ScRNAseq

Seurat is great for scRNAseq analysis and it provides many easy-to-use ggplot2 wrappers for visualization. However, this brings the cost of flexibility. For example, In FeaturePlot, one can specify multiple genes and also split.by to further split to multiple the conditions in the meta.data. If split.by is not NULL, the ncol is ignored so you can not arrange the grid. This is best to understand with an example. library(dplyr) library(Seurat) library(patchwork) library(ggplot2) # Load the PBMC dataset pbmc.

In scanpy, there is a function to create a stacked violin plot. There is no such function in Seurat, and many people were asking for this feature. e.g. https://github.com/satijalab/seurat/issues/300 https://github.com/satijalab/seurat/issues/463 The developers have not implemented this feature yet. In this post, I am trying to make a stacked violin plot in Seurat. The idea is to create a violin plot per gene using the VlnPlot in Seurat, then customize the axis text/tick and reduce the margin for each plot and finally concatenate by cowplot::plot_grid or patchwork::wrap_plots.

In my last post, I tried to include transgenes to the cellranger reference and want to get the counts for the transgenes. However, even after I extended the Tdtomato and Cre with the potential 3’UTR, I still get very few cells express them. This is confusing to me. My next thought is: maybe the STAR aligner is doing something weird that excluded those reads? At this point, I want to give kb-python, a python wrapper on kallisto and bustools a try.

