Skip to content

Instantly share code, notes, and snippets.

@sckott

sckott/phylometa_fxn.R

Created April 24, 2011 23:19
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 2 You must be signed in to fork a gist
  • Save sckott/939970 to your computer and use it in GitHub Desktop.
Save sckott/939970 to your computer and use it in GitHub Desktop.
Download ZIP
Conduct phylogenetic meta-analysis from R calling Phylometa - These are functions needed in the code in file phylometa_inR_run.R.
Raw
phylometa_fxn.R
#####################################################################
# Created by Scott Chamberlain
# Ecology and Evolutionary Biology Dept., Rice University
# Houston, TX 77005, USA
# myrmecocystus@gmail.com
#####################################################################
# Function to split confidence intervals, this function is required in the below functions
CI_split <- function(a){
dd<-c(gsub("\\(","",unlist(strsplit(a[5],","))[1]),gsub("\\)","",unlist(strsplit(a[5],","))[2]))
b22<-a[2:8]
b222<-b22[-4]
b222<-append(b222,dd,after=3)
b222
}
### Functions to run meta-analysis in Phylometa
# 1 group in the moderator variable
maketables_1group <- function(output){ #DONE IF/ELSE REFORMATTING
#Table 1, traditional
#a1<-unlist(strsplit(output[6]," +"))[-1] #Table header
a2 <- unlist(strsplit(output[8]," +"))[-1] #Data row 1, between groups, fixed effects
a3 <- unlist(strsplit(output[9]," +"))[-1] #Data row 2, within groups
a4 <- if(length(unlist(strsplit(output[10]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[10]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:8])} else
end
a7 <- unlist(strsplit(output[11]," +"))[-1] #Data row 6, Total
a8 <- unlist(strsplit(output[14]," +"))[-1] #Data row 7, between groups, random effects
rownaames<-c(a2[1],a3[1],a4[1],a7[1],a8[1]) #make vector of first column names
numbers<-t(data.frame(as.numeric(a2[2:4]),as.numeric(a3[2:4]),as.numeric(a4[2:4]),append(as.numeric(a7[2]),c(-9999,-9999),after=1),as.numeric(a8[2:4]))) #make data frame of numbers
numbers<-data.frame(numbers,rownaames) #Make data frame of row names and numbers
colnames(numbers)<-c("Q","df","P","Source") #Assign column names
summaryfitstats_table1_trad<-data.frame(Source=numbers$Source,Q=numbers$Q,df=numbers$df,P=numbers$P) #Reorder columns
#summaryfitstats_table1_trad #See table
#Table 2, traditional
#b1<-unlist(strsplit(output[23]," +"))[-1] #Table header
b2 <- if(length(unlist(strsplit(output[27]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[27]," +"))[-1])[1],(unlist(strsplit(output[27]," +"))[-1])[2]),unlist(strsplit(output[27]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[27]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[27]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[27]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[27]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[27]," +"))[-1][6:7])} else
end
b3 <- if(length(unlist(strsplit(output[28]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[28]," +"))[-1])[1],(unlist(strsplit(output[28]," +"))[-1])[2]),unlist(strsplit(output[28]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[28]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[28]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[28]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[28]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[28]," +"))[-1][6:7])} else
end
b6 <- if(length(unlist(strsplit(output[32]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[32]," +"))[-1])[1],(unlist(strsplit(output[32]," +"))[-1])[2]),unlist(strsplit(output[32]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[32]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[32]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[32]," +"))[-1][6:7])} else
end
b7 <- if(length(unlist(strsplit(output[33]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[33]," +"))[-1])[1],(unlist(strsplit(output[33]," +"))[-1])[2]),unlist(strsplit(output[33]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[33]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[33]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[33]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[33]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[33]," +"))[-1][6:7])} else
end
rownaames1<-c(b2[1],b3[1],b6[1],b7[1]) #make vector of first column names
b22<-CI_split(b2) #Split confidence interval numbers apart and reinsert to vector of data
b33<-CI_split(b3)
b66<-CI_split(b6)
b77<-CI_split(b7)
numbers1<-t(data.frame(as.numeric(b22),as.numeric(b33),as.numeric(b66),as.numeric(b77))) #make data frame of numbers
