In my previous post, I explained how to quantify phenotypic plasticity and introduced the concept of ‘responsiveness.’
□ Quantifying Phenotypic Plasticity of Crops
I introduced a formula to calculate responsiveness as (Treatment - Control) / Control.
| Genotype | Control | Treatment | Responsiveness |
| A | 100 | 90 | -10.0% |
| B | 120 | 70 | -41.7% |
| C | 115 | 90 | -21.7% |
| D | 95 | 85 | -10.5% |
| E | 110 | 105 | -4.5% |
However, when analyzing data, the format may not always be the same as above. Mostly, treatments (independent variable) are arranged in one column, and the dependent variable is arranged in another column. In this case, how to calculate responsiveness calculated as (Treatment - Control) / Control?
Let’s upload one data.
if(!require(readr)) install.packages("readr")
library(readr)
github="https://raw.githubusercontent.com/agronomy4future/raw_data_practice/main/fertilizer_treatment.csv"
dataA= data.frame(read_csv(url(github), show_col_types = FALSE))
print(head(dataA,5))
Genotype Block variable value
1 Genotype_A I Control 42.9
2 Genotype_A II Control 41.6
3 Genotype_A III Control 28.9
4 Genotype_A IV Control 30.8
5 Genotype_B I Control 53.3
.
.
.
This data has two treatments (independent variables) with 4 replicates.
xtabs(~variable + Genotype, data=dataA)
Genotype variable Genotype_A Genotype_B Genotype_C Genotype_D Control 4 4 4 4 Fertilizer1 4 4 4 4 Fertilizer2 4 4 4 4 Fertilizer3 4 4 4 4
Now, I’ll calculate the ‘responsiveness’ of yield for each fertilizer in response to control.
if(!require(dplyr)) install.packages("dplyr")
library(dplyr)
dataA=data.frame(dataA %>%
group_by(Genotype) %>%
mutate(responsiveness=ifelse(variable!="Control",
(value-value[variable=="Control"])/value[variable=="Control"], NA)))
print(head(dataA,5))
Genotype Block variable value responsiveness
1 Genotype_A I Control 42.9 NA
2 Genotype_A II Control 41.6 NA
3 Genotype_A III Control 28.9 NA
4 Genotype_A IV Control 30.8 NA
5 Genotype_B I Control 53.3 NA
I’ll check if this calculation is correct. Let’s download this data as an Excel file.
if(!require(writexl)) install.packages("writexl")
library(writexl)
write_xlsx (dataA,"C:/Users/Desktop/dataA.xlsx")
Then, I compared what I calculated manually with what R calculated, and they are the same.
How to calculate when several groups exist?
The previous data only includes replicates from a single group (genotype). My next question is how to calculate when multiple groups are present?
This dataset involves different planting rows, including outside, east, west, and middle divisions. It analyzes biomass yield per wheat head and stem in response to three conditions: control, tr1, and tr2. How should I calculate responsiveness for each group combination in this scenario? I already calculated responsiveness using excel.
library(readr) github="https://raw.githubusercontent.com/agronomy4future/raw_data_practice/main/responsiveness.csv" dataA= data.frame(read_csv(url(github),show_col_types = FALSE))
print(head(dataA,5)) tissue row plot treatment yield responsiveness 1 Head outside 5 Control 159.30 <NA> 2 Head outside 5 Tr1 59.70 -62.5% 3 Head outside 5 Tr2 97.50 -38.8% 4 Stem outside 5 Control 74.65 <NA> 5 Stem outside 5 Tr1 34.05 -54.4%
Now I’ll calculate the responsiveness using R by groups.
if(!require(dplyr)) install.packages("dplyr")
library(dplyr)
dataB=data.frame(dataA %>%
group_by(tissue, row, plot) %>%
mutate(Calculation=ifelse(treatment!="Control",
(yield-yield[treatment=="Control"])/yield[treatment=="Control"], NA)))
print(head(dataB,5)) tissue row plot treatment yield responsiveness Calculation 1 Head outside 5 Control 159.30 <NA> NA 2 Head outside 5 Tr1 59.70 -62.5% -0.6252354 3 Head outside 5 Tr2 97.50 -38.8% -0.3879473 4 Stem outside 5 Control 74.65 <NA> NA 5 Stem outside 5 Tr1 34.05 -54.4% -0.5438714
However, as you can see, the calculated value differs from what I calculated using excel This issue can occur when the ‘plyr’ package is installed. To resolve it, you can either uninstall the ‘plyr’ package or use the dplyr::mutate function.
if(!require(dplyr)) install.packages("dplyr")
library(dplyr)
dataB=data.frame(dataA %>%
group_by(tissue, row, plot) %>%
dplyr::mutate(Calculation=ifelse(treatment!="Control",
(yield-yield[treatment=="Control"])/yield[treatment=="Control"], NA)))
print(head(dataB,5)) tissue row plot treatment yield responsiveness Calculation 1 Head outside 5 Control 159.30 <NA> NA 2 Head outside 5 Tr1 59.70 -62.5% -0.6252354 3 Head outside 5 Tr2 97.50 -38.8% -0.3879473 4 Stem outside 5 Control 74.65 <NA> NA 5 Stem outside 5 Tr1 34.05 -54.4% -0.5438714
Now, the value will be the same.
