Omar Halawa (ohalawa@ucsd.edu) & Julia Kononova (jkononova@ucsd.edu) of the GenePattern Team @ Mesirov Lab - UCSD
The following repository is a GPU-enabled GenePattern module of Tangram (see publication), a deep learning method for mapping sc/snRNA-seq data to various forms of spatial data collected from the same anatomical region or tissue type.
The module can be used for various purposes (all of which are shown in example runs), including:
- obtaining a spatial localization of cell types (see image below)
- extending gene throughput
- correcting low-quality data
- performing single-cell deconvolution
The module was written in Python 3.11 and uses a Singularity image built from its Docker counterpart. The same environment can also be replicated using the corresponding YAML environment file. This repository utilizes the code and logic found in the original Broad Institute Tangram repository.
Detailed documentation on all module parameters can be found here.
All source files, including input and output datasets for replicating runs from the original Broad Institute tutorial notebooks (tutorial_tangram_with_squidpy.ipynb & tutorial_tangram_without_squidpy.ipynb), can be found through referencing the following directory.
Biancalani, Tommaso, et al. “Deep learning and alignment of spatially resolved single-cell transcriptomes with Tangram.” Nature Methods, vol. 18, no. 11, 28 Oct. 2021, pp. 1352–1362.
| Version | Release Date | Description |
|---|---|---|
| 1.0 | Aug 23, 2024 | Initial version for public use. |