numbers1<-data.frame(numbers1,rownaames1) #Make data frame of row names and numbers
colnames(numbers1)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_trad<-data.frame(Group=numbers1[,9],k=numbers1[,1],effsize=numbers1[,2],var=numbers1[,3],CI_low=numbers1[,4],CI_high=numbers1[,5],Z=numbers1[,6],df=numbers1[,7],P=numbers1[,8]) #Reorder columns
#summaryeffsizes_table2_trad #See table
#Table 1, phylogenetic
#d1 <- unlist(strsplit(output[47]," +"))[-1] #Table header
d2 <- unlist(strsplit(output[49]," +"))[-1] #Data row 1, between groups, fixed effects
d3 <- unlist(strsplit(output[50]," +"))[-1] #Data row 2, within groups
d4 <- if(length(unlist(strsplit(output[51]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[51]," +"))[-1][1],unlist(strsplit(output[51]," +"))[-1][2],unlist(strsplit(output[51]," +"))[-1][3]),unlist(strsplit(output[51]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[51]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[51]," +"))[-1][1],unlist(strsplit(output[51]," +"))[-1][2],unlist(strsplit(output[51]," +"))[-1][3]),unlist(strsplit(output[51]," +"))[-1][4:8])} else
end
d7 <- unlist(strsplit(output[52]," +"))[-1] #Data row 6, Total
d8 <- unlist(strsplit(output[55]," +"))[-1] #Data row 7, between groups, random effects
rownaames2<-c(d2[1],d3[1],d4[1],d7[1],d8[1]) #make vector of first column names
numbers2<-t(data.frame(append(as.numeric(d2[2:4]),c(-9999,-9999),after=3),as.numeric(d3[2:6]),as.numeric(d4[2:6]),append(as.numeric(d7[2]),c(-9999,-9999,-9999,-9999),after=1),append(as.numeric(d8[2:4]),c(-9999,-9999),after=3))) #make data frame of numbers
numbers2<-data.frame(numbers2,rownaames2) #Make data frame of row names and numbers
colnames(numbers2)<-c("Q","df","P","df_polytadj","P_polytadj","Source") #Assign column names
summaryfitstats_table1_phyl<-data.frame(Source=numbers2[,6],Q=numbers2[,1],df=numbers2[,2],P=numbers2[,3],df_polytadj=numbers2[,4],P_polytadj=numbers2[,5]) #Reorder columns
#summaryfitstats_table1_phyl #See table
#Table 2, phylogenetic
#e1<-unlist(strsplit(output[64]," +"))[-1] #Table header
e2 <- if(length(unlist(strsplit(output[68]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[68]," +"))[-1])[1],(unlist(strsplit(output[68]," +"))[-1])[2]),unlist(strsplit(output[68]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[68]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[68]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[68]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[68]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[68]," +"))[-1][6:7])} else
end
e3 <- if(length(unlist(strsplit(output[69]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[69]," +"))[-1])[1],(unlist(strsplit(output[69]," +"))[-1])[2]),unlist(strsplit(output[69]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[69]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[69]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[69]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[69]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[69]," +"))[-1][6:7])} else
end
e6 <- if(length(unlist(strsplit(output[73]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[73]," +"))[-1])[1],(unlist(strsplit(output[73]," +"))[-1])[2]),unlist(strsplit(output[73]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[73]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[73]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[73]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[73]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[73]," +"))[-1][6:7])} else
end
e7 <- if(length(unlist(strsplit(output[74]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[74]," +"))[-1])[1],(unlist(strsplit(output[74]," +"))[-1])[2]),unlist(strsplit(output[74]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[74]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[74]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[74]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[74]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[74]," +"))[-1][6:7])} else
end
rownaames3<-c(e2[1],e3[1],e6[1],e7[1]) #make vector of first column names
e22<-CI_split(e2) #Split confidence interval numbers apart and reinsert to vector of data
e33<-CI_split(e3)
e66<-CI_split(e6)
e77<-CI_split(e7)
numbers3<-t(data.frame(as.numeric(e22),as.numeric(e33),as.numeric(e66),as.numeric(e77))) #make data frame of numbers
numbers3<-data.frame(numbers3,rownaames3) #Make data frame of row names and numbers
colnames(numbers3)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_phyl<-data.frame(Group=numbers3[,9],k=numbers3[,1],effsize=numbers3[,2],var=numbers3[,3],CI_low=numbers3[,4],CI_high=numbers3[,5],Z=numbers3[,6],df=numbers3[,7],P=numbers3[,8]) #Reorder columns
