scez 0.0.1__py2.py3-none-any.whl

scez/__init__.py

@@ -0,0 +1,36 @@
+"""scez – Single Cell Analysis, Easy Mode!"""
+
+from . import diffexp as de
+from . import preprocess as pp
+from . import representation as rp
+from . import utils
+import scanpy as sc
+import matplotlib.pyplot as plt
+
+import tomli
+
+toml_dict = tomli.load(open('pyproject.toml','rb'))
+__version__ = toml_dict['tool']['poetry']['version']
+
+
+sc.settings.verbosity = 1             # verbosity: errors (0), warnings (1), info (2), hints (3)
+sc.settings.set_figure_params(dpi=100, dpi_save=300, frameon=False, figsize=(5, 5), facecolor='white')
+sc.logging.print_header()
+
+# https://stackoverflow.com/questions/21884271/warning-about-too-many-open-figures
+plt.rcParams.update({'figure.max_open_warning': 0})
+plt.close('all')
+
+# https://stackoverflow.com/questions/3899980/how-to-change-the-font-size-on-a-matplotlib-plot
+
+SMALL_SIZE = 6
+MEDIUM_SIZE = 8
+BIGGER_SIZE = 10
+
+plt.rc('font', size=SMALL_SIZE)          # controls default text sizes
+plt.rc('axes', titlesize=SMALL_SIZE)     # font size of the axes title
+plt.rc('axes', labelsize=MEDIUM_SIZE)    # font size of the x and y labels
+plt.rc('xtick', labelsize=SMALL_SIZE)    # font size of the tick labels
+plt.rc('ytick', labelsize=SMALL_SIZE)    # font size of the tick labels
+plt.rc('legend', fontsize=SMALL_SIZE)    # legend font size
+plt.rc('figure', titlesize=BIGGER_SIZE)  # font size of the figure title

