LTMG

LTMG (Left Truncated Mixture Gaussian) is a software tool to model and analyze single-cell RNA-seq data. It addresses the challenges of capturing the diversity of gene expression states across individual cells, which are regulated by different transcriptional regulatory inputs and complicated by the presence of zero and low expressions.

Key features and capabilities of LTMG:

1. Infers expression multi-modalities across single cells while treating dropouts and low expressions as left truncated.

2. Demonstrates better goodness of fit compared to other state-of-the-art models when tested on a large number of scRNA-seq datasets.

3. Validates the biological assumption of low non-zero expressions, the rationality of the multimodality setting, and the ability to extract expression states specific to cell types or functions using independent experimental data sets.

4. It includes a differential gene expression test with higher sensitivity and specificity than other popular methods.

5. Provides a co-regulation module identification method capable of retrieving gene co-regulation modules corresponding to perturbed transcriptional regulations.\