Chapter 1 Introduction

This is the supplemental material associated with the 6th chapter in my dissertation.

1.1 Computer Setup

These analyses were conducted in the following computing environment:

print(version)

##                _                           
## platform       x86_64-pc-linux-gnu         
## arch           x86_64                      
## os             linux-gnu                   
## system         x86_64, linux-gnu           
## status                                     
## major          4                           
## minor          2.3                         
## year           2023                        
## month          03                          
## day            15                          
## svn rev        83980                       
## language       R                           
## version.string R version 4.2.3 (2023-03-15)
## nickname       Shortstop Beagle

1.2 Experimental setup

Setting up required variables variables.

# libraries we are using
library(ggplot2)
library(cowplot)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(PupillometryR)

## Loading required package: rlang

p_theme <- theme(
  text = element_text(size = 28),
  plot.title = element_text( face = "bold", size = 22, hjust = 0.5),
  panel.border = element_blank(),
  panel.grid.minor = element_blank(),
  legend.title=element_text(size=18),
  legend.text=element_text(size=18),
  axis.title = element_text(size=20),
  axis.text = element_text(size=18),
  legend.position="bottom",
  panel.background = element_rect(fill = "#f1f2f5",
                                  colour = "white",
                                  size = 0.5, linetype = "solid")
)

## Warning: The `size` argument of `element_rect()` is deprecated as of ggplot2 3.4.0.
## i Please use the `linewidth` argument instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

# default variables

MODEL = c('EA','IS','NMIS')
STRUCTURE = c('Well-mixed','Standard islands','Isolated islands')
EXPERIMENTS = c('BASE-EXPERIMENTS/','MI50/','MI5000/')
SCHEME = c('TRUNCATION','TOURNAMENT','LEXICASE')
DIAGNOSTIC = c('EXPLOITATION_RATE', 'ORDERED_EXPLOITATION', 'CONTRADICTORY_OBJECTIVES', 'MULTIPATH_EXPLORATION')
DIMENSIONALITY = 100
cb_palette <- c('#D81B60','#1E88E5','#FFC107')
cb_palette_mi <- c('#E66100','#1E88E5','#5D3A9B')
SHAPE = c(15,16,17)
TSIZE = 20
GENERATIONS = 50000

# data related
DATA_DIR = '/opt/Diagnosing-Island-Structures/DATA-FINAL/'

# go through each diagnostic and collect over time data for cross comparison (cc)
base_over_time = data.frame()
mi50_over_time = data.frame()
mi5000_over_time = data.frame()
print('over time data')

## [1] "over time data"

for(model in MODEL)
{
  print(model)
  for(scheme in SCHEME)
  {
    base_dir = paste(DATA_DIR,EXPERIMENTS[1],model,'/over-time-',scheme, '.csv', sep = "", collapse = NULL)
    base_over_time = rbind(base_over_time, read.csv(base_dir, header = TRUE, stringsAsFactors = FALSE))

    mi50_dir = paste(DATA_DIR,EXPERIMENTS[2],model,'/over-time-',scheme, '.csv', sep = "", collapse = NULL)
    mi50_over_time = rbind(mi50_over_time, read.csv(mi50_dir, header = TRUE, stringsAsFactors = FALSE))

    mi5000_dir = paste(DATA_DIR,EXPERIMENTS[3],model,'/over-time-',scheme, '.csv', sep = "", collapse = NULL)
    mi5000_over_time = rbind(mi5000_over_time, read.csv(mi5000_dir, header = TRUE, stringsAsFactors = FALSE))

  }
}

## [1] "EA"
## [1] "IS"
## [1] "NMIS"

colnames(base_over_time)[colnames(base_over_time) == "SEL"] = 'Selection\nScheme'
base_over_time$Structure <- factor(base_over_time$Structure, levels = MODEL)
base_over_time$sel_pre = base_over_time$sel_pre * -1.0
base_over_time$`Population structure` <-
  with(base_over_time, ifelse(Structure == 'EA', 'Well-mixed',
                        ifelse(Structure == 'IS', 'Standard islands',
                               ifelse(Structure == 'NMIS', 'Isolated islands',''))))
base_over_time$`Population structure` <- factor(base_over_time$`Population structure`, levels = STRUCTURE)


colnames(mi50_over_time)[colnames(mi50_over_time) == "SEL"] = 'Selection\nScheme'
mi50_over_time$Structure <- factor(mi50_over_time$Structure, levels = MODEL)
mi50_over_time$sel_pre = mi50_over_time$sel_pre * -1.0
mi50_over_time$`Population structure` <-
  with(mi50_over_time, ifelse(Structure == 'EA', 'Well-mixed',
                              ifelse(Structure == 'IS', 'Standard islands',
                                     ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi50_over_time$`Population structure` <- factor(mi50_over_time$`Population structure`, levels = STRUCTURE)


colnames(mi5000_over_time)[colnames(mi5000_over_time) == "SEL"] = 'Selection\nScheme'
mi5000_over_time$Structure <- factor(mi5000_over_time$Structure, levels = MODEL)
mi5000_over_time$sel_pre = mi5000_over_time$sel_pre * -1.0
mi5000_over_time$`Population structure` <-
  with(mi5000_over_time, ifelse(Structure == 'EA', 'Well-mixed',
                              ifelse(Structure == 'IS', 'Standard islands',
                                     ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi5000_over_time$`Population structure` <- factor(mi5000_over_time$`Population structure`, levels = STRUCTURE)


# go through each diagnostic and collect best over time for cross comparison (cc)
base_best = data.frame()
mi50_best = data.frame()
mi5000_best = data.frame()
print('best data')

