README.md

In NCBI-Hackathons/robustSingleCell: Robust Subpopulation dEcision Trees

Discovery of cell subpopulations using single cell sequencing data and ReSET

What is ReSET (Robust Subpopulation dEcision Trees)?

ReSET is A tool that leverages the knowledge of cell subpopulation hierarchy to discover robust populations defined from single-cell RNA-Seq experiments. ReSET creates a hierarchical model that fits two independent single cell RNA-Seq datasets.

Overview

Single-cell RNA sequencing enables unbiased analysis of expression patterns. However, researchers don’t have the tools for appropriate decision making during the analysis. Our general aim is to introduce a data-driven strategy to identify the appropriate number of robust subpopulations, their discriminatory defining markers and the relationships between populations.

Workflow diagram

Getting started

To run ReSET on your experimental data, describe your samples in a CSV file sample_sheet.csv, provide a settings.yaml to override the defaults defaults, and select the pipeline.

To generate a settings file template for any pipeline:

ReSET [pipeline] --init=settings

To generate a sample sheet template for any pipeline:

ReSET [pipeline] --init=sample-sheet

Here's a simple example to run the RNAseq pipeline:\

ReSET rnaseq my-sample-sheet.csv --settings my-settings.yaml

To see all available options run ReSET --help.

Install

A pre-built package is available in this repository.

First, you need to install the devtools package which is available from CRAN. Invoke R and then type

install.packages("devtools")

Load the devtools package.

library(devtools)

Install ReSET

install_github("NCBI-Hackathons/robustSingleCell")

Example Data Sources

10X Genomics, 4k Pan T Cells from a Healthy Donor

10X Genomics, 3k Pan T Cells from a Healthy Donor

Dependencies

python = 2.7

R >= 3.5

Project Team (Alphabetical Order)

Main team

• Assaf Magen (Team Lead)
• Mamie Wang
• Billy Kim

Contributors

• Jonathan Badger
• Claire Malley
• Chris Rhodes
• Amulya Shastry

Presentations

• day 1
• day 2
• day 3 (final)
• Plan

License

NCBI-Hackathons/robustSingleCell is licensed under the MIT License. See LICENSE for further details.

NCBI-Hackathons/robustSingleCell documentation built on May 9, 2019, 3:27 a.m.