Skip to content

An R Shiny app to help with scRNA-seq benchmarking and analysis with clustifyr

Notifications You must be signed in to change notification settings

SidhantPuntambekar/clustifyr-web-app

 
 

Repository files navigation

Clustifyr Shiny App

The purpose of this app is to enable quick classification of single-cell RNA sequencing data through an interactive web interface. Users can directly upload their matrix and metadata files or Seurat/SCE objects generated from single-cell RNA sequencing analyses and produce useful cell identity inference and visualization,using a built-in library of references (clustifyrdatahub) compiled from reference bulk RNA-seq experiments, microarray expression data, and single-cell gene signatures. An additional purpose of the app is to enable quick browsing, preview, and reference building directly from NCBI Gene Expression Omnibus (GEO) records. Data reuse for this application and many other reanalysis/extension purposes require accurate metadata, which is frustratingly rare. We call on data repositories, journals, and investigators to work together towards ensuring proper cell-level annotation deposition. Please see someta for further discussions.

Workflow

  1. Upload expression, either raw counts or normalized, matrix. Or Seurat/SCE object. You can also retrieve GEO data through accession #.
  2. Upload cell-level metadata in text formats. Or Seurat/SCE object. You can also retrieve GEO data through accession #.
  3. Choose (in dropdown menu or just click on the preview) the column in metadata that represent clustering information.
  4. Choose or upload reference dataset, a matrix containing average expression of each cell type. Mouse MCA is the default.
  5. Go to clustify step and look at results: correlation matrix, called cell-types, and heatmap. Results can be downloaded as xlsx.

Clustifyr Background

Single cell transcriptomes are difficult to annotate without knowledge of the underlying biology. Even with this knowledge, accurate identification can be challenging due to the lack of detectable expression of common marker genes. clustifyr aims to alleviate this problem by automatically annotating single cells or clusters of cells using single-cell RNA-seq, bulk RNA-seq data, microarray, or marker gene lists. Additional functions enable exploratory analysis of similarities between single cell RNA-seq datasets and reference data.

Clustifyr Data Hub

Reference cell type gene signatures are located in the accompanying Clustifyr Data Hub. Descriptions of each data set are present in the table below.

Available references include

Title Species Description RDataPath BiocVersion Genome SourceType SourceUrl
ref_MCA Mus musculus Mouse Cell Atlas clustifyrdatahub/ref_MCA.rda 3.12 mm10 Zip https://ndownloader.figshare.com/files/10756795
ref_tabula_muris_drop Mus musculus Tabula Muris (10X) clustifyrdatahub/ref_tabula_muris_drop.rda 3.12 mm10 Zip https://ndownloader.figshare.com/articles/5821263
ref_tabula_muris_facs Mus musculus Tabula Muris (SmartSeq2) clustifyrdatahub/ref_tabula_muris_facs.rda 3.12 mm10 Zip https://ndownloader.figshare.com/articles/5821263
ref_mouse.rnaseq Mus musculus Mouse RNA-seq from 28 cell types clustifyrdatahub/ref_mouse.rnaseq.rda 3.12 mm10 RDA https://github.com/dviraran/SingleR/tree/master/data
ref_moca_main Mus musculus Mouse Organogenesis Cell Atlas (main cell types) clustifyrdatahub/ref_moca_main.rda 3.12 mm10 RDA https://oncoscape.v3.sttrcancer.org/atlas.gs.washington.edu.mouse.rna/downloads
ref_immgen Mus musculus Mouse sorted immune cells clustifyrdatahub/ref_immgen.rda 3.12 mm10 RDA https://github.com/dviraran/SingleR/tree/master/data
ref_hema_microarray Homo sapiens Human hematopoietic cell microarray clustifyrdatahub/ref_hema_microarray.rda 3.12 hg38 TXT https://ftp.ncbi.nlm.nih.gov/geo/series/GSE24nnn/GSE24759/matrix/GSE24759_series_matrix.txt.gz
ref_cortex_dev Homo sapiens Human cortex development scRNA-seq clustifyrdatahub/ref_cortex_dev.rda 3.12 hg38 TSV https://cells.ucsc.edu/cortex-dev/exprMatrix.tsv.gz
ref_pan_indrop Homo sapiens Human pancreatic cell scRNA-seq (inDrop) clustifyrdatahub/ref_pan_indrop.rda 3.12 hg38 RDA https://scrnaseq-public-datasets.s3.amazonaws.com/scater-objects/baron-human.rds
ref_pan_smartseq2 Homo sapiens Human pancreatic cell scRNA-seq (SmartSeq2) clustifyrdatahub/ref_pan_smartseq2.rda 3.12 hg38 RDA https://scrnaseq-public-datasets.s3.amazonaws.com/scater-objects/segerstolpe.rds
ref_mouse_atlas Mus musculus Mouse Atlas scRNA-seq from 321 cell types clustifyrdatahub/ref_mouse_atlas.rda 3.12 mm10 RDA https://github.com/rnabioco/scRNA-seq-Cell-Ref-Matrix/blob/master/atlas/musMusculus/MouseAtlas.rda

About

An R Shiny app to help with scRNA-seq benchmarking and analysis with clustifyr

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 100.0%