Home

Welcome to the CONICSmat wiki!

CONICSmat (COpy-Number analysis In single-Cell RNA-Sequencing from an expression matrix ) is a tool to infer large-scale copy number variations (CNVs) from single-cell RNA-seq data.

Typically, tumor biopsies consist of a mixture of neoplastic cells and non-neoplastic cells from the tumor micro-environment. While clustering of cells based on expression of highly variable genes may provide preliminary evidence for their identity, only the presence/absence of somatic mutations can be interpreted as proof for cell identity.

To infer the copy number status of each cell, CONICSmat fits a two component Gaussian Mixture Model for each user-provided chromosomal region. The mixture model is fit to the average gene expression of genes within a region, for example all genes on chromosome 10, across all cells. Cells with a deletion of the region will show an on average lower expression from the region than cells without the deletion. The posterior probabilities for each cell belonging to one of the components can then be used to construct a heatmap that visualizes the copy number status of each cell.

It is important to note that CONICSmat can be run without a definitive normal control, so single cell RNA-seq of paired non-malignant tissue is not needed. The inferred CNV profiles can be used to triage malignant and non-malignant cells, an to subsequently infer mutational phylogenies of cancer cells.

On the right side you find different tutorials on how to use CONICSmat. Depending on the availability of Exome-seq data, there is different approaches to analyze your data with CONICSmat.

For questions regarding the usage of CONICSmat use the issues section or email Soren.