1 Create a connection to SDY144

First, we initialize the connection to SDY144 using CreateConnection.

library(ImmuneSpaceR)
library(data.table)
library(ggplot2)

con <- CreateConnection("SDY144")

## Warning in matrix(data = unlist(curfld[resultCols]), nrow = 1, ncol =
## length(resultCols), : data length [4] is not a sub-multiple or multiple of the
## number of columns [7]

2 Retrieve and manipulate data

We grab the datasets of interests with the getDataset method.

flow <- con$getDataset("fcs_analyzed_result")
hai <- con$getDataset("hai")
vn <- con$getDataset("neut_ab_titer")

Then, we select the cell populations and time points of intereset.

pb <- flow[population_name_reported %in% c("Plasma cells,Freq. of,B lym CD27+",
                                           "Plasmablast,Freq. of,Q3: CD19+, CD20-")]
pb <- pb[, population_cell_number := as.numeric(population_cell_number)]
pb <- pb[study_time_collected == 7 & study_time_collected_unit == "Days"] # 13 subjects
pb <- pb[, list(participant_id, population_cell_number, population_name_reported)]

We compute the HI and VN titer as the fold-increase between baseline and day 30.

# HAI
hai <- hai[, response := value_preferred / value_preferred[study_time_collected == 0],
           by = "virus,cohort,participant_id"][study_time_collected == 30]
hai <- hai[, list(participant_id, virus, response)]
dat_hai <- merge(hai, pb, by = "participant_id", allow.cartesian = TRUE)

# VN
vn <- vn[, response:= value_preferred/value_preferred[study_time_collected == 0],
         by = "virus,cohort,participant_id"][study_time_collected == 30]
vn <- vn[, list(participant_id, virus, response)]
dat_vn <- merge(vn, pb, by = "participant_id", allow.cartesian = TRUE)

3 Visualize using ggplot2

ggplot(dat_hai, aes(x = population_cell_number, y = response)) +
  geom_point() + 
  geom_smooth(method = "lm") +
  facet_grid(virus ~ population_name_reported, scale = "free") +
  xlab("Number of cells") + 
  ylab("HI fold-increase Day 30 vs. baseline") + 
  theme_IS()

ggplot(dat_vn, aes(x = population_cell_number, y = response)) +
  geom_point() + 
  geom_smooth(method = "lm") +
  facet_grid(virus ~ population_name_reported, scale = "free") +
  xlab("Number of cells") + 
  ylab("VN fold-increase Day 30 vs. baseline") + 
  theme_IS()