VAMF

The software tool 'VAMF' is designed for the analysis of single-cell RNA-Seq (scRNA-Seq) data, a pivotal technology for gene expression profiling at the individual cell level. This high-throughput method offers insights into cell-to-cell variability across a genomic scale, opening new avenues in research. Standard techniques involve dimension reduction and clustering to discover distinct cell types or phenotypes. Yet, scRNA-Seq datasets are often sparse, with many measurements reported as zeros due to experimental limitations and technical artifacts. Conventional approaches treat these zeros as genuine, leading to distorted low-dimensional factors that result in erroneous biological group identification. 'VAMF' introduces a novel approach that considers cell-specific censoring using a varying-censoring aware matrix factorization (VAMF) model. This model addresses systematic bias and enables accurate factor identification, enhancing clustering results. The benefits of the 'VAMF' approach are demonstrated using published scRNA-Seq data and validated with simulated data.