr>
+    </thead>
+    <tbody>
+        <tr>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_tiobe_index_001_2_00x.png" height="300px">
+            </td>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_oil_well_001_2_00x.png" height="300px">
+            </td>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_arc_diagram_001_2_00x.png" width="300px">
+            </td>
+        </tr>
+    </tbody>
+</table>
+
+<table>
+    <thead>
+        <tr>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_pbmc3k.html">
+                    Heatmap
+                </a>
+            </th>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_oncoprint.html">
+                    Oncoprint
+                </a>
+            </th>
+            <th>
+                <a href="https://marsilea.readthedocs.io/en/latest/examples/Gallery/plot_upset.html">
+                    Upsetplot
+                </a>
+            </th>
+        </tr>
+    </thead>
+    <tbody>
+        <tr>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_pbmc3k_001_2_00x.png" width="300px">
+            </td>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_oncoprint_005_2_00x.png" width="300px">
+            </td>
+            <td>
+                <img src="https://marsilea.readthedocs.io/en/latest/_images/sphx_glr_plot_upset_001_2_00x.png" width="300px">
+            </td>
+        </tr>
+    </tbody>
+</table>

marsilea-0.4.4.dist-info/RECORD

@@ -0,0 +1,30 @@
+marsilea/__init__.py,sha256=N_-2Ce1mWmyFqjfe6217db2yfJ5eZfGWxJjEP8krBX0,541
+marsilea/_api.py,sha256=tymWZHfjhx8-0NNd9762znfdIu36NrARRweEIr5L1mA,283
+marsilea/_deform.py,sha256=QRz4OGXMsQzbiIkC3ASzZayMPhHhoFsEK38oBzSeQG8,14440
+marsilea/base.py,sha256=aj15043lJURGR83T3q8Y8UUpeVHHLsi8T9VLol6KB1I,47552
+marsilea/dataset.py,sha256=a0mXjPu9_tRGHofnnQaTryFpxftkfqldq_ZLXMSBf7A,4410
+marsilea/dendrogram.py,sha256=Ung43zseybZKzTEvH5P_ge3WGfsr7i7qsX7YEVDlC74,15590
+marsilea/exceptions.py,sha256=wN5ElUZxuaJKSnnwWdkNx6P-Oc16dzSuaRPbRKWIBEM,1046
+marsilea/heatmap.py,sha256=UhRcBFhx1JKrtndZPvQCDaiMt8Pb314PT1BtL-JndCY,4208
+marsilea/layers.py,sha256=puXLlGGpEqAzaTqadpgpsYmIDPH33WyyHIuysRSqFZQ,12163
+marsilea/layout.py,sha256=X8MGPlAbbr7dcZiqW4pI7sEb8U3jVaiS7t1DKOqMYLI,27758
+marsilea/upset.py,sha256=U1Rsmo1WpCAV9z3LBlE2L4T0nAW9ols8Z36fXzmXycw,30388
+marsilea/utils.py,sha256=y_KYs4ToiuKEsiBdmcIVtmxMXFpD4wKiJ0k7iBa11z8,2854
+marsilea/plotter/__init__.py,sha256=vtusGk_3B5QKw0k3EMV219FT1U8ZeJse6nMOB_wmQHA,768
+marsilea/plotter/_seaborn.py,sha256=svo1YNenW_EBOKpR-w5suYhCJoL99VrEGzL7Q8-7sX0,8330
+marsilea/plotter/_utils.py,sha256=Efhdk-TrrAanhbXRiEVWThMYvZ4vVHZMVYMs5X3JvIM,710
+marsilea/plotter/arc.py,sha256=44BKVGvDc_OpghfgEvaiVCovZe7_OwZiM20iepaRMFw,8139
+marsilea/plotter/area.py,sha256=zjjAhvgKHYe9rqzcseqZqhwfpgvzm0w2FRJ_vr9Fxm4,2650
+marsilea/plotter/bar.py,sha256=tXM5iR4cTkTrt55ff-aWoh7lxupKtA5vJ9Yn9hr7RkA,12067
+marsilea/plotter/base.py,sha256=TbnZDjm161RF5OlK4TCfhCFR9lM96MJUrvAk_xTk_kQ,20657
+marsilea/plotter/bio.py,sha256=34tucmxs4LM3TFZoGsrjnXTolyrzYaHVEiRe4dzDH68,5040
+marsilea/plotter/images.py,sha256=gb0xIQhUch3rNAt3FfvuUoamSGEynoBBBky2eE754ec,9560
+marsilea/plotter/mesh.py,sha256=eUMXX9PNHJXf9O4wpcRgO6uEFKlpft1LwshHgMuojp8,24176
+marsilea/plotter/text.py,sha256=6S4mnAxLJLMkduKiyor03lPd86oTOJ5TojVREA9oU6s,37466
+oncoprinter/__init__.py,sha256=efshcAD1h9s-NVJj4HLU9-hXc_LtTeIrNYqLHl-sm_g,106
+oncoprinter/core.py,sha256=5KPnKW5ivlxPp14uJd0OtfTv-pXV2UEym8EbII2VCcw,11846
+oncoprinter/preset.py,sha256=mBk2tFCqoTj_1ZZKRYuv4j2I3NTBa6Swc9wjzbmxRVw,8238
+marsilea-0.4.4.dist-info/METADATA,sha256=jA6MqUk97yh7Gapgbgij-DGr_W7260WAY8l_IZJrE6c,6576
+marsilea-0.4.4.dist-info/WHEEL,sha256=1yFddiXMmvYK7QYTqtRNtX66WJ0Mz8PYEiEUoOUUxRY,87
+marsilea-0.4.4.dist-info/licenses/LICENSE,sha256=2TLD8FnLJqXzg8YBRs7W3VZBwfWfp4ArDfBl-rn96Qc,1074
+marsilea-0.4.4.dist-info/RECORD,,

