Demultiplexing barcoded single-cell sequencing data with ESGI

ESGI - Efficient Splitting of Generic Indices - makes demultiplexing singl-cell data easy. Download the App to easily run demultiplexing on your laptop (without gene-mapping) or download to CLI to run ESGI on a server for more functionality.

ESGI

ESGI demultiplexes single-cell data by assigning reads to their cell of origin and counting them per cell. The input is a file specifying the mapping pattern, which can be written quickly and makes running ESGI straightforward.

Features

ESGI works on any multiplexed single-cell data. Its generic input pattern allows you to demultiplex almost any barcoded single-cell data. Here are some features and use-cases that ESGI can handle:

• ESGI can demultiplex barcoded data like 10X, Combinatorial-indexing experiments, spatial sequencing data, etc.
• ESGI takes a pattern as input, which encodes the barcode-structure. It allows barcodes with fixed bases, variable barcodes that can be given by a file, RNA sequences or UMIs (15X for 15 nucleotide long UMI), e.g.: [GACGCATACGT][15X][SINGLE_CELL_BARCODES_FILE.txt][PROTEIN_BARCODES_FILE.txt]
• ESGI can map the RNA part in the read, but can also count other modalities, like CITE-seq or intra-cellular protein measurement technologies.
• For RNA mapping we recommend to run ESGI (together with a mapper like STAR) on a server
• For simpler experiments (like counting protein tags) ou can easily run the App on your local laptop
• ESGI allows also barcodes on different length
• For flexability ESGI can also take several patterns at once as input, in case you pooled different modalities, or read-structures

For features, getting started with development, see the Getting Started page. Would you like to request a feature or contribute? Open an issue