How about calculating responsiveness using the mean?
Next, I want to determine responsiveness by averaging the yield. Therefore, the first step is to calculate the average yield.
library (plyr)
dataC=ddply(dataA, c("tissue","row","treatment"), summarise, mean=mean(yield))
colnames(dataC)[4]=c("yield")
dataC
tissue row treatment yield
1 Head east Control 71.8000
2 Head east Tr1 52.7375
3 Head east Tr2 45.1750
4 Head middle Control 62.9000
5 Head middle Tr1 57.8250
6 Head middle Tr2 37.2250
7 Head outside Control 118.0500
8 Head outside Tr1 137.6500
9 Head outside Tr2 96.8250
10 Head west Control 73.8750
11 Head west Tr1 60.9125
12 Head west Tr2 47.6000
13 Stem east Control 48.5500
14 Stem east Tr1 35.1875
15 Stem east Tr2 53.9375
16 Stem middle Control 41.3750
17 Stem middle Tr1 36.3750
18 Stem middle Tr2 63.6250
19 Stem outside Control 66.9500
20 Stem outside Tr1 71.7750
21 Stem outside Tr2 79.1000
22 Stem west Control 50.6250
23 Stem west Tr1 38.1500
24 Stem west Tr2 55.2250
Then, I’ll calculate responsiveness again.
library(dplyr)
dataD=data.frame(dataC %>%
group_by(tissue, row) %>%
dplyr::mutate(responsiveness=ifelse(treatment!="Control",
(yield-yield[treatment=="Control"])/yield[treatment=="Control"], NA)))
dataD
tissue row treatment yield responsiveness
1 Head east Control 71.8000 NA
2 Head east Tr1 52.7375 -0.26549443
3 Head east Tr2 45.1750 -0.37082173
4 Head middle Control 62.9000 NA
5 Head middle Tr1 57.8250 -0.08068362
6 Head middle Tr2 37.2250 -0.40818760
7 Head outside Control 118.0500 NA
8 Head outside Tr1 137.6500 0.16603134
9 Head outside Tr2 96.8250 -0.17979670
10 Head west Control 73.8750 NA
11 Head west Tr1 60.9125 -0.17546531
12 Head west Tr2 47.6000 -0.35566836
13 Stem east Control 48.5500 NA
14 Stem east Tr1 35.1875 -0.27523172
15 Stem east Tr2 53.9375 0.11096807
16 Stem middle Control 41.3750 NA
17 Stem middle Tr1 36.3750 -0.12084592
18 Stem middle Tr2 63.6250 0.53776435
19 Stem outside Control 66.9500 NA
20 Stem outside Tr1 71.7750 0.07206871
21 Stem outside Tr2 79.1000 0.18147872
22 Stem west Control 50.6250 NA
23 Stem west Tr1 38.1500 -0.24641975
24 Stem west Tr2 55.2250 0.09086420
Tip!! How about we consider reshaping the data format??
Let’s go back to dataA. After going through the above process to calculate responsiveness due to variables being in the same column, how about we reshape the data format to facilitate the calculation of responsiveness more easily?
dataA
Genotype Block variable value
1 Genotype_A I Control 42.9
2 Genotype_A II Control 41.6
3 Genotype_A III Control 28.9
4 Genotype_A IV Control 30.8
5 Genotype_B I Control 53.3
6 Genotype_B II Control 69.6
7 Genotype_B III Control 45.4
8 Genotype_B IV Control 35.1
9 Genotype_C I Control 62.3
10 Genotype_C II Control 58.5
.
.
.