{marsilea-0.4.2.dist-info → marsilea-0.4.4.dist-info}/WHEEL

@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: flit 3.9.0
+Generator: hatchling 1.25.0
 Root-Is-Purelib: true
 Tag: py3-none-any

oncoprinter/__init__.py

@@ -0,0 +1,4 @@
+"""Draw oncoprint in Python"""
+
+from .core import OncoPrint
+from .preset import Rect, FracRect, FrameRect

oncoprinter/core.py

@@ -0,0 +1,360 @@
+import warnings
+from collections import Counter
+from copy import deepcopy
+from dataclasses import dataclass
+from itertools import count
+from typing import Any
+
+import numpy as np
+import pandas as pd
+from marsilea import ClusterBoard
+from marsilea.layers import LayersMesh, FrameRect, Piece
+from marsilea.plotter import Labels, StackBar, Numbers, ColorMesh
+from marsilea.utils import get_canvas_size_by_data
+
+from .preset import SHAPE_BANK, MATCH_POOL, Alteration
+
+
+def guess_alteration(event: str):
+    for alt, rule in MATCH_POOL.items():
+        if rule.is_match(event):
+            return alt
+    return Alteration.OTHER
+
+
+@dataclass(repr=False)
+class LayerData:
+    matrix: np.ndarray
+    piece: Piece
+    color: Any
+    event: Any
+
+    def __repr__(self):
+        return f"{self.piece.label} ({self.color})"
+
+
+class GenomicData:
+    """Handle class for genomics data
+
+    Parameters
+    ----------
+    data : pd.DataFrame
+        Each column is:
+            1) Sample ID
+            2) Track name
+            3) Alteration
+    samples_order : list, optional
+        The order of samples, by default None
+    tracks_order : list, optional
+        The order of tracks, by default None
+    custom_pieces : dict, optional
+        Custom pieces for each alteration, by default None
+
+    """
+
+    def __repr__(self):
+        ntrack, nsample = self.shape
+        return (
+            f"{ntrack} Tracks, {nsample} Samples with "
+            f"{len(self.events)} Alterations"
+        )
+
+    def __init__(
+        self,
+        data,
+        samples_order=None,
+        tracks_order=None,
+        custom_pieces=None,
+    ):
+        self.data = data.copy()
+        self.data.columns = ["sample", "track", "event"]
+
+        if samples_order is None:
+            samples_order = self.data["sample"].unique()
+
+        self.samples = samples_order
+        self._patients_ix = dict(zip(samples_order, count(0, 1)))
+
+        if tracks_order is None:
+            tracks_order = self.data["track"].unique()
+        self.tracks = tracks_order
+        self._tracks_ix = dict(zip(tracks_order, count(0, 1)))
+
+        self._shape = (len(self.tracks), len(self.samples))
+
+        if custom_pieces is None:
+            custom_pieces = {}
+        self.custom_pieces = custom_pieces
+
+        self._process_alterations()
+
+        layers = {}
+        for e in self.events:
+            layers[e] = np.zeros(self._shape, dtype=bool)
+
+        for _, row in self.data.iterrows():
+            patient, track, event = row
+            row_ix = self._tracks_ix[track]
+            col_ix = self._patients_ix[patient]
+            layers[event][row_ix, col_ix] = True
+
+        self.layers = layers
+
+    def _process_alterations(self):
+        events_alt = dict()
+        custom_events = list(self.custom_pieces.keys())
+        raw_events = self.data["event"].unique()
+
+        unknown_alterations = []
+        for e in raw_events:
+            alt = guess_alteration(e)
+            if alt == Alteration.OTHER:
+                alt = e
+                if e not in custom_events:
+                    unknown_alterations.append(e)
+            events_alt[e] = alt
+        self.data["event"] = [events_alt[e] for e in self.data["event"]]
+        self.events = self.data["event"].unique()
+        if len(unknown_alterations) > 0:
+            msg = (
+                f"Found unknown alterations: {unknown_alterations}, "
+                f"please specify a piece for this alteration."
+            )
+            warnings.warn(msg)
+
+    @property
+    def shape(self):
+        return self._shape
+
+    def get_layers_data(self, background_color="#BEBEBE"):
+        # explicitly make copy
+        bg_pieces = deepcopy(SHAPE_BANK[Alteration.BACKGROUND])