scez/diffexp.py

@@ -0,0 +1,199 @@
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import seaborn as sns
+import anndata as ad
+
+from pydeseq2.dds import DeseqDataSet
+from pydeseq2.default_inference import DefaultInference
+from pydeseq2.ds import DeseqStats
+from .utils import run_adjust_text
+from adpbulk import ADPBulk
+
+
+def pseudobulk_by_clusters(adt, condition, cluster_col='leiden', method="mean"):
+    # initialize the object
+    adpb = ADPBulk(adt, [cluster_col, condition], method=method)
+
+    # perform the pseudobulking
+    pseudobulk_matrix = adpb.fit_transform()
+
+    # retrieve the sample metadata (useful for easy incorporation with edgeR)
+    sample_meta = adpb.get_meta()
+
+    out = ad.AnnData(
+        X=pseudobulk_matrix,
+        obs=sample_meta.set_index('SampleName')
+    )
+
+    return out
+
+
+def run_deseq(adata, design, tested_level, ref_level, n_cpus=8):
+
+    inference = DefaultInference(n_cpus=n_cpus)
+    
+    dds = DeseqDataSet(
+        counts=adata.to_df().astype(int),
+        metadata=adata.obs,
+        design_factors=design,  # compare samples based on the "condition"
+        refit_cooks=True,
+        inference=inference,
+    )
+
+    dds.deseq2()
+
+    stat_res = DeseqStats(
+        dds, 
+        contrast=[design, tested_level, ref_level], 
+        inference=inference
+    )
+    stat_res.summary()
+
+    df = stat_res.results_df
+
+    return df
+
+
+def plot_volcano(df, title=None, labels=None, n_genes=False, side='both', 
+                 font_scale=1, dot_size = 5,
+                 color = '#1f77b4', color_highlight = '#FFA500',
+                 ax = None, **kwargs):
+    dot_size_highlight = dot_size * 1.1
+    annotate_font_size = 5 * font_scale
+    scatter_font_size = 8 * font_scale
+    label_font_size = 9 * font_scale
+    title_font_size = 10 * font_scale
+
+    if 'name' not in df.columns: df['name'] = df.index.to_list()
+    df['-log10(pvalue)'] = - np.log10(df.pvalue)
+
+    if not ax: fig, ax = plt.subplots(figsize=(3, 3))
+
+    # Scatter plot
+    ax.scatter(
+        df['log2FoldChange'],
+        df['-log10(pvalue)'],
+        alpha=0.9, s=dot_size, c=color,
+        **kwargs
+    )
+
+    # Set background color to transparent
+    ax.set_facecolor('none')
+
+    # Set smaller font size
+    ax.tick_params(axis='both', which='both', labelsize=scatter_font_size)
+
+    # Set labels
+    ax.set_xlabel('log2FoldChange', fontsize=label_font_size)
+    ax.set_ylabel('-log10(pvalue)', fontsize=label_font_size)
+
+    # Set plot title
+    if not title:
+        ax.set_title('Volcano Plot', fontsize=title_font_size)
+    else:
+        ax.set_title(title, fontsize=title_font_size)
+
+    ax.grid(False)
+
+    # check if `labels` is provided or set that based on `n_genes` and `side`
+    if labels and n_genes:
+        # error message if both labels and n_genes are provided and say one of them is allowed
+        raise ValueError('Provide either labels or n_genes, not both!')
+
+    elif n_genes and side == 'positive':
+        # Highlight top genes
+        top_genes = df.query('log2FoldChange > 0').nlargest(n_genes, '-log10(pvalue)')
+        labels = [row['name'] for _, row in top_genes.iterrows()]
+
+    elif n_genes and side == 'negative':
+        # Highlight top genes
+        top_genes = df.query('log2FoldChange < 0').nlargest(n_genes, '-log10(pvalue)')
+        labels = [row['name'] for _, row in top_genes.iterrows()]
+
+    elif n_genes and side == 'both':
+        # Highlight top genes
+        top_genes = df.nlargest(n_genes, '-log10(pvalue)')
+        labels = [row['name'] for _, row in top_genes.iterrows()]
+
+    # Highlight the points from given labels
+    if labels:
+        for label in labels:
+            ax.scatter(
+                df.loc[label, 'log2FoldChange'],
+                df.loc[label, '-log10(pvalue)'],
+                s=dot_size_highlight, c=color_highlight
+            )
+        run_adjust_text(
+            df.loc[labels, 'log2FoldChange'], 
+            df.loc[labels, '-log10(pvalue)'], 
+            labels, 
+            font_size=annotate_font_size, ax=ax, use_arrow=False
+        )
+
+    if not ax: 
+        plt.tight_layout()
+        plt.show()
+
+
+def plot_top_DEG_violinplot(adata, df, title=None, labels=None, n_genes=False, side='both', font_scale=1, figsize=(10, 4), **kwargs):
+    label_font_size = 9 * font_scale
+    title_font_size = 10 * font_scale
+
+    if 'name' not in df.columns: df['name'] = df.index.to_list()
+
+    if labels and n_genes:
+        # error message if both labels and n_genes are provided and say one of them is allowed
+        raise ValueError('Provide either labels or n_genes, not both!')
+
+    if not labels and not n_genes:
+        # error message if neither labels nor n_genes are provided
+        raise ValueError('Provide either labels or n_genes!')
+
+    if labels:
+        # Highlight the points from given list
+        selected_genes = df.loc[labels]
+
+    elif n_genes and side == 'positive':
+        # Highlight top genes
+        selected_genes = df.query('log2FoldChange > 0').nlargest(n_genes, '-log10(pvalue)')
+
+    elif n_genes and side == 'negative':
+        # Highlight top genes
+        selected_genes = df.query('log2FoldChange < 0').nlargest(n_genes, '-log10(pvalue)')
+
+    elif n_genes and side == 'both':
+        # Highlight top genes
+        selected_genes = df.nlargest(n_genes, '-log10(pvalue)')
+
+    # Filter the single-cell dataset for the selected genes
+    subset_adata = adata[:, selected_genes.index]
+    subset_adata.var.index = subset_adata.var.index.str.split('_').str[0]
+
+    # Convert the subset of adata to a DataFrame
+    subset_df = subset_adata.to_df()
+
+    # Merge the DataFrame with .obs to include the 'sample' information
+    merged_df = pd.merge(subset_df, adata.obs[['sample']], left_index=True, right_index=True)
+
+    # Melt the DataFrame to prepare for violin plot
+    melted_df = pd.melt(merged_df, id_vars='sample', var_name='Gene', value_name='Counts')
+
+    # Create a violin plot
+    plt.figure(figsize=figsize)
+    sns.violinplot(x='Gene', y='Counts', hue='sample', data=melted_df, split=True, inner='quartile', palette='Set2', **kwargs)
+    sns.stripplot(x='Gene', y='Counts', hue='sample', data=melted_df, dodge=True, jitter=True, color='black', size=1, alpha=0.3, **kwargs)
+
+    plt.xticks(rotation=45, ha='right', fontsize=label_font_size)
+    plt.legend(bbox_to_anchor=(1.05, 1), loc='upper left', fontsize=label_font_size)
+
+    if not title:
+        plt.title('Top Differentially Expressed Genes', fontsize=title_font_size)
+    else:
+        plt.title(title, fontsize=title_font_size)
+    plt.show()
+
+
+def write_top_DEGs(df, sample_id, result_dir='.', n_hits=200):
+    df['-log10(pvalue)'] = - np.log10(df.pvalue)
+    df.nlargest(n_hits, '-log10(pvalue)').to_csv(f'{result_dir}/{sample_id}_top_{n_hits}.csv')  # Adjust the number as needed

