scDehashR

The purpose of this app is to develop an interactive Shiny portal that can dehash any number of pooled samples from single cell techniques for easy visualization & downstream analysis for Bench scientists.

Table of Contents

• Background
• Example Data
• Usage
  • Installation
  • Requirements
• Plan
  • Module-1
  • Module-2
  • Module-3
  • Module-4
  • Module-5
  • Module-6
  • Module-7
  • Future Development
• Results
• Team Members

Background

Analysis of cost effective methods spawned techniques like CITE-Seq that pool samples from different experimental conditions/replicates/library preps which are sequenced to get a high throughput. There are tools like Seurat (Sajita Lab - https://satijalab.org/seurat/articles/hashing_vignette.html); who already developed a method (HTODemux) that can demultiplex pooled samples. So, through this Shiny app, we will be making it easy for anyone in the field of omics to QC, Demultiplex, Visualize, normalize and perform downstream analysis for multiplex single-cell data.

ExampleData

Summary of the Study:

Study show that modified vaccinia Ankara (MVA)-based COVID-19 vaccine expressing membrane anchored pre-fusion stabilized spike (MVA/S) induces both neutralizing antibodies and CD8+ T cells in the blood and lung and protects from SARS-CoV-2 challenge. Single-cell RNA sequencing analysis of lung cells at day 4 post-infection revealed that MVA/S vaccination also protected macaques from infection-induced inflammation and B cell abnormalities, and lowered induction of interferon stimulated genes.

(https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE165747)

Usage

Installation

Install Rstudio and install Rshiny package. We will provide the github link to the app, which can be installed using the following commands:

install.packages("devtools")
install_github("scDehashR")

Once you start the app, it should open up a Rshiny pop-up window (allow pop-up on your internet browser) which can be used to load and analyze the data.

Requirements

OS:

Works on Mac, windows and Linux (gui)

R Libraries required:

Seurat, dplyr, plotly, ggplot2, clusterprofiler, knitr, kableExtra, cowplot, gridExtra, tidyverse, biomaRt, Matrix, stringr, DoubletFinder (optional), SingleR (optional - Cell annotation Tool)

Plan

Incudes Following Modules:

Module-1

Quality Control - Visualization

Module-2

Demultiplexing and Doublet detection module – Visualization (optional)

Module-3

Identifying Cell subset module by unsupervised learning – Visualization (UMAPS, TSNE, PCA, Elbow plot).

Module-4

Integration – no visualization (background process – time taking step, dependent on no. of Capture or samples per capture or no. of cells in the experiment).

Module-5

Differential expression module – Visualization (Heatmaps, RidgePlots, FeaturePlots, DotPlots).

Module-6

Gene set enrichment analysis module – Visualization (Upset plots, enrichment plots).

Module-7

Visualization module – Additional plots like Marker genes expression across multiple treatment groups.

FutureDevelopment

1. Comparison of Demultiplexing method - HTODemux, Demuxlet, scSplit.
2. Multi-gene set comparison module – Comparison of genes in Hallmark vs KEGG pathways.
3. Option of ADT assay.
4. T-Cell and B-Cell repertoire Analysis.

Results

Yet to be posted

Team Members

Arun Boddapati | arunbodd@outlook.com | Team Leader
Srinivas Nallandighal | srinivasnvnk55@gmail.com | Team Member
Ojonugwa Abubakar | Otabu01@gmail.com | Team Member
JaMor Hairston | jmhairs@uab.edu | Team Member
Jason Needham | jneedham@uab.edu | Team Member
Aaron Lynch | aaron.lynch24@gmail.com | Team Member
Antony Linto | lintoantony@uabmc.edu | Team Member