#summaryeffsizes_table2_phyl #See table
### Model fit table, AIC
f1<-unlist(strsplit(output[89]," +"))[-1] #Table header
f2<-unlist(strsplit(output[90]," +"))[-1] #Data row 1, fixed effects, all studies
rownaames4<-c(f1[1],f2[1]) #make vector of first column names
numbers4<-t(data.frame(as.numeric(f1[2:3]),as.numeric(f2[2:3]))) #make data frame of numbers
numbers4<-data.frame(numbers4,rownaames4) #Make data frame of row names and numbers
colnames(numbers4)<-c("AIC_fixed","AIC_random","Analysis") #Assign column names
summarymodelfits_AIC_table<-data.frame(Analysis=numbers4[,3],AIC_fixed=numbers4[,1],AIC_random=numbers4[,2]) #Reorder columns
#summarymodelfits_table #See table
### Model fit table, -2(likelihood)
g1 <- unlist(strsplit(output[101]," +"))[-1] #Traditional AIC data
g2 <- unlist(strsplit(output[102]," +"))[-1] #Phylogenetically controlled AIC data
rownaames5 <- c(g1[1],g2[1]) #make vector of first column names
numbers5 <- t(data.frame(as.numeric(g1[2:3]), as.numeric(g2[2:3]))) #make data frame of numbers
numbers5 <- data.frame(numbers5, rownaames5) #Make data frame of row names and numbers
colnames(numbers5) <- c("twoln_fixed", "twoln_random", "Analysis") #Assign column names
summarymodelfits_twoln_table <- data.frame(Analysis = numbers5[,3], twoln_fixed = numbers5[,1], twoln_random = numbers5[,2]) #Reorder columns
#summarymodelfits_twoln_table #See table
#Make list of tables
tables <- list(summaryfitstats_table1_trad, summaryeffsizes_table2_trad, summaryfitstats_table1_phyl, summaryeffsizes_table2_phyl, summarymodelfits_AIC_table, summarymodelfits_twoln_table)
tables
}
# 2 groups in the moderator variable
maketables_2group <- function(output){ #DONE IF/ELSE REFORMATTING
#Table 1, Traditional meta-analysis
a1 <- unlist(strsplit(output[6]," +"))[-1] #Table header
a2 <- unlist(strsplit(output[8]," +"))[-1] #Data row 1, between groups, fixed effects
a3 <- unlist(strsplit(output[9]," +"))[-1] #Data row 2, within groups
a4 <- if(length(unlist(strsplit(output[10]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[10]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:8])} else
end
a5 <- if(length(unlist(strsplit(output[11]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[11]," +"))[-1][1],unlist(strsplit(output[11]," +"))[-1][2],unlist(strsplit(output[11]," +"))[-1][3]),unlist(strsplit(output[11]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[11]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[11]," +"))[-1][1],unlist(strsplit(output[11]," +"))[-1][2],unlist(strsplit(output[11]," +"))[-1][3]),unlist(strsplit(output[11]," +"))[-1][4:8])} else
end
a7 <- unlist(strsplit(output[12]," +"))[-1] #Data row 6, Total
a8 <- unlist(strsplit(output[15]," +"))[-1] #Data row 7, between groups, random effects
rownaames<-c(a2[1],a3[1],a4[1],a5[1],a7[1],a8[1]) #make vector of first column names
numbers<-t(data.frame(as.numeric(a2[2:4]),as.numeric(a3[2:4]),as.numeric(a4[2:4]),as.numeric(a5[2:4]),append(as.numeric(a7[2]),c(-9999,-9999),after=1),as.numeric(a8[2:4]))) #make data frame of numbers
numbers<-data.frame(numbers,rownaames) #Make data frame of row names and numbers
colnames(numbers)<-c("Q","df","P","Source") #Assign column names
summaryfitstats_table1_trad<-data.frame(Source=numbers$Source,Q=numbers$Q,df=numbers$df,P=numbers$P) #Reorder columns
#summaryfitstats_table1_trad #See table
#Table 2, Traditional meta-analysis
#b1<-unlist(strsplit(output[24]," +"))[-1] #Table header
b2 <- if(length(unlist(strsplit(output[28]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[28]," +"))[-1])[1],(unlist(strsplit(output[28]," +"))[-1])[2]),unlist(strsplit(output[28]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[28]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[28]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[28]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[28]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[28]," +"))[-1][6:7])} else
end
b3 <- if(length(unlist(strsplit(output[29]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[29]," +"))[-1])[1],(unlist(strsplit(output[29]," +"))[-1])[2]),unlist(strsplit(output[29]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[29]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[29]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[29]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[29]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[29]," +"))[-1][6:7])} else
end