scez/preprocess.py

@@ -0,0 +1,74 @@
+import pandas as pd
+import scanpy as sc
+import scar
+
+
+def normalization(adata, target_sum=1e4, max_value=10, final_layer='scaled', keep_initial_layer=True):
+    if keep_initial_layer == True:
+        adata.layers['raw_counts'] = adata.X.copy()
+    elif type(keep_initial_layer) == str:
+        adata.layers[keep_initial_layer] = adata.X.copy()
+    
+    # normalize counts to target_sum (default 1e4)
+    counts = sc.pp.normalize_total(adata, target_sum=target_sum, inplace=False)
+    # log1p transform
+    adata.layers["log1p_norm"] = sc.pp.log1p(counts["X"], copy=True)
+    # scale counts
+    adata.layers['scaled'] = sc.pp.scale(adata, max_value=max_value, copy=True).X
+    # set the final layer
+    adata.X = adata.layers[final_layer]
+
+
+def remove_ambient_rna(adata_filtered_feature_bc, adata_raw_feature_bc):
+    scar.setup_anndata(
+        adata = adata_filtered_feature_bc,
+        raw_adata = adata_raw_feature_bc,
+        prob = 0.995,
+        kneeplot = True
+    )
+
+    adata_scar = scar.model(
+        raw_count=adata_filtered_feature_bc.to_df(), # In the case of Anndata object, scar will automatically use the estimated ambient_profile present in adata.uns.
+        # ambient_profile=adata_filtered_feature_bc.uns['ambient_profile_Gene Expression'],
+        feature_type='mRNA',
+        sparsity=1,
+        # device=device # Both cpu and cuda are supported.
+    )
+
+    adata_scar.train(
+        epochs=200,
+        batch_size=64,
+        verbose=True
+    )
+
+    # After training, we can infer the native true signal
+    adata_scar.inference(batch_size=256)  # by defaut, batch_size = None, set a batch_size if getting a memory issue
+
+    denoised_count = pd.DataFrame(
+        adata_scar.native_counts, 
+        index=adata_filtered_feature_bc.obs_names, 
+        columns=adata_filtered_feature_bc.var_names
+    )
+
+    adata = adata_filtered_feature_bc.copy()
+    adata.layers['raw_counts'] = adata.X
+    adata.layers['scar_denoised_counts'] = denoised_count.to_numpy()
+
+    return adata
+
+
+def clustering(
+        adata
+        ):
+    pass
+    # , n_pcs=50, n_neighbors=30, use_highly_variable='Yes',
+    #     use_rep=None, resolution=None
+    
+    # if use_highly_variable == 'Yes':
+    #     sc.pp.highly_variable_genes(adata, min_mean=0.0125, max_mean=3, min_disp=0.5)
+    #     sc.tl.pca(adata, svd_solver='arpack', use_highly_variable=True)
+    # else:
+    #     sc.pp.pca(adata, n_comps=n_pcs)
+    # sc.pp.neighbors(adata, use_rep=use_rep, n_neighbors=n_neighbors)#, n_pcs=n_pcs)
+    # sc.tl.umap(adata)
+    # sc.tl.leiden(adata, resolution=resolution)