+        bg_pieces.background_color = background_color
+
+        # Add background layer
+        layers_data = [
+            LayerData(
+                np.ones(self._shape), bg_pieces, background_color, Alteration.BACKGROUND
+            )
+        ]
+        for alt in self.events:
+            layer = self.layers[alt]
+            if not isinstance(alt, Alteration):
+                piece = self.custom_pieces.get(alt)
+                if piece is None:
+                    # The default style for OTHER
+                    piece = FrameRect(color="pink", label=alt)
+                else:
+                    piece = deepcopy(piece)
+            else:
+                piece = deepcopy(SHAPE_BANK[alt])
+            color = piece.color
+            piece.background_color = background_color
+
+            layers_data.append(LayerData(layer, piece, color, alt))
+
+        return layers_data
+
+    def get_track_mutation_rate(self):
+        gb = self.data.groupby("track", sort=False, observed=True)
+        ts = {}
+        for track, df in gb:
+            ts[track] = len(df["sample"].unique())
+        counts = np.array([ts[t] for t in self.tracks])
+        return counts / len(self.samples)
+
+    def get_track_mutation_types(self):
+        gb = self.data.groupby("track", sort=False, observed=True)
+        cs = {}
+        for track, df in gb:
+            cs[track] = Counter(df["event"])
+        return pd.DataFrame(cs).fillna(0.0)
+
+    def get_sample_mutation_types(self):
+        gb = self.data.groupby("sample", sort=False, observed=True)
+        cs = {}
+        for track, df in gb:
+            cs[track] = Counter(df["event"])
+        return pd.DataFrame(cs).fillna(0.0)
+
+
+class OncoPrint(ClusterBoard):
+    """OncoPrint plot
+
+    The oncoprint plot is a visualization for genomics data in cancer research.
+    It's first introduced by the cBioPortal project.
+    See https://www.cbioportal.org/oncoprinter for more details.
+
+    To use this class, import from oncoprinter
+
+        >>> from oncoprinter import OncoPrint
+
+    Parameters
+    ----------
+    genomic_data : pd.DataFrame
+        Genomics data, each column is:
+            1) Sample ID
+            2) Track name
+            3) Alteration
+
+    patients_order : list, optional
+        The order of samples, by default None
+    tracks_order : list, optional
+        The order of tracks, by default None
+    pieces : dict, optional
+        Custom pieces for each alteration, by default None
+        See :class:`Piece <marsilea.layers.Piece>` for details
+    background_color : str, optional, default: "#BEBEBE"
+        The background color
+    shrink : tuple, optional, default: (0.8, 0.8)
+        The shrink ratio for each layer
+    width, height : float, optional
+        The size in inches to define the size of main canvas
+    aspect : float, optional, default: 2.5
+        The aspect ratio of the main canvas
+    legend_kws : dict, optional
+        The options for legend, by default None
+        See :class:`cat_legend <legendkit.cat_legend>` for details
+    name : str, optional
+        The name of this OncoPrint
+    add_tracks_names : str, optional, default: "left"
+        The position to add tracks names
+        If None, will not add tracks names
+    add_samples_names : str, optional, default: "bottom"
+        The position to add samples names
+        If None, will not add samples names
+    add_mut_perc : str, optional, default: "right"
+        The position to add mutation percentage
+        If None, will not add mutation percentage
+    add_tracks_counts : str, optional, default: "right"
+        The position to add tracks mutation counts
+        If None, will not add tracks mutation counts
+    add_mut_counts : str, optional, default: "top"
+        The position to add mutation counts
+        If None, will not add mutation counts
+    add_tracks_counts_size : float, optional, default: 0.2
+        The size of tracks mutation counts
+    add_tracks_counts_pad : float, optional, default: 0
+        The padding of tracks mutation counts
+    add_mut_counts_size : float, optional, default: 0.2