b4 <- if(length(unlist(strsplit(output[30]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[30]," +"))[-1])[1],(unlist(strsplit(output[30]," +"))[-1])[2]),unlist(strsplit(output[30]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[30]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[30]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[30]," +"))[-1][6:7])} else
end
b6 <- if(length(unlist(strsplit(output[34]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[34]," +"))[-1])[1],(unlist(strsplit(output[34]," +"))[-1])[2]),unlist(strsplit(output[34]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[34]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[34]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[34]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[34]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[34]," +"))[-1][6:7])} else
end
b7 <- if(length(unlist(strsplit(output[35]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[35]," +"))[-1])[1],(unlist(strsplit(output[35]," +"))[-1])[2]),unlist(strsplit(output[35]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[35]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[35]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[35]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[35]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[35]," +"))[-1][6:7])} else
end
b8 <- if(length(unlist(strsplit(output[36]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[36]," +"))[-1])[1],(unlist(strsplit(output[36]," +"))[-1])[2]),unlist(strsplit(output[36]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[36]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[36]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[36]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[36]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[36]," +"))[-1][6:7])} else
end
rownaames1<-c(b2[1],b3[1],b4[1],b6[1],b7[1],b8[1]) #make vector of first column names
b22<-CI_split(b2) #Split confidence interval numbers apart and reinsert to vector of data
b33<-CI_split(b3)
b44<-CI_split(b4)
b66<-CI_split(b6)
b77<-CI_split(b7)
b88<-CI_split(b8)
numbers1<-t(data.frame(as.numeric(b22),as.numeric(b33),as.numeric(b44),as.numeric(b66),as.numeric(b77),as.numeric(b88))) #make data frame of numbers
numbers1<-data.frame(numbers1,rownaames1) #Make data frame of row names and numbers
colnames(numbers1)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_trad<-data.frame(Group=numbers1[,9],k=numbers1[,1],effsize=numbers1[,2],var=numbers1[,3],CI_low=numbers1[,4],CI_high=numbers1[,5],Z=numbers1[,6],df=numbers1[,7],P=numbers1[,8]) #Reorder columns
#summaryeffsizes_table2_trad #See table
###Phylogenetic meta-analysis
#Table 1, Phylogenetic meta-analysis
d1<-unlist(strsplit(output[50]," +"))[-1] #Table header
d2<-unlist(strsplit(output[52]," +"))[-1] #Data row 1, between groups, fixed effects
d3<-unlist(strsplit(output[53]," +"))[-1] #Data row 2, within groups
d4 <- if(length(unlist(strsplit(output[54]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[54]," +"))[-1][1],unlist(strsplit(output[54]," +"))[-1][2],unlist(strsplit(output[54]," +"))[-1][3]),unlist(strsplit(output[54]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[54]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[54]," +"))[-1][1],unlist(strsplit(output[54]," +"))[-1][2],unlist(strsplit(output[54]," +"))[-1][3]),unlist(strsplit(output[54]," +"))[-1][4:8])} else
end
d5 <- if(length(unlist(strsplit(output[55]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[55]," +"))[-1][1],unlist(strsplit(output[55]," +"))[-1][2],unlist(strsplit(output[55]," +"))[-1][3]),unlist(strsplit(output[55]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[55]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[55]," +"))[-1][1],unlist(strsplit(output[55]," +"))[-1][2],unlist(strsplit(output[55]," +"))[-1][3]),unlist(strsplit(output[55]," +"))[-1][4:8])} else
end
d7<-unlist(strsplit(output[56]," +"))[-1] #Data row 6, Total
d8<-unlist(strsplit(output[59]," +"))[-1] #Data row 7, between groups, random effects
rownaames2<-c(d2[1],d3[1],d4[1],d5[1],d7[1],d8[1]) #make vector of first column names
numbers2<-t(data.frame(append(as.numeric(d2[2:4]),c(-9999,-9999),after=3),as.numeric(d3[2:6]),as.numeric(d4[2:6]),as.numeric(d5[2:6]),append(as.numeric(d7[2]),c(-9999,-9999,-9999,-9999),after=1),append(as.numeric(d8[2:4]),c(-9999,-9999),after=3))) #make data frame of numbers
numbers2<-data.frame(numbers2,rownaames2) #Make data frame of row names and numbers
colnames(numbers2)<-c("Q","df","P","df_polytadj","P_polytadj","Source") #Assign column names