scez/representation.py

@@ -0,0 +1,44 @@
+from itertools import product
+import matplotlib.pyplot as plt
+import scanpy as sc
+import numpy as np
+
+
+def optimising_umap_layout(adata, cluster_key='leiden',MIN_DISTS = [0.1, 1, 2], SPREADS = [0.5, 1, 5]):
+    # https://scanpy-tutorials.readthedocs.io/en/latest/plotting/advanced.html
+    # Copy adata not to modify UMAP in the original adata object
+    adata_temp = adata.copy()
+    
+    # Create grid of plots, with a little extra room for the legends
+    fig, axes = plt.subplots(
+        len(MIN_DISTS), len(SPREADS), figsize=(len(SPREADS) * 3 + 2, len(MIN_DISTS) * 3)
+    )
+
+    # Loop through different umap parameters, recomputting and replotting UMAP for each of them
+    for (i, min_dist), (j, spread) in product(enumerate(MIN_DISTS), enumerate(SPREADS)):
+        ax = axes[i][j]
+        param_str = " ".join(["min_dist =", str(min_dist), "and spread =", str(spread)])
+        # Recompute UMAP with new parameters
+        sc.tl.umap(adata_temp, min_dist=min_dist, spread=spread)
+        # Create plot, placing it in grid
+        sc.pl.umap(
+            adata_temp,
+            color=[cluster_key],
+            title=param_str,
+            s=40,
+            ax=ax,
+            show=False,
+        )
+    plt.tight_layout()
+    plt.show()
+    plt.close()
+    del adata_temp
+
+
+def random_ordering(adata):
+    # Randomly order cells by making a random index and subsetting AnnData based on it
+    # Set a random seed to ensure that the cell ordering will be reproducible
+    np.random.seed(0)
+    random_indices = np.random.permutation(list(range(adata.shape[0])))
+
+    return random_indices

scez/tests/test_scez.py

@@ -0,0 +1,18 @@
+import unittest
+import matplotlib.pyplot as plt
+import scanpy as sc
+import scez
+import tomli
+
+toml_dict = tomli.load(open('pyproject.toml','rb'))
+version = toml_dict['tool']['poetry']['version']
+
+class TestScezConfig(unittest.TestCase):
+    def test_version(self):
+        self.assertEqual(scez.__version__, version)
+
+    def test_scanpy_settings(self):
+        self.assertEqual(sc.settings.verbosity, 1)
+
+if __name__ == '__main__':
+    unittest.main()

scez/utils.py

@@ -0,0 +1,64 @@
+import pandas as pd
+from matplotlib import pyplot as plt
+from adjustText import adjust_text
+
+
+def rank_genes_to_df(adata, n=50):
+    result = adata.uns['rank_genes_groups']
+
+    groups = result['names'].dtype.names
+
+    df = pd.DataFrame(
+        {group + '_' + key: result[key][group]
+         for group in groups for key in ['names', 'scores']}).head(n)
+
+    return df
+
+
+def add_marker_feature(adata, marker, marker_name, clusters_name, thr = 0, figsize=(10, 4)):
+
+    adata.obs[marker_name] = ''
+    adata.obs.loc[adata.to_df().loc[:,marker] <= thr, marker_name] = f'{marker}-'
+    adata.obs.loc[adata.to_df().loc[:,marker] > thr, marker_name] = f'{marker}+'
+
+    df = pd.concat([
+        adata.obs.groupby([marker_name,clusters_name]).size()[f'{marker}+'],
+        adata.obs.groupby([marker_name,clusters_name]).size()[f'{marker}-']
+    ],axis=1).rename(columns={0:f'{marker}+',1:f'{marker}-'})
+
+    # Make some labels.
+    labels = df[f'{marker}+'] / df.sum(axis=1) * 100
+    labels = labels.round(decimals=1)
+    labels.sort_values(ascending=False,inplace=True)
+    df = df.loc[labels.index,]
+
+    ax = df.plot.bar(stacked=True,rot=0,figsize=figsize)
+
+    rects = ax.patches
+
+    for rect, label in zip(rects, labels):
+        height = rect.get_height()
+        ax.text(
+            rect.get_x() + rect.get_width() / 2, height + 5, str(label) + "%",
+            ha="center", va="bottom", fontsize=8
+        )
+
+    ax.set_yscale('log')
+    ax.set_ylabel('# of cells')
+    return ax
+
+
+def run_adjust_text(x, y, labels, ax=None, use_arrow=True, font_weight='bold', font_size=8):
+    texts = [
+        plt.text(
+            x[i], y[i], 
+            labels[i],
+            fontdict={'weight': font_weight, 'size': font_size},
+            ha='center', va='center'
+        ) for i in range(len(x))
+    ]
+    
+    if use_arrow:
+        adjust_text(texts, arrowprops=dict(arrowstyle='->', color='red'), ax = ax)
+    else:
+        adjust_text(texts, ax = ax)

