This experiment is done in September 2023 with 5 Arabidopsis genotypes for DUF247 project. In brief, seeds were germinate in 1/2 MS media for one week, and then trasplanted to the soil with 100% WHC for one more week until the soil reached 50% WHC, then salt stress was applied by soaking pots in 200 mM NaCl, resulting in final concentration of 100 mM NaCl. The shoot growth was measure over the period of 12 days, by measuring growth rate for 3 days before stress exposure and 9 days of stress exposure.
getwd()## [1] "C:/Users/Julkowska Lab/Desktop/R codes by Maryam/20240307_At_5geno_soil_phenorigs_experiment_shoot_growth"#Let’s import ALL the files with data:
my_files <- list.files(pattern="-single-value-traits.csv")
my_files## [1] "raspiK_cameraA_KA.result.csv-single-value-traits.csv"
## [2] "raspiM_cameraA-single-value-traits.csv"
## [3] "raspiM_cameraB-single-value-traits.csv"
## [4] "raspiT_cameraA-single-value-traits.csv"
## [5] "raspiT_cameraB-single-value-traits.csv"
## [6] "raspiU_cameraA-single-value-traits.csv"
## [7] "raspiU_cameraB-single-value-traits.csv"
## [8] "raspiV_cameraA_VA.result.csv-single-value-traits.csv"
## [9] "raspiV_cameraB_VB.result.csv-single-value-traits.csv"Let’s import ALL the data as one dataset:
M_A <- read.csv("raspiM_cameraA-single-value-traits.csv" )
M_B <- read.csv("raspiM_cameraB-single-value-traits.csv" )
T_A <- read.csv("raspiT_cameraA-single-value-traits.csv" )
T_B <- read.csv("raspiT_cameraB-single-value-traits.csv" )
V_A <- read.csv("raspiV_cameraA_VA.result.csv-single-value-traits.csv")
V_B <- read.csv("raspiV_cameraB_VB.result.csv-single-value-traits.csv")
U_A <- read.csv("raspiU_cameraA-single-value-traits.csv" )
U_B <- read.csv("raspiU_cameraB-single-value-traits.csv" )
colnames(M_A) %in% colnames(M_B)## [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [16] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [31] TRUEall_data <- rbind(M_A, M_B)
all_data <- rbind(all_data, T_A)
all_data <- rbind(all_data, T_B)
all_data <- rbind(all_data, V_A)
all_data <- rbind(all_data, V_B)
all_data <- rbind(all_data, U_A)
all_data <- rbind(all_data, U_B)Let’s remove the columns we are not using anyways:
colnames(all_data)## [1] "camera" "imgtype" "zoom"
## [4] "exposure" "gain" "frame"
## [7] "lifter" "timestamp" "id"
## [10] "plantbarcode" "treatment" "cartag"
## [13] "measurementlabel" "other" "image"
## [16] "sample" "in_bounds" "area"
## [19] "convex_hull_area" "solidity" "perimeter"
## [22] "width" "height" "longest_path"
## [25] "convex_hull_vertices" "object_in_frame" "ellipse_major_axis"
## [28] "ellipse_minor_axis" "ellipse_angle" "ellipse_eccentricity"
## [31] "plantID"all_data2 <- all_data[,c(31,18:30)]
all_data2Let’s get all of the mega-data isolated from our PlantID - so we can fuse it with the identifying information - to look at differences between treatments + genotypes:
all_data2$info <- strsplit(all_data2$plantID, "_")
all_data2$info[1]## [[1]]
## [1] "raspiM" "cameraA.2023.10.12-11.00.01"
## [3] "11"all_data2$info[1][[1]][1]## [1] "raspiM"all_data2$info[1][[1]][3]## [1] "11"all_data2$info[2][[1]][1]## [1] "raspiM"all_data2$info[2][[1]][3]## [1] "6"We can extract Raspi info and Position info from here
for(i in 1:nrow(all_data2)){
all_data2$RasPi[i] <- all_data2$info[i][[1]][1]
all_data2$Position[i] <- all_data2$info[i][[1]][3]
all_data2$info2[i] <- all_data2$info[i][[1]][2]
}
all_data2Now - let’s isolate TIME by strsplitting on dash:
all_data2$info3 <- strsplit(all_data2$info2, "-")
all_data2$info3[1]## [[1]]
## [1] "cameraA.2023.10.12" "11.00.01"for(i in 1:nrow(all_data2)){
all_data2$time[i] <- all_data2$info3[i][[1]][2]
all_data2$info4[i] <- all_data2$info3[i][[1]][1]
}then - let’s isolate Camera info and Date info:
all_data2$info4 <- gsub(".2023", "-2023", all_data2$info4)
all_data2$info5 <- strsplit(all_data2$info4, "-")
all_data2$info5[2]## [[1]]
## [1] "cameraA" "2023.10.04"for(i in 1:nrow(all_data2)){
all_data2$cameraID[i] <- all_data2$info5[i][[1]][1]
all_data2$Date[i] <- all_data2$info5[i][[1]][2]
}
all_data2Let’s re-shape the dataframe once again:
colnames(all_data2)## [1] "plantID" "area" "convex_hull_area"
## [4] "solidity" "perimeter" "width"
## [7] "height" "longest_path" "convex_hull_vertices"
## [10] "object_in_frame" "ellipse_major_axis" "ellipse_minor_axis"
## [13] "ellipse_angle" "ellipse_eccentricity" "info"
## [16] "RasPi" "Position" "info2"
## [19] "info3" "time" "info4"
## [22] "info5" "cameraID" "Date"data <- all_data2[,c(1,16,23,24,20,17,2:14)]
head(data)In order to fuse this with our decoding file - we need to create unique plantID by fusing RasPi, CameraID and position:
data$ID <- paste(data$RasPi, data$cameraID, data$Position, sep="_")
data