summaryfitstats_table1_phyl<-data.frame(Source=numbers2[,6],Q=numbers2[,1],df=numbers2[,2],P=numbers2[,3],df_polytadj=numbers2[,4],P_polytadj=numbers2[,5]) #Reorder columns
#summaryfitstats_table1_phyl #See table
#Table 2, Phylogenetic meta-analysis
#e1<-unlist(strsplit(output[68]," +"))[-1] #Table header
e2 <- if(length(unlist(strsplit(output[72]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[72]," +"))[-1])[1],(unlist(strsplit(output[72]," +"))[-1])[2]),unlist(strsplit(output[72]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[72]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[72]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[72]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[72]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[72]," +"))[-1][6:7])} else
end
e3 <- if(length(unlist(strsplit(output[73]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[73]," +"))[-1])[1],(unlist(strsplit(output[73]," +"))[-1])[2]),unlist(strsplit(output[73]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[73]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[73]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[73]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[73]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[73]," +"))[-1][6:7])} else
end
e4 <- if(length(unlist(strsplit(output[74]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[74]," +"))[-1])[1],(unlist(strsplit(output[74]," +"))[-1])[2]),unlist(strsplit(output[74]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[74]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[74]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[74]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[74]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[74]," +"))[-1][6:7])} else
end
e6 <- if(length(unlist(strsplit(output[78]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[78]," +"))[-1])[1],(unlist(strsplit(output[78]," +"))[-1])[2]),unlist(strsplit(output[78]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[78]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[78]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[78]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[78]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[78]," +"))[-1][6:7])} else
end
e7 <- if(length(unlist(strsplit(output[79]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[79]," +"))[-1])[1],(unlist(strsplit(output[79]," +"))[-1])[2]),unlist(strsplit(output[79]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[79]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[79]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[79]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[79]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[79]," +"))[-1][6:7])} else
end
e8 <- if(length(unlist(strsplit(output[80]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[80]," +"))[-1])[1],(unlist(strsplit(output[80]," +"))[-1])[2]),unlist(strsplit(output[80]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[80]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[80]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[80]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[80]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[80]," +"))[-1][6:7])} else
end
rownaames3<-c(e2[1],e3[1],e4[1],e6[1],e7[1],e8[1]) #make vector of first column names
e22<-CI_split(e2) #Split confidence interval numbers apart and reinsert to vector of data
e33<-CI_split(e3)
e44<-CI_split(e4)
e66<-CI_split(e6)
e77<-CI_split(e7)
e88<-CI_split(e8)
numbers3<-t(data.frame(as.numeric(e22),as.numeric(e33),as.numeric(e44),as.numeric(e66),as.numeric(e77),as.numeric(e88))) #make data frame of numbers
numbers3<-data.frame(numbers3,rownaames3) #Make data frame of row names and numbers
colnames(numbers3)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_phyl<-data.frame(Group=numbers3[,9],k=numbers3[,1],effsize=numbers3[,2],var=numbers3[,3],CI_low=numbers3[,4],CI_high=numbers3[,5],Z=numbers3[,6],df=numbers3[,7],P=numbers3[,8]) #Reorder columns
#summaryeffsizes_table2_phyl #See table
### Model fit table, AIC
f1<-unlist(strsplit(output[95]," +"))[-1] #Traditional AIC data
f2<-unlist(strsplit(output[96]," +"))[-1] #Phylogenetically controlled AIC data
rownaames4<-c(f1[1],f2[1]) #make vector of first column names
numbers4<-t(data.frame(as.numeric(f1[2:3]),as.numeric(f2[2:3]))) #make data frame of numbers
numbers4<-data.frame(numbers4,rownaames4) #Make data frame of row names and numbers
colnames(numbers4)<-c("AIC_fixed","AIC_random","Analysis") #Assign column names
summarymodelfits_AIC_table<-data.frame(Analysis=numbers4[,3],AIC_fixed=numbers4[,1],AIC_random=numbers4[,2]) #Reorder columns
