Hi there,
I am Quang, a current PhD student at Oxford. In my research, I applied computational tools to analyze hi-dimensional dataset, such as scRNAseq and flow cytometry data.
Welcome back to my channel, the Series of #learntoRfromNothing ! The series will skip all the boring (though important lol) theories about coding, and jump right into the codes.
~~~~~~~~~~~~~~~~~~~~
Seurat is a popular R package for the analysis of single-cell RNA-sequences. Tutorials for Seurat is abundant on the Internet, but these tutorials do not alert users, especially beginners, about the key considerations/issues and potential approaches to address them. Moreover, Seurat has various alternative workflows that are often not clear to and hence overlooked by beginners. Anyone can plug and chuck the codes, but practical knowledge of these issues is critical to the validity of the results obtained.
~~~~~~~~~~~~~~~~~~~
Learning Objectives for Day :
0:00 – Objective for today’s workshop: How to decide certain QC thresholds for scRNAseq data with Seurat R package?
1:34 – What does the Seurat pipeline involve from count matrix to clusters?
5:23 – How to download and load the sample scRNAseq PBMC data and all the codes in this workshop?
7:23 – How to install needed packages in R Studio for today’s workshop?
7:41 – Examine the sample data
9:10 – How to check if certain genes are detected in your dataset?
10:07 – How to decide threshold (min.cells) to remove low-quality genes?
10:24 – How to compute quantitle in R?
15:11 – How to check how many genes you removed with min.cells threshold?
16:44 – How to pull out the cells containing the genes you removed with min.cells threshold?
17:52 – How to decide threshold to remove low-quality cells?
18:09 – How to decide threshold for percent.mt?
19:25 – How to visualize percent.mt metric with a violin VlnPlot?
21:18 – SMALL EXERCISE: Why this nFeatures threshold?
22:19 – Solution to small exercise: How to decide threshold for nFeatures?
23:26 – How to visualize two QC metrics in the same plot (FeatureScatter)?
27:55 – How to summarize how many cells we have after QC in R?
28:23 – How to know many cells expressed a gene in R with Seurat package?
28:55 – What are other QC metrics for your scRNAseq data? (#doublets, #ribosomalgenes, cell cycle genes, novelty score, #sequencingdepth, #samplingerror)
32:07 – How to automatically determine QC threshold for scRNAseq with miQC R package?
33:01 – How to regress out unwanted variation in scRNAseq data with Seurat R package?
33:59 – What is the latent.vars in FindMarker function in Seurat?
34:41 – How to perform clustering in Seurat?
35:20 – How to do several #clusteringresolutions at once by FindClusters in Seurat?
37:24 – How to determine the right clustering resolution with the #clustreeRpackage?
37:46 – How to assess the stability of clusters as we increase clustering resolution with clustree R package?
38:53 – How to assess the expression of a gene across clusters as we increase clustering resolution with clustree R package?
40:30 – How to count how many cells per cluster in Seurat R package?
41:31 – How to QC your clusters by visualization using ggplot and Seurat R package? #batcheffect #confoundingvariable #dimplot
44:13 – How to subset cells graphically with CellSelector function in Seurat on a UMAP?
44:58 – How to peform DEG analysis and use differentially expressed genes to annotate clusters?
46:27 – MEGA EXERCISE: Are there more CD8 or CD4 T cells in PBMCs?
47:04 – Best practice for QC your scRNAseq data in Seurat R package
48:07 – Additional online resources to learn about Seurat and scRNAseq analysis
~~~~~~~~~~~~~~~~~~~
Intended audience: Anyone who would like to use Seurat R package to analyze single-cell RNA-sequencing data
~~~~~~~~~~~~~~~~~~~~
Resource for today:
All materials (codes, sample data, and PowerPoint slide) are available to download at https://github.com/denvercal1234GitHub/learntoRfromNothing
~~~~~~~~~~~~~~~~~~~~
#Seurat #RNAseqanalysis #rprogrammingforbeginners #rprogramming #dataframes #dataanalysis #RStudio #Rmarkdown #Seuratpackage #RNAseq #SeuratData #oxfordstudent #codingtutorials #learntoRfromNothing #coding #programming #computational #computationalbiology #immunology #computationalimmunology #vaccinology #coding #computationalbiologist #QC #singlecellRNAseq #singlecellRNAsequencing #RNAseq #RNAsequencing #QCdataset #scRNAseqworkshops #clustree #miQC #nFeatures #nCounts #UMI #cellranger #Rprogramming #RStudio #QNNDukeNguyen #mincells #ggplot #ggplot2 #visualization #visualizationinr #histogram #dotplot #violinplots #vlnplot #clustering #doubletfinder #scrublet #normalization #SCTransform #differentialexpressedgenes #deganalysis #dimentionalityreduction #pca #cellselector #pbmc #scRNAseqworkshop #Seuratworkshop
source
