Getting started#

bulkdgd is a Python package containing the BulkDGD generative model for the gene expression of human tissues from bulk RNA-seq data described in the work of Schuster and Krogh [SK21a, SK21b] and Prada and coworkers [PLSL+23].

bulkdgd can be used, for instance, to find differentially expressed genes between normal human samples and diseased samples.

Installation#

We provide detailed instructions to install bulkdgd in a Python virtual environment or in a conda environment in the Installation section.

Usage#

The modules of bulkdgd can be imported and used to build customized scripts and pipelines.

The API reference section provides a detailed descriptions of bulkdgd’s sub-packages and modules.

However, we also provide a small command-line interface to automate some of the most common tasks for which the DGD model can be used for more bio-oriented audiences.

Tutorials#

Our tutorials provide detailed, step-by-step explanations of how to perform different tasks using bulkdgd.

Issues, bugs, and questions#

You can report to us any bug, issue, or question about the package by opening an issue in the issues section of our GitHub repository.

Citing#

If you use our software for your research, please cite the following articles:

  • Prada-Luengo, Inigo, et al. “N-of-one differential gene expression without control samples using a deep generative model.” Genome Biology 24.1 (2023): 263.

  • Schuster, Viktoria, and Anders Krogh. “A manifold learning perspective on representation learning: Learning decoder and representations without an encoder.” Entropy 23.11 (2021): 1403.

  • Schuster, Viktoria, and Anders Krogh. “The deep generative decoder: Using MAP estimates of representations.” arXiv preprint arXiv:2110.06672 (2021).

References#

[PLSL+23]

Inigo Prada-Luengo, Viktoria Schuster, Yuhu Liang, Thilde Terkelsen, Valentina Sora, and Anders Krogh. N-of-one differential gene expression without control samples using a deep generative model. bioRxiv, pages 2023–01, 2023.

[SK21a]

Viktoria Schuster and Anders Krogh. A manifold learning perspective on representation learning: learning decoder and representations without an encoder. Entropy, 23(11):1403, 2021.

[SK21b]

Viktoria Schuster and Anders Krogh. The deep generative decoder: using map estimates of representations. arXiv, 2021.