Features
CellKb can be used to assign cell types to cluster-specific gene lists from scRNA-seq or spatial RNA-seq data to find matching cell type signatures and canonical markers published in literature.
CellKb uses a rank-based method to identify the cell type marker gene sets matching the users gene list. Advanced search features allow filtering by cell types, tissues, diseases and publications.
Cell type annotation can be performed for individual cells in scRNA-seq datasets using CellKb. Given the raw gene expression values, CellKb assigns a cell type
to each cell with a cell type score without the need for data clustering. The large cell type reference dataset provided by CellKb can be used for cell type deconvolution in spatial RNA-seq data.
Spatial RNA-seq data can be directly uploaded to CellKb which provides a cell type score for each spot. CellKb calculates this score for up to 3 cell types per spot.
Contact us for a trial to the scRNA-seq and spatial RNA-seq data annotation service in CellKb.
The CellKb advantage
CellKb provides extensive search capabilities, more data and regular updates compared to other cell type marker databases, in addition to:
- Manual curation of all data and annotations.
- Multiple marker gene sets for every cell type with information about publication, disease, tissue and experimental conditions, and reliability scores.
- Marker gene sets from single-cell and bulk RNA-seq experiments.
- Canonical markers for cell types.
- Consensus signatures for every cell type in different tissues and disease conditions.
- Cell type specific protein-protein interaction networks.
1,298 Publications
Cell type markers
The cell type marker gene sets in CellKb are extracted from tables, figures or supplementary materials
of publications describing single-cell and bulk mRNA-seq experiments selected using stringent criteria.
Cell type marker gene sets in CellKb
- Name, anatomical structure, canonical markers, condition and any special characteristics of each cell type is manually extracted
- Cell types, anatomical structures and disease conditions are assigned standardized ontology terms.
- Associated values like average expression, fold change, statistical significance are stored.
- Reliability scores are calculated for each cell type marker gene set.
- Consensus signatures are calculated for each cell type in every tissue, disease and condition.
For more details about our data collection and curation methods, please refer to our preprint:
CellKb Immune: a manually curated database of mammalian hematopoietic marker gene sets for rapid cell type identification. biorxiv, 2020.
CellKb Immune: a manually curated database of mammalian hematopoietic marker gene sets for rapid cell type identification. biorxiv, 2020.
Biomarker discovery
Since CellKb calculates consensus signatures for all cell types across all tissues and conditions in which they are found,
it can be used to compare cell type markers between diseases or tissues to identify biomarkers that are common or unique to each condition.