#summarymodelfits_AIC_table #See table
### Model fit table, -2(likelihood)
g1 <- unlist(strsplit(output[107]," +"))[-1] #Traditional AIC data
g2 <- unlist(strsplit(output[108]," +"))[-1] #Phylogenetically controlled AIC data
rownaames5 <- c(g1[1],g2[1]) #make vector of first column names
numbers5 <- t(data.frame(as.numeric(g1[2:3]), as.numeric(g2[2:3]))) #make data frame of numbers
numbers5 <- data.frame(numbers5, rownaames5) #Make data frame of row names and numbers
colnames(numbers5) <- c("twoln_fixed", "twoln_random", "Analysis") #Assign column names
summarymodelfits_twoln_table <- data.frame(Analysis = numbers5[,3], twoln_fixed = numbers5[,1], twoln_random = numbers5[,2]) #Reorder columns
#summarymodelfits_twoln_table #See table
#Make list of tables
tables <- list(summaryfitstats_table1_trad, summaryeffsizes_table2_trad, summaryfitstats_table1_phyl, summaryeffsizes_table2_phyl, summarymodelfits_AIC_table, summarymodelfits_twoln_table)
tables
}
# 3 groups in the moderator variable
maketables_3group <- function(output){ #DONE IF/ELSE REFORMATTING
#Table 1, Traditional meta-analysis
#a1<-unlist(strsplit(output[6]," +"))[-1] #Table header
a2<-unlist(strsplit(output[8]," +"))[-1] #Data row 1, between groups, fixed effects
a3<-unlist(strsplit(output[9]," +"))[-1] #Data row 2, within groups
a4 <- if(length(unlist(strsplit(output[10]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[10]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[10]," +"))[-1][1],unlist(strsplit(output[10]," +"))[-1][2],unlist(strsplit(output[10]," +"))[-1][3]),unlist(strsplit(output[10]," +"))[-1][4:8])} else
end
a5 <- if(length(unlist(strsplit(output[11]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[11]," +"))[-1][1],unlist(strsplit(output[11]," +"))[-1][2],unlist(strsplit(output[11]," +"))[-1][3]),unlist(strsplit(output[11]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[11]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[11]," +"))[-1][1],unlist(strsplit(output[11]," +"))[-1][2],unlist(strsplit(output[11]," +"))[-1][3]),unlist(strsplit(output[11]," +"))[-1][4:8])} else
end
a6 <- if(length(unlist(strsplit(output[12]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[12]," +"))[-1][1],unlist(strsplit(output[12]," +"))[-1][2],unlist(strsplit(output[12]," +"))[-1][3]),unlist(strsplit(output[12]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[12]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[12]," +"))[-1][1],unlist(strsplit(output[12]," +"))[-1][2],unlist(strsplit(output[12]," +"))[-1][3]),unlist(strsplit(output[12]," +"))[-1][4:8])} else
end
a7<-unlist(strsplit(output[13]," +"))[-1] #Data row 6, Total
a8<-unlist(strsplit(output[16]," +"))[-1] #Data row 7, between groups, random effects
rownaames<-c(a2[1],a3[1],a4[1],a5[1],a6[1],a7[1],a8[1]) #make vector of first column names
numbers<-t(data.frame(as.numeric(a2[2:4]),as.numeric(a3[2:4]),as.numeric(a4[2:4]),as.numeric(a5[2:4]),as.numeric(a6[2:4]),append(as.numeric(a7[2]),c(-9999,-9999),after=1),as.numeric(a8[2:4]))) #make data frame of numbers
numbers<-data.frame(numbers,rownaames) #Make data frame of row names and numbers
colnames(numbers)<-c("Q","df","P","Source") #Assign column names
summaryfitstats_table1_trad<-data.frame(Source=numbers$Source,Q=numbers$Q,df=numbers$df,P=numbers$P) #Reorder columns
#summaryfitstats_table1_trad #See table
#Table 2, Traditional meta-analysis
#b1<-unlist(strsplit(output[25]," +"))[-1] #Table header
b2 <- if(length(unlist(strsplit(output[29]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[29]," +"))[-1])[1],(unlist(strsplit(output[29]," +"))[-1])[2]),unlist(strsplit(output[29]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[29]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[29]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[29]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[29]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[29]," +"))[-1][6:7])} else
end
b3 <- if(length(unlist(strsplit(output[30]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[30]," +"))[-1])[1],(unlist(strsplit(output[30]," +"))[-1])[2]),unlist(strsplit(output[30]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[30]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[30]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[30]," +"))[-1][6:7])} else
end
b4 <- if(length(unlist(strsplit(output[31]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[31]," +"))[-1])[1],(unlist(strsplit(output[31]," +"))[-1])[2]),unlist(strsplit(output[31]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[31]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[31]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[31]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[31]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[31]," +"))[-1][6:7])} else
