brie 2.0.3

BRIE: Bayesian regression for isoform estimate

BRIE: Bayesian Regression for Isoform Estimate

About BRIE

Welcome to the new BRIE2 (Bayesian regression for isoform estimate, v2), a scalable Bayesian method to robustly identify splicing phenotypes in single cells RNA-seq designs and accurately estimate isoform proportions and its uncertainty.

BRIE2 supports isoform quantification for different needs:

1. cell features: informative prior is learned from shared cell processes. It also allows to effectively detect splicing phenotypes by using Evidence Lower Bound gain, an approximate of Bayes factor.

2. gene features: informative prior is learned from shared gene regulatory features, e.g., sequences and RNA protein binding

3. no feature: use zero-mean logit-normal as uninformative prior, namely merely data deriven

Note, BRIE1 CLI is still available in this version but changed to brie1 and brie1-diff.

Installation

BRIE2 is available through PyPI. To install, type the following command line, and add -U for upgrading:

pip install -U brie

Alternatively, you can install from this GitHub repository for latest (often development) version by following command line

pip install -U git+https://github.com/huangyh09/brie

In either case, add --user if you don’t have the write permission for your Python environment.

For more instructions, see the installation manual.

Manual and examples

The full manual is at https://brie.readthedocs.io More examples and tutorials are coming soon.

In brief, you need to run brie-count first, which will return a count matrix and hdf5 file for AnnData. Then you can use brie-quant to perform quantification in different settings. Type command line brie-count -h and brie-quant -h to see the full arguments.

References

Yuanhua Huang and Guido Sanguinetti. BRIE: transcriptome-wide splicing quantification in single cells. Genome Biology, 2017; 18(1):123.