Let’s use the below code to reshape data format.
library(dplyr)
library(tidyr)
dataB= data.frame(dataA %>%
group_by(Genotype, Block, variable) %>%
spread(key=variable, value=value))
dataB
Genotype Block Control Fertilizer1 Fertilizer2 Fertilizer3
1 Genotype_A I 42.9 53.8 49.5 44.4
2 Genotype_A II 41.6 58.5 53.8 41.8
3 Genotype_A III 28.9 43.9 40.7 28.3
4 Genotype_A IV 30.8 46.3 39.4 34.7
5 Genotype_B I 53.3 57.6 59.8 64.1
6 Genotype_B II 69.6 69.6 65.8 57.4
7 Genotype_B III 45.4 42.4 41.4 44.1
8 Genotype_B IV 35.1 51.9 45.4 51.6
9 Genotype_C I 62.3 63.4 64.5 63.6
10 Genotype_C II 58.5 50.4 46.1 56.1
11 Genotype_C III 44.6 45.0 62.6 52.7
12 Genotype_C IV 50.3 46.7 50.3 51.8
13 Genotype_D I 75.4 70.3 68.8 71.6
14 Genotype_D II 65.6 67.3 65.3 69.4
15 Genotype_D III 54.0 57.6 45.6 56.6
16 Genotype_D IV 52.7 58.5 51.0 47.4
Then we can easily calculate the responsiveness.
dataB$R_Fertilizer1=(dataB$Fertilizer1-dataB$Control)/dataB$Control
dataB$R_Fertilizer2=(dataB$Fertilizer2-dataB$Control)/dataB$Control
dataB$R_Fertilizer3=(dataB$Fertilizer2-dataB$Control)/dataB$Control
dataB
Genotype Block Control Fertilizer1 Fertilizer2 Fertilizer3 R_Fertilizer1 R_Fertilizer2 R_Fertilizer3
1 Genotype_A I 42.9 53.8 49.5 44.4 0.25407925 0.153846154 0.153846154
2 Genotype_A II 41.6 58.5 53.8 41.8 0.40625000 0.293269231 0.293269231
3 Genotype_A III 28.9 43.9 40.7 28.3 0.51903114 0.408304498 0.408304498
4 Genotype_A IV 30.8 46.3 39.4 34.7 0.50324675 0.279220779 0.279220779
5 Genotype_B I 53.3 57.6 59.8 64.1 0.08067542 0.121951220 0.121951220
6 Genotype_B II 69.6 69.6 65.8 57.4 0.00000000 -0.054597701 -0.054597701
7 Genotype_B III 45.4 42.4 41.4 44.1 -0.06607930 -0.088105727 -0.088105727
8 Genotype_B IV 35.1 51.9 45.4 51.6 0.47863248 0.293447293 0.293447293
9 Genotype_C I 62.3 63.4 64.5 63.6 0.01765650 0.035313002 0.035313002
10 Genotype_C II 58.5 50.4 46.1 56.1 -0.13846154 -0.211965812 -0.211965812
11 Genotype_C III 44.6 45.0 62.6 52.7 0.00896861 0.403587444 0.403587444
12 Genotype_C IV 50.3 46.7 50.3 51.8 -0.07157058 0.000000000 0.000000000
13 Genotype_D I 75.4 70.3 68.8 71.6 -0.06763926 -0.087533156 -0.087533156
14 Genotype_D II 65.6 67.3 65.3 69.4 0.02591463 -0.004573171 -0.004573171
15 Genotype_D III 54.0 57.6 45.6 56.6 0.06666667 -0.155555556 -0.155555556
16 Genotype_D IV 52.7 58.5 51.0 47.4 0.11005693 -0.032258065 -0.032258065
Then, let’s reshape data again.
dataC= data.frame(pivot_longer(dataB,
cols= starts_with("R_Fertilizer"),
names_to= "Responsiveness",
values_to= "Value"))
dataC
Genotype Block Control Fertilizer1 Fertilizer2 Fertilizer3 Responsiveness Value
1 Genotype_A I 42.9 53.8 49.5 44.4 R_Fertilizer1 0.254079254
2 Genotype_A I 42.9 53.8 49.5 44.4 R_Fertilizer2 0.153846154
3 Genotype_A I 42.9 53.8 49.5 44.4 R_Fertilizer3 0.153846154
4 Genotype_A II 41.6 58.5 53.8 41.8 R_Fertilizer1 0.406250000
5 Genotype_A II 41.6 58.5 53.8 41.8 R_Fertilizer2 0.293269231
6 Genotype_A II 41.6 58.5 53.8 41.8 R_Fertilizer3 0.293269231
7 Genotype_A III 28.9 43.9 40.7 28.3 R_Fertilizer1 0.519031142
8 Genotype_A III 28.9 43.9 40.7 28.3 R_Fertilizer2 0.408304498
9 Genotype_A III 28.9 43.9 40.7 28.3 R_Fertilizer3 0.408304498
10 Genotype_A IV 30.8 46.3 39.4 34.7 R_Fertilizer1 0.503246753
We aim to develop open-source code for agronomy ([email protected])
© 2022 – 2025 https://agronomy4future.com – All Rights Reserved.
Last Updated: 05/11/2023