Cargando…

GeneMarkeR: A Database and User Interface for scRNA-seq Marker Genes

Single-cell sequencing (scRNA-seq) has enabled researchers to study cellular heterogeneity. Accurate cell type identification is crucial for scRNA-seq analysis to be valid and robust. Marker genes, genes specific for one or a few cell types, can improve cell type classification; however, their speci...

Descripción completa

Detalles Bibliográficos
Autores principales: Paisley, Brianna M., Liu, Yunlong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8577352/
https://www.ncbi.nlm.nih.gov/pubmed/34764987
http://dx.doi.org/10.3389/fgene.2021.763431
Descripción
Sumario:Single-cell sequencing (scRNA-seq) has enabled researchers to study cellular heterogeneity. Accurate cell type identification is crucial for scRNA-seq analysis to be valid and robust. Marker genes, genes specific for one or a few cell types, can improve cell type classification; however, their specificity varies across species, samples, and cell subtypes. Current marker gene databases lack standardization, cell hierarchy consideration, sample diversity, and/or the flexibility for updates as new data become available. Most of these databases are derived from a single statistical analysis despite many such analyses scattered in the literature to identify marker genes from scRNA-seq data and pure cell populations. An R Shiny web tool called GeneMarkeR was developed for researchers to retrieve marker genes demonstrating cell type specificity across species, methodology and sample types based on a novel algorithm. The web tool facilitates online submission and interfaces with MySQL to ensure updatability. Furthermore, the tool incorporates reactive programming to enable researchers to retrieve standardized public data supporting the marker genes. GeneMarkeR currently hosts over 261,000 rows of standardized marker gene results from 25 studies across 21,012 unique genomic entities and 99 unique cell types mapped to hierarchical ontologies.