+        The size of mutation counts
+    add_mut_counts_pad : float, optional, default: 0.1
+        The padding of mutation counts
+
+    """
+
+    def __init__(
+        self,
+        genomic_data=None,
+        patients_order=None,
+        tracks_order=None,
+        pieces=None,
+        background_color="#BEBEBE",
+        shrink=(0.8, 0.8),
+        width=None,
+        height=None,
+        aspect=2.5,
+        legend_kws=None,
+        name=None,
+        add_tracks_names="left",
+        add_samples_names="bottom",
+        add_mut_perc="right",
+        add_tracks_counts="right",
+        add_mut_counts="top",
+        add_tracks_counts_size=0.2,
+        add_tracks_counts_pad=0,
+        add_mut_counts_size=0.2,
+        add_mut_counts_pad=0.1,
+    ):
+        data = GenomicData(
+            genomic_data,
+            samples_order=patients_order,
+            tracks_order=tracks_order,
+            custom_pieces=pieces,
+        )
+        self.genomic_data = data
+        width, height = get_canvas_size_by_data(
+            data.shape, width=width, height=height, scale=0.2, aspect=aspect
+        )
+
+        self.canvas = super().__init__(
+            name=name, cluster_data=np.zeros(data.shape), width=width, height=height
+        )
+
+        legend_options = dict(title="Alterations", handleheight=aspect, handlelength=1)
+        legend_kws = {} if legend_kws is None else legend_kws
+        legend_options.update(legend_kws)
+
+        layers, pieces, colors_mapper = [], [], {}
+        for layer in data.get_layers_data(background_color):
+            layers.append(layer.matrix)
+            pieces.append(layer.piece)
+            colors_mapper[layer.event] = layer.color
+        mesh = LayersMesh(
+            layers=layers, pieces=pieces, shrink=shrink, legend_kws=legend_options
+        )
+        self.add_layer(mesh)
+
+        if add_tracks_names:
+            self.add_plot(add_tracks_names, Labels(data.tracks))
+
+        # Add other statistics
+        track_mut_rate = data.get_track_mutation_rate()
+        # Convert to percentage string
+        if add_mut_perc:
+            rates = [_format_percentage(t) for t in track_mut_rate]
+            self.add_plot(add_mut_perc, Labels(rates))
+        if add_samples_names:
+            self.add_plot(add_samples_names, Labels(data.samples))
+
+        if add_tracks_counts:
+            track_counter = data.get_track_mutation_types()
+            colors = [colors_mapper[e] for e in track_counter.index]
+            track_bar = StackBar(track_counter, colors=colors, show_value=False)
+            self.add_plot(
+                add_tracks_counts,
+                track_bar,
+                legend=False,
+                size=add_tracks_counts_size,
+                pad=add_tracks_counts_pad,
+            )
+
+        if add_mut_counts:
+            patients_counter = data.get_sample_mutation_types()
+            colors = [colors_mapper[e] for e in patients_counter.index]
+            patients_bar = StackBar(patients_counter, colors=colors, show_value=False)
+            self.add_plot(
+                add_mut_counts,
+                patients_bar,
+                legend=False,
+                size=add_mut_counts_size,
+                pad=add_mut_counts_pad,
+            )
+        self.add_legends()
+
+    clinical_plots = {
+        "bar": Numbers,
+        "stack_bar": StackBar,
+    }
+
+    def add_clinical_data(self, data, plot="bar", size=None, pad=0.1, **kwargs):
+        data = data.loc[self.samples_order]
+        plotter = self.clinical_plots[plot]
+        self.add_bottom(plotter(data, **kwargs), size=size, pad=pad)
+
+    def add_heatmap_data(self, data, size=0.2, pad=0.1, **kwargs):
+        data = data.loc[self.samples_order]
+        options = {"cmap": "Blues", "label_loc": "left", **kwargs}
+        self.add_bottom(ColorMesh(data, **options), size=size, pad=pad)
+
+    @property
+    def samples_order(self):
+        return self.genomic_data.samples
+
+    @property
+    def tracks_order(self):
+        return self.genomic_data.tracks
+
+
+def _format_percentage(t):
+    return f"{float(t) * 100:.2f}".rstrip("0").rstrip(".") + "%"