end
b5 <- if(length(unlist(strsplit(output[32]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[32]," +"))[-1])[1],(unlist(strsplit(output[32]," +"))[-1])[2]),unlist(strsplit(output[32]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[32]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[32]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[32]," +"))[-1][6:7])} else
end
b6 <- if(length(unlist(strsplit(output[36]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[36]," +"))[-1])[1],(unlist(strsplit(output[36]," +"))[-1])[2]),unlist(strsplit(output[36]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[36]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[36]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[36]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[36]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[36]," +"))[-1][6:7])} else
end
b7 <- if(length(unlist(strsplit(output[37]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[37]," +"))[-1])[1],(unlist(strsplit(output[37]," +"))[-1])[2]),unlist(strsplit(output[37]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[37]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[37]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[37]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[37]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[37]," +"))[-1][6:7])} else
end
b8 <- if(length(unlist(strsplit(output[38]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[38]," +"))[-1])[1],(unlist(strsplit(output[38]," +"))[-1])[2]),unlist(strsplit(output[38]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[38]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[38]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[38]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[38]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[38]," +"))[-1][6:7])} else
end
b9 <- if(length(unlist(strsplit(output[39]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[39]," +"))[-1])[1],(unlist(strsplit(output[39]," +"))[-1])[2]),unlist(strsplit(output[39]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[39]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[39]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[39]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[39]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[39]," +"))[-1][6:7])} else
end
rownaames1<-c(b2[1],b3[1],b4[1],b5[1],b6[1],b7[1],b8[1],b9[1]) #make vector of first column names
b22<-CI_split(b2) #Split confidence interval numbers apart and reinsert to vector of data
b33<-CI_split(b3)
b44<-CI_split(b4)
b55<-CI_split(b5)
b66<-CI_split(b6)
b77<-CI_split(b7)
b88<-CI_split(b8)
b99<-CI_split(b9)
numbers1<-t(data.frame(as.numeric(b22),as.numeric(b33),as.numeric(b44),as.numeric(b55),as.numeric(b66),as.numeric(b77),as.numeric(b88),as.numeric(b99))) #make data frame of numbers
numbers1<-data.frame(numbers1,rownaames1) #Make data frame of row names and numbers
colnames(numbers1)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_trad<-data.frame(Group=numbers1[,9],k=numbers1[,1],effsize=numbers1[,2],var=numbers1[,3],CI_low=numbers1[,4],CI_high=numbers1[,5],Z=numbers1[,6],df=numbers1[,7],P=numbers1[,8]) #Reorder columns
#summaryeffsizes_table2_trad #See table
###Phylogenetic meta-analysis
#Table 1, Phylogenetic meta-analysis
#d1<-unlist(strsplit(output[53]," +"))[-1] #Table header
d2 <- unlist(strsplit(output[55]," +"))[-1] #Data row 1, between groups, fixed effects
d3 <- unlist(strsplit(output[56]," +"))[-1] #Data row 2, within groups
d4 <- if(length(unlist(strsplit(output[57]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[57]," +"))[-1][1],unlist(strsplit(output[57]," +"))[-1][2],unlist(strsplit(output[57]," +"))[-1][3]),unlist(strsplit(output[57]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[57]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[57]," +"))[-1][1],unlist(strsplit(output[57]," +"))[-1][2],unlist(strsplit(output[57]," +"))[-1][3]),unlist(strsplit(output[57]," +"))[-1][4:8])} else
end
d5 <- if(length(unlist(strsplit(output[58]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[58]," +"))[-1][1],unlist(strsplit(output[58]," +"))[-1][2],unlist(strsplit(output[58]," +"))[-1][3]),unlist(strsplit(output[58]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[58]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[58]," +"))[-1][1],unlist(strsplit(output[58]," +"))[-1][2],unlist(strsplit(output[58]," +"))[-1][3]),unlist(strsplit(output[58]," +"))[-1][4:8])} else