scez-0.0.1.dist-info/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 Abolfazl (Abe)
+
+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.

scez-0.0.1.dist-info/METADATA

@@ -0,0 +1,48 @@
+Metadata-Version: 2.1
+Name: scez
+Version: 0.0.1
+Summary: Single Cell Analysis, Easy Mode!
+License: MIT
+Author: Abe Arab
+Author-email: abarbiology@gmail.com
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 2
+Classifier: Programming Language :: Python :: 2.7
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.4
+Classifier: Programming Language :: Python :: 3.5
+Classifier: Programming Language :: Python :: 3.6
+Classifier: Programming Language :: Python :: 3.7
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Project-URL: Source, https://github.com/abearab/scez
+Description-Content-Type: text/markdown
+
+## scez – single cell, easy mode
+[![package](https://github.com/abearab/scez/actions/workflows/main.yml/badge.svg)](https://github.com/abearab/scez/actions/workflows/main.yml)
+
+### Installation
+Make sure you have mamba installed in your base environment. If not, install it with:
+```bash
+conda install mamba -n base -c conda-forge
+```
+Then, create a new conda environment with the provided `environment.yml` file and activate it. This will install all necessary dependencies for scez.
+```bash
+conda env create -f environment.yml
+
+conda activate scez
+```
+Finally, install scez with:
+
+```bash
+pip install scez
+```
+
+Or, if you want to install the latest version from the repository:
+```bash
+pip install git+https://github.com/abearab/scez.git
+```
+

scez-0.0.1.dist-info/RECORD

@@ -0,0 +1,11 @@
+scez/__init__.py,sha256=kmN55oxZUHJfsInsnDWvAtxSNn5KnESir65NGr_ssyo,1356
+scez/diffexp.py,sha256=V8qjqpTZVaoAb-iCa3jYmPDmDBtg5AK9OqNzRow6ic0,6770
+scez/preprocess.py,sha256=3iaxACkmKQjKZdaFBLRAXz1kwkNnhbdnd_gwF8rI4nw,2572
+scez/representation.py,sha256=IW0pwa_yoKf-2a3lbtxRwLTdCjB2PkS9oN4WGCLwED0,1596
+scez/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+scez/tests/test_scez.py,sha256=hoBSYB5k1WtF1hY_kyWXw1W7NolBornmTdAng6tDpjw,452
+scez/utils.py,sha256=lPmt01JILiKoBUhhxZeyhl2CfAzailTp4dIefP4FqGQ,1955
+scez-0.0.1.dist-info/LICENSE,sha256=59TS1D5RmGh66RJikXvFoq_ZW9pGirB4zUOUXLfptJM,1071
+scez-0.0.1.dist-info/METADATA,sha256=H5YLkU2mzh80sL9WrXsNX740nA7tTt4NWQFIHi93Kio,1647
+scez-0.0.1.dist-info/WHEEL,sha256=IrRNNNJ-uuL1ggO5qMvT1GGhQVdQU54d6ZpYqEZfEWo,92
+scez-0.0.1.dist-info/RECORD,,

scez-0.0.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: poetry-core 1.9.0
+Root-Is-Purelib: true
+Tag: py2.py3-none-any