gene-platform-utils

Python helper functions for the gene-platform to use in the browser via Piodide.

Directory Structure

.
├── gene_platform_utils/
│   ├── __init__.py
│   ├── generate_html_3d_embedding.py
│   ├── plot_de.py
│   ├── plot_eda.py
│   ├── plot_gsea.py
│   └── transformation.py
├── README.md
├── pyproject.toml
└── requirements.txt

Modules

transformation

This module provides functions for variance stabilizing transformation (VST) and other data transformations.

Functions

• vst(counts: np.ndarray, min_mu: float = 0.5, min_disp: float = 1e-8, max_disp: float = 10.0, beta_tol: float = 1e-8, subset_n_genes: Optional[int] = 1200, rand_seed: Optional[int] = 123) -> NDArray

  Performs variance stabilizing transformation on the input counts matrix.

• log2_1p(counts: np.ndarray) -> np.ndarray

  Applies log2(x + 1) transformation.

• ln_1p(counts: np.ndarray) -> np.ndarray

  Applies natural log(x + 1) transformation.

• log10_1p(counts: np.ndarray) -> np.ndarray

  Applies log10(x + 1) transformation.

plot_de

This module provides functions for creating differential expression plots.

Functions

• create_volcano_plot(data: np.ndarray, row_names: list[str], column_names: list[str], lfc_thresh: float, pval_thresh: float, cohort_name: str) -> str

  Creates an interactive volcano plot using Plotly.

• create_mean_difference_plot(data: np.ndarray, row_names: list[str], column_names: list[str], fdr: float, cohort_name: str) -> str

  Creates an interactive mean difference plot using Plotly.

plot_eda

This module provides functions for creating exploratory data analysis plots.

Functions

• calculate_correlation(counts: np.ndarray) -> np.ndarray

  Calculates the sample-to-sample correlation matrix.

• create_correlation_heatmap(counts: np.ndarray, sample_ids: list[str]) -> str

  Creates an interactive heatmap of sample-to-sample correlation using Plotly.

• create_pca_plot(counts, sample_ids)

  Creates a 3D PCA plot.

• create_tsne_plot(counts, sample_ids)

  Creates a 3D t-SNE plot.

plot_gsea

This module provides functions for creating gene set enrichment analysis plots.

Functions

• gene_concept_network_plotgsea_res: dict[str, dict[str, Union[str, float]]], de_res: dict[str, dict[str, Union[str, float]]], ensembl_to_symbol: dict[str, str], color_metric: str = "log2FoldChange", pvalue_threshold: float = 0.05, layout_seed: int = 0, color_seed: int = 0) -> str

  Creates an interactive gene concept network plot using Plotly.

generate_html_3d_embedding

This module provides functions for generating HTML content for 3D embeddings.

Functions

• generate_scattergl_html(title1: str, title2: str, data_csv_content: str, metadata_csv_content: str) -> str

  Generates HTML content for a 3D scatter plot using Plotly.

Dependencies

The package has the following dependencies:

• networkx
• numpy
• plotly
• scipy
• scikit-learn

These dependencies are specified in the pyproject.toml and requirements.txt files.

[build-system]
requires = ["setuptools"]
build-backend = "setuptools.build_meta"

[project]
name = "gene-platform-utils"
version = "0.0.1"
description = "Python helpers for the Gene Platform (run with pyodide)"
dependencies = ["networkx", "numpy", "plotly", "scipy", "scikit-learn"]
requires-python = ">= 3.9"

classifiers = [
    "License :: OSI Approved :: MIT License",
]

networkx
numpy
plotly
scipy
scikit-learn