knitr::opts_chunk$set(echo = TRUE, warning = FALSE)
Import Data
library(tidyverse)
library(readxl)
library(janitor)
fhb_spe1<- read_excel("data/dat-isolates.xlsx")
Toxin type frequency
## Frequency of genotypes
fhb_spe1 %>%
filter(tri!="NA",
year!="2012") %>%
tabyl(tri)
### 2018
tri_2018 <- fhb_spe1 %>%
filter(year =="2018") %>%
filter(tri!="NA")
tri_2018 %>%
tabyl(location, tri, year)
## $`2018`
## location 15ADON 3ADON NIV
## Armstrong 25 2 1
## Centre 156 5 0
## Lancaster 35 0 0
## Lebanon 3 0 0
## Potter 7 6 2
## Tioga 32 3 0
## York 34 0 0
### 2019
tri_2019 <- fhb_spe1 %>%
filter(year =="2019") %>%
filter(tri!="NA")
tri_2019 %>%
tabyl(location, tri, year)
## $`2019`
## location 15ADON 3ADON NX-2
## Centre 14 0 0
## Crawford 3 0 0
## Erie 11 0 0
## Lancaster 38 0 0
## Lebanon 30 1 1
## Potter 5 0 0
### 2012
tri_2012 <- fhb_spe1 %>%
filter(year =="2012") %>%
filter(tri!="NA")
tri_2012 %>%
tabyl(year, tri)
Independence tests
Toxin type vs. Year
tri_year <- fhb_spe1 %>%
filter(tri!="NA",
year!="2012") %>%
tabyl(tri, year)
fisher.test(tri_year, simulate.p.value = TRUE)
##
## Fisher's Exact Test for Count Data with simulated p-value
## (based on 2000 replicates)
##
## data: tri_year
## p-value = 0.05347
## alternative hypothesis: two.sided
# No significant association between trichothecene genotype and year of sampling.
Toxin type vs. location - 2018
tri_county_2018 <- fhb_spe1 %>%
filter(year == "2018") %>%
filter(tri!="NA") %>%
tabyl(county, tri)
fisher.test(tri_county_2018, simulate.p.value = TRUE)
##
## Fisher's Exact Test for Count Data with simulated p-value
## (based on 2000 replicates)
##
## data: tri_county_2018
## p-value = 0.0004998
## alternative hypothesis: two.sided
# Significant association between trichothecene genotype and sampled location in 2018.
Toxin type vs. location - 2019
tri_county_2019 <- fhb_spe1 %>%
filter(year == "2019") %>%
filter(tri!="NA") %>%
tabyl(county, tri)
fisher.test(tri_county_2019, simulate.p.value = TRUE)
##
## Fisher's Exact Test for Count Data with simulated p-value
## (based on 2000 replicates)
##
## data: tri_county_2019
## p-value = 0.6447
## alternative hypothesis: two.sided
# No significant association between trichothecene genotype and sampled location in 2019.
Toxin type vs. host - 2018
host_tri_2018 <- fhb_spe1 %>%
filter(year == "2018") %>%
filter(tri!="NA") %>%
tabyl(crop, tri)
chisq.test(host_tri_2018)
##
## Pearson's Chi-squared test
##
## data: host_tri_2018
## X-squared = 2.9445, df = 2, p-value = 0.2294
fisher.test(host_tri_2018, simulate.p.value = TRUE)
##
## Fisher's Exact Test for Count Data with simulated p-value
## (based on 2000 replicates)
##
## data: host_tri_2018
## p-value = 0.2414
## alternative hypothesis: two.sided
# No significant association between trichothecene genotype and sampled host in 2018.
Toxin type vs. host - 2019
host_tri_2019 <- fhb_spe1 %>%
filter(year == "2019") %>%
filter(tri!="NA") %>%
tabyl(crop, tri)
fisher.test(host_tri_2019, simulate.p.value = TRUE)
##
## Fisher's Exact Test for Count Data with simulated p-value
## (based on 2000 replicates)
##
## data: host_tri_2019
## p-value = 1
## alternative hypothesis: two.sided
# No significant association between trichothecene genotype and sampled host in 2019.
FGSC frequency
## Frequency of FGSC members
fhb_spe1 %>%
filter(Fg_16!="NA") %>%
tabyl(Fg_16)
### 2018
spc_2018 <- fhb_spe1 %>%
filter(Fg_16!="NA") %>%
filter(year =="2018")
spc_2018 %>%
tabyl(Fg_16)
### 2019
spc_2019 <- fhb_spe1 %>%
filter(Fg_16!="NA") %>%
filter(year =="2019")
spc_2019 %>%
tabyl(Fg_16)
Non-FGSC frequency
## TEF gene
fhb_spe1 %>%
filter(year!="2012",
TEF!="NA" ,
TEF!="Fusarium graminearum",
TEF!="Unidentified") %>%
tabyl(TEF)
Copyright 2020 Maira R. Duffeck