## [1] "best data"

for(model in MODEL)
{
  print(model)
  for(scheme in SCHEME)
  {
    base_dir = paste(DATA_DIR,EXPERIMENTS[1],model,'/best-',scheme, '.csv', sep = "", collapse = NULL)
    base_best = rbind(base_best, read.csv(base_dir, header = TRUE, stringsAsFactors = FALSE))

    mi50_dir = paste(DATA_DIR,EXPERIMENTS[2],model,'/best-',scheme, '.csv', sep = "", collapse = NULL)
    mi50_best = rbind(mi50_best, read.csv(mi50_dir, header = TRUE, stringsAsFactors = FALSE))

    mi5000_dir = paste(DATA_DIR,EXPERIMENTS[3],model,'/best-',scheme, '.csv', sep = "", collapse = NULL)
    mi5000_best = rbind(mi5000_best, read.csv(mi5000_dir, header = TRUE, stringsAsFactors = FALSE))
  }
}

## [1] "EA"
## [1] "IS"
## [1] "NMIS"

colnames(base_best)[colnames(base_best) == "SEL"] = 'Selection\nScheme'
base_best$Structure <- factor(base_best$Structure, levels = MODEL)
base_best$`Population structure` <-
  with(base_best, ifelse(Structure == 'EA', 'Well-mixed',
                              ifelse(Structure == 'IS', 'Standard islands',
                                     ifelse(Structure == 'NMIS', 'Isolated islands',''))))
base_best$`Population structure` <- factor(base_best$`Population structure`, levels = STRUCTURE)


colnames(mi50_best)[colnames(mi50_best) == "SEL"] = 'Selection\nScheme'
mi50_best$Structure <- factor(mi50_best$Structure, levels = MODEL)
mi50_best$`Population structure` <-
  with(mi50_best, ifelse(Structure == 'EA', 'Well-mixed',
                         ifelse(Structure == 'IS', 'Standard islands',
                                ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi50_best$`Population structure` <- factor(mi50_best$`Population structure`, levels = STRUCTURE)


colnames(mi5000_best)[colnames(mi5000_best) == "SEL"] = 'Selection\nScheme'
mi5000_best$Structure <- factor(mi5000_best$Structure, levels = MODEL)
mi5000_best$`Population structure` <-
  with(mi5000_best, ifelse(Structure == 'EA', 'Well-mixed',
                         ifelse(Structure == 'IS', 'Standard islands',
                                ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi5000_best$`Population structure` <- factor(mi5000_best$`Population structure`, levels = STRUCTURE)

# get generation a satisfactory solution is found for cross comparison (cc)
base_ssf = data.frame()
mi50_ssf = data.frame()
mi5000_ssf = data.frame()
print('ssf data')

## [1] "ssf data"

for(model in MODEL)
{
  print(model)
  for(scheme in SCHEME)
  {
    base_dir = paste(DATA_DIR,EXPERIMENTS[1],model,'/ssf-',scheme, '.csv', sep = "", collapse = NULL)
    base_ssf = rbind(base_ssf, read.csv(base_dir, header = TRUE, stringsAsFactors = FALSE))

    mi50_dir = paste(DATA_DIR,EXPERIMENTS[2],model,'/ssf-',scheme, '.csv', sep = "", collapse = NULL)
    mi50_ssf = rbind(mi50_ssf, read.csv(mi50_dir, header = TRUE, stringsAsFactors = FALSE))

    mi5000_dir = paste(DATA_DIR,EXPERIMENTS[3],model,'/ssf-',scheme, '.csv', sep = "", collapse = NULL)
    mi5000_ssf = rbind(mi5000_ssf, read.csv(mi5000_dir, header = TRUE, stringsAsFactors = FALSE))
  }
}

## [1] "EA"
## [1] "IS"
## [1] "NMIS"

colnames(base_ssf)[colnames(base_ssf) == "SEL"] = 'Selection\nScheme'
base_ssf$Structure <- factor(base_ssf$Structure, levels = MODEL)
base_ssf$`Population structure` <-
  with(base_ssf, ifelse(Structure == 'EA', 'Well-mixed',
                  ifelse(Structure == 'IS', 'Standard islands',
                  ifelse(Structure == 'NMIS', 'Isolated islands',''))))
base_ssf$`Population structure` <- factor(base_ssf$`Population structure`, levels = STRUCTURE)



colnames(mi50_ssf)[colnames(mi50_ssf) == "SEL"] = 'Selection\nScheme'
mi50_ssf$Structure <- factor(mi50_ssf$Structure, levels = MODEL)
mi50_ssf$`Population structure` <-
  with(mi50_ssf, ifelse(Structure == 'EA', 'Well-mixed',
                        ifelse(Structure == 'IS', 'Standard islands',
                               ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi50_ssf$`Population structure` <- factor(mi50_ssf$`Population structure`, levels = STRUCTURE)


colnames(mi5000_ssf)[colnames(mi5000_ssf) == "SEL"] = 'Selection\nScheme'
mi5000_ssf$Structure <- factor(mi5000_ssf$Structure, levels = MODEL)
mi5000_ssf$`Population structure` <-
  with(mi5000_ssf, ifelse(Structure == 'EA', 'Well-mixed',
                        ifelse(Structure == 'IS', 'Standard islands',
                               ifelse(Structure == 'NMIS', 'Isolated islands',''))))
mi5000_ssf$`Population structure` <- factor(mi5000_ssf$`Population structure`, levels = STRUCTURE)