1. EuPathDB: Provides access to pathogen annotation resources available on EuPathDB databases

This package helps us to retrieve annotation data from EuPathDB.

  1. AnnotationForge: Tools for building SQLite-based annotation data packages

Databases for fungal pathogens were provided on different platforms, we need a way to make it compatible to input into other analysis packages. Annotation Forge is a robust tool for this task.

  1. AnnotationHub: Client to access AnnotationHub resources

The Bioconductor team wants to collect popular web resources into their package called AnnotationHub. This beneficial tool provides genomic and transcriptomic profiles for many applications.

  1. Biomart: Interface to BioMart databases (i.e. Ensembl)

This tool provides a way to access BioMart (Ensembl) annotation database directly from R. This is very useful to query annotation data from Ensembl.

  1. clusterProfiler: A universal enrichment tool for interpreting omics data

This package provides an interface for functional annotation using data from various sources. They also provide excellent functions to visualize enrichment results obtained from its analysis.

  1. hpgltools: A pile of (hopefully) useful R functions

This is a set of functions really helpful for host-pathogen interaction analysis.

One of my very first bioinformatic tutors develop it!!! I should say that the author of this package is a super lovely scientist. I was an amateur guy in bioinformatics when I started my Ph.D. There were so many simple troubles I was struggling with every day. I have some problems with the database on AnnotationHub. I contacted him, and he was so helpful and kind to help me a lot. He made an Rmd file to guide me in handling my data and try to explain basic things via email. I am very grateful to him. Hope to see him in person somewhere soon!

If you are working in microbiology and usually do growth curves, this is a topic for you.

I worked with growth curves during my working time. I have done this on 96-well plates and gotten a lot of data points. I saw some people copy and process those data types with Excel and GraphPad. It is terrible for me!

For Candida auris, I measured every hour for at least 2 days which is not able to plot manually with MS Excel or GraphPad. Therefore, preparing some R scripts for data processing and visualizing is make sense.

1. Prepare your data

When you get data from the plate reader, it is in CSV or excel formats. You need to name your files by the same pattern with different time points, we will extract the number for plotting later. All data files are put in the same folder.

I guest your data from the plate reader will look like below with 96 data points.

2. Import your data into R and transform it into the long format

#call library

#get file name
#change pattern following your file names
listFiles <- list.files(pattern = "mBioRev_GR_*")

#read data from excel into long data and annotate the data
listData <- lapply(listFiles, function(x) { #read all data into a list
  df = melt(readxl::read_excel(x, #use melt() from reshape 2 to make long data
              range = "A13:M20", #select your data region
              col_names = c("row", seq(1:12))), #add column names from 1-12
              value.name = gsub(".*_(.+)h.xlsx", "\\1", x), 
              variable.name = "col") 
  df$ID <- str_c(df$row, df$col) #generate an ID column from row names and column names. Result of ID should be A1-H12
  df<- df[,c(4,3)] #select ID and OD value that we need

After this step, each file will be an element of listData, you need to convert listData into data frame.

#convert list to dataframe
dat <- Reduce(function(x,y) merge(x=x, y=y, by="ID", all=T), listData)

#transform into long data
ldat <- melt(dat, variable.name = "time", value.name = "OD600")

Your final result should look like below

3. Load metadata

Prepare metadata for your 96-well plates.

  • for medium

  • for strain list

#load metadata into R and anotate
meta_medium <- readxl::read_excel("meta.xlsx", sheet = "medium",
                   col_names = c("row", seq(1:12)), skip = 1)
meta_medium <- melt(meta_medium, id.vars = c("row"), value.name = "medium") #transform into long format
meta_medium$ID <- str_c(meta_medium$row, meta_medium$variable) #generate ID (A1-H12)
meta_medium <- meta_medium[,c(4,3)] #select essential columns

#do similar for strains
meta_strains <- readxl::read_excel("meta.xlsx", sheet = "strains",
                                  col_names = c("row", seq(1:12)), skip = 1)
meta_strains <- melt(meta_strains, id.vars = c("row"), value.name = "strains")
meta_strains$ID <- str_c(meta_strains$row, meta_strains$variable)
meta_strains <- meta_strains[,c(4,3)]

#merge metadata from strains and mediums
meta <- merge(meta_medium, meta_strains)

4. Join your data with meta

#Join data
pdat <- merge(ldat, meta, all= T)

5. Visualize data

#call library

#check your data structure first to make sure numeric value of time and OD

#convert to numeric
pdat$time <- as.numeric(as.character(pdat$time))

###plot for each medium
p <- ggplot(data = pdat[pdat$medium=="YPD"&pdat$strains!="control",],
       aes(time, OD600, col = strains)) + 
  geom_point() + 
  geom_smooth(se = FALSE,alpha=0.5, span = 0.3) +
  theme_bw() + 
  scale_color_npg()  + #set nature journal colors
  theme(text = element_text(size = 20), legend.position=c(0.8, 0.4))

#Saving as svg
svg("YPD.svg",                      # File name
    width = 8, height = 7)          # Paper size
#plot name
# Closing the graphical device

Your result should look like the figure below