In compbiomed/singleCellTK: Comprehensive and Interactive Analysis of Single Cell RNA-Seq Data

knitr::opts_chunk$set(echo = TRUE, eval = TRUE, message=FALSE, warning=FALSE)

````{=html}

## Introduction

Single Cell Toolkit (singleCellTK, SCTK) is a package that works on single-cell RNA-seq (scRNAseq) dataset. SCTK allows users to import multiple datasets, perform quality control and a series of preprocessing, get clustering on cells and markers of clusters, and run various downstream analysis. Meanwhile, SCTK also wraps curated workflows for [celda](https://www.camplab.net/celda/) and [Seurat](https://satijalab.org/seurat/). 

The SCTK tutorial series start from importing, QC and filtering. After this, users can go through either the [A La Carte Workflow](02_a_la_carte_workflow.html) or the Curated Workflows ([Seurat](seurat_curated_workflow.html) or [Celda](celda_curated_workflow.html)). These tutorials will take the real-world scRNAseq dataset as an example, which consists of 2,700 Peripheral Blood Mononuclear Cells (PBMCs) collected from a healthy donor, namingly PBMC3K. This dataset is available from 10X Genomics and can be found on the 10X website.

Before starting the analysis from either UI or console, users should have SCTK properly installed first. Please refer to the [Installation](installation.html) documentation for detail of installing the `singleCellTK` package and relevant dependencies. Alternatively, users can also access the functionality by visiting our <a href="https://sctk.bu.edu/" target="_blank">online web server deployment https://sctk.bu.edu/</a>, or using a stand-alone [docker image](https://hub.docker.com/r/campbio/sctk_shiny). 

## Importing Data

SCTK allows importing dataset from various type of sources, such as preprocessing tools like cellranger or flat text files. SCTK also allows importing multiple batches (samples) of dataset at the same time. Please refer to the [detailed documentation of Importing](import_data.html) for different ways of importing. This tutorial only shows how we can import the example PBMC3K data.

````{=html}
<div class="tabset" id="tabset1">
<div class="tab">
  <button class="tablinks ia" onclick="openTab(event, 'ia', 'tabset1')">Interactive Analysis</button>
  <button class="tablinks console" onclick="openTab(event, 'console', 'tabset1')">Console Analysis</button>
</div>

<div class="tabcontent ia">

After starting the UI, the landing page will be for importing the scRNAseq data. To import the PBMC3K dataset previously mentioned, users should follow these steps:

\

1. If you are not already at the starting import page, you start by going to the Data tab and selecting the Import Single Cell Data option (2).

\

1. Select the relevant way to import data. For the purpose of this example, we use pbmc3k filtered raw data using the CellRanger option. This data can be downloaded from here.

\

1. In the appropriate input boxes, we select the matrix, feature and barcode files and upload each sample individually by giving a unique sample name (5).

\

1. Now either import samples or Add one more sample. You can keep adding more samples until all samples have been added and then finally import all of them together.

\

After successfully importing, the third collapse box will pop up and show users the basic summary stats of the imported dataset. Meanwhile, users can set feature display options here. Most of the time, the dataset has default feature ID (usually seen as row names of a matrix) together with other types of ID (e.g. symbol) in the feature metadata. The first option "Set feature ID" sets the type of default feature ID, which should be unique and has no NA value. The second option "Set feature names to be displayed in downstream visualization" does what it says. When we need to show features on a plot, such as a heatmap or a volcano plot, usually it is better to show gene symbols rather than ensembl IDs.

````{=html}