The problem I am working on some 10x scRNAseq data from transgenic mouse. The cells express Tdtomato and Cre genes. I need to add those to the cellranger reference to get the counts for those two genes. The journey to the solution Following https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/advanced/references#addgene I created a fasta file for the two transgenes: tdTomato and Cre: tdtomato_cre.fa: >tdtomato dna:chromosome chromosome:GRCm38:tdtomato:1:1431:1 REF ATGGTGAGCAAGGGCGAGGAGGTCATCAAAGAGTTCATGCGCTTCAAGGTGCGCATGGAGGGCTCCATGAACGGCCACGAGTTCGAGATCGAGGGCGAGGGCGAGGGCCGCCCCTACGAGGGCACCCAGACCGCCAAGCTGAAGGTGACCAAGGGCGGCCCCCTGCCCTTCGCCTGGGACATCCTGTCCCCCCAGTTCATGTACGGCTCCAAGGCGTACGTGAAGCACCCCGCCGACATCCCCGATTACAAGAAGCTGTCCTTCCCCGAGGGCTTCAAGTGGGAGCGCGTGATGAACTTCGAGGACGGCGGTCTGGTGACCGTGACCCAGGACTCCTCCCTGCAGGACGGCACGCTGATCTACAAGGTGAAGATGCGCGGCACCAACTTCCCCCCCGACGGCCCCGTAATGCAGAAGAAGACCATGGGCTGGGAGGCCTCCACCGAGCGCCTGTACCCCCGCGACGGCGTGCTGAAGGGCGAGATCCACCAGGCCCTGAAGCTGAAGGACGGCGGCCACTACCTGGTGGAGTTCAAGACCATCTACATGGCCAAGAAGCCCGTGCAACTGCCCGGCTACTACTACGTGGACACCAAGCTGGACATCACCTCCCACAACGAGGACTACACCATCGTGGAACAGTACGAGCGCTCCGAGGGCCGCCACCACCTGTTCCTGGGGCATGGCACCGGCAGCACCGGCAGCGGCAGCTCCGGCACCGCCTCCTCCGAGGACAACAACATGGCCGTCATCAAAGAGTTCATGCGCTTCAAGGTGCGCATGGAGGGCTCCATGAACGGCCACGAGTTCGAGATCGAGGGCGAGGGCGAGGGCCGCCCCTACGAGGGCACCCAGACCGCCAAGCTGAAGGTGACCAAGGGCGGCCCCCTGCCCTTCGCCTGGGACATCCTGTCCCCCCAGTTCATGTACGGCTCCAAGGCGTACGTGAAGCACCCCGCCGACATCCCCGATTACAAGAAGCTGTCCTTCCCCGAGGGCTTCAAGTGGGAGCGCGTGATGAACTTCGAGGACGGCGGTCTGGTGACCGTGACCCAGGACTCCTCCCTGCAGGACGGCACGCTGATCTACAAGGTGAAGATGCGCGGCACCAACTTCCCCCCCGACGGCCCCGTAATGCAGAAGAAGACCATGGGCTGGGAGGCCTCCACCGAGCGCCTGTACCCCCGCGACGGCGTGCTGAAGGGCGAGATCCACCAGGCCCTGAAGCTGAAGGACGGCGGCCACTACCTGGTGGAGTTCAAGACCATCTACATGGCCAAGAAGCCCGTGCAACTGCCCGGCTACTACTACGTGGACACCAAGCTGGACATCACCTCCCACAACGAGGACTACACCATCGTGGAACAGTACGAGCGCTCCGAGGGCCGCCACCACCTGTTCCTGTACGGCATGGACGAGCTGTACAAGTAA >cre dna:chromosome chromosome:GRCm38:cre:1:1032:1 REF ATGGCCAATTTACTGACCGTACACCAAAATTTGCCTGCATTACCGGTCGATGCAACGAGTGATGAGGTTCGCAAGAACCTGATGGACATGTTCAGGGATCGCCAGGCGTTTTCTGAGCATACCTGGAAAATGCTTCTGTCCGTTTGCCGGTCGTGGGCGGCATGGTGCAAGTTGAATAACCGGAAATGGTTTCCCGCAGAACCTGAAGATGTTCGCGATTATCTTCTATATCTTCAGGCGCGCGGTCTGGCAGTAAAAACTATCCAGCAACATTTGGGCCAGCTAAACATGCTTCATCGTCGGTCCGGGCTGCCACGACCAAGTGACAGCAATGCTGTTTCACTGGTTATGCGGCGGATCCGAAAAGAAAACGTTGATGCCGGTGAACGTGCAAAACAGGCTCTAGCGTTCGAACGCACTGATTTCGACCAGGTTCGTTCACTCATGGAAAATAGCGATCGCTGCCAGGATATACGTAATCTGGCATTTCTGGGGATTGCTTATAACACCCTGTTACGTATAGCCGAAATTGCCAGGATCAGGGTTAAAGATATCTCACGTACTGACGGTGGGAGAATGTTAATCCATATTGGCAGAACGAAAACGCTGGTTAGCACCGCAGGTGTAGAGAAGGCACTTAGCCTGGGGGTAACTAAACTGGTCGAGCGATGGATTTCCGTCTCTGGTGTAGCTGATGATCCGAATAACTACCTGTTTTGCCGGGTCAGAAAAAATGGTGTTGCCGCGCCATCTGCCACCAGCCAGCTATCAACTCGCGCCCTGGAAGGGATTTTTGAAGCAACTCATCGATTGATTTACGGCGCTAAGGATGACTCTGGTCAGAGATACCTGGCCTGGTCTGGACACAGTGCCCGTGTCGGAGCCGCGCGAGATATGGCCCGCGCTGGAGTTTCAATACCGGAGATCATGCAAGCTGGTGGCTGGACCAATGTAAATATTGTCATGAACTATATCCGTAACCTGGATAGTGAAACAGGGGCAATGGTGCGCCTGCTGGAAGATGGCGATTAG edit the genome.

Using nested dataframe and list column has transformed my way of data wrangling in R. For more on this topic, I highly recommend purrr tutorial from Jenney Bryan. In this post, I am going to show you how I use this to solve a problem for adding pct_in column from the differential scRNAseq result table. I am going to use presto for differential gene expression test. presto performs a fast Wilcoxon rank sum test and auROC analysis.

In a typical “barnyard” experiment in which cells from different species are mixed before loading to the 10x controller, the identification of the species of origin after mapping/counting with the hybrid reference is a problem. People tend to use the ratio of reads mapped to each reference genome to determine which species a cell is from. In this paper https://www.biorxiv.org/content/10.1101/630087v1.full To deconvolute species, detect doublets and low quality cells, the mixed-species mapped data was used.

I was reading Feature Selection and Dimension Reduction for Single Cell RNA-Seq based on a Multinomial Model. In the paper, the authors model the scRNAseq counts using a multinomial distribution. I was using negative binomial distribution for modeling in my last post, so I asked the question on twitter: for modeling RNAseq counts, what’s the difference/advantages using negative binomial and multinomial distribution? — Ming Tang (@tangming2005) November 26, 2019 some quotes from the answers I get from Matthew