In matthewhirschey/tidybiology: Utility Functions and Data Sets for Tidy Biology

Load data

library('scRNAseq')
scRNAData <- scRNAseq #Because ggplot2 builder needs the dataset to be in the Environment

Preview data structure

library('dplyr')
glimpse(scRNAData)

head(scRNAData,4) #Dummy datasets will be generated from these 4 rows

Generate dummy df

heart <- scRNAData[scRNAData$organ=='heart',]
brain <- scRNAData[scRNAData$organ=='brain',]
kidney <- scRNAData[scRNAData$organ=='kidney',]

Does the original df have any transcript appeared in more than 1 organ?

• Clue for more than 2 df join(): try to use %>%

Assembly the df from 3 organs together in a proper way and name the new df as three_organ_df

FIlter the genes from heart out from three_organ_df

• You can use either a join function or a negated %in% (%!in% <- Negate(%in%))
• If you want to use a join function, make sure you understand which is the "left" and which is the "right"

Conditional coloring in ggplot by stringr pattern detect

What gene types do you have in scRNAData?

sort(table(scRNAData$gene_type), decreasing = TRUE)

What is the average transcript lengthn for non-pseudogenes from brain? Make a plot to answer this.

Additional requirements:

• Filter the scRNAData by transcript length by a range that you think is reasonable to remove potential artifacts
• Change the x-axis label of a boxplot to Pseudogenes if its gene type belongs to pseudogenes or Non-pseudogenes if not
• Use the organ label to fill in the color of boxplots. If a gene is from brain, label it as Brain-related; if not from brain, label it as Other-organs-related

matthewhirschey/tidybiology documentation built on June 11, 2022, 7:32 p.m.