end
d6 <- if(length(unlist(strsplit(output[59]," +"))[-1]) == 6) {c(paste(unlist(strsplit(output[59]," +"))[-1][1],unlist(strsplit(output[59]," +"))[-1][2],unlist(strsplit(output[59]," +"))[-1][3]),unlist(strsplit(output[59]," +"))[-1][4:6])} else
if(length(unlist(strsplit(output[59]," +"))[-1]) == 8) {c(paste(unlist(strsplit(output[59]," +"))[-1][1],unlist(strsplit(output[59]," +"))[-1][2],unlist(strsplit(output[59]," +"))[-1][3]),unlist(strsplit(output[59]," +"))[-1][4:8])} else
end
d7 <- unlist(strsplit(output[60]," +"))[-1] #Data row 6, Total
d8 <- unlist(strsplit(output[63]," +"))[-1] #Data row 7, between groups, random effects
rownaames2<-c(d2[1],d3[1],d4[1],d5[1],d6[1],d7[1],d8[1]) #make vector of first column names
numbers2<-t(data.frame(append(as.numeric(d2[2:4]),c(-9999,-9999),after=3),as.numeric(d3[2:6]),as.numeric(d4[2:6]),as.numeric(d5[2:6]),as.numeric(d6[2:6]),append(as.numeric(d7[2]),c(-9999,-9999,-9999,-9999),after=1),append(as.numeric(d8[2:4]),c(-9999,-9999),after=3))) #make data frame of numbers
numbers2<-data.frame(numbers2,rownaames2) #Make data frame of row names and numbers
colnames(numbers2)<-c("Q","df","P","df_polytadj","P_polytadj","Source") #Assign column names
summaryfitstats_table1_phyl<-data.frame(Source=numbers2[,6],Q=numbers2[,1],df=numbers2[,2],P=numbers2[,3],df_polytadj=numbers2[,4],P_polytadj=numbers2[,5]) #Reorder columns
#summaryfitstats_table1_phyl #See table
#Table 2, Phylogenetic meta-analysis
#e1<-unlist(strsplit(output[72]," +"))[-1] #Table header
e2 <- if(length(unlist(strsplit(output[76]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[76]," +"))[-1])[1],(unlist(strsplit(output[76]," +"))[-1])[2]),unlist(strsplit(output[76]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[76]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[76]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[76]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[76]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[76]," +"))[-1][6:7])} else
end
e3 <- if(length(unlist(strsplit(output[77]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[77]," +"))[-1])[1],(unlist(strsplit(output[77]," +"))[-1])[2]),unlist(strsplit(output[77]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[77]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[77]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[77]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[77]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[77]," +"))[-1][6:7])} else
end
e4 <- if(length(unlist(strsplit(output[78]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[78]," +"))[-1])[1],(unlist(strsplit(output[78]," +"))[-1])[2]),unlist(strsplit(output[78]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[78]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[78]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[78]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[78]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[78]," +"))[-1][6:7])} else
end
e5 <- if(length(unlist(strsplit(output[79]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[79]," +"))[-1])[1],(unlist(strsplit(output[79]," +"))[-1])[2]),unlist(strsplit(output[79]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[79]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[79]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[79]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[79]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[79]," +"))[-1][6:7])} else
end
e6 <- if(length(unlist(strsplit(output[83]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[83]," +"))[-1])[1],(unlist(strsplit(output[83]," +"))[-1])[2]),unlist(strsplit(output[83]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[83]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[83]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[83]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[83]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[83]," +"))[-1][6:7])} else
end
e7 <- if(length(unlist(strsplit(output[84]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[84]," +"))[-1])[1],(unlist(strsplit(output[84]," +"))[-1])[2]),unlist(strsplit(output[84]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[84]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[84]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[84]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[84]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[84]," +"))[-1][6:7])} else
end
e8 <- if(length(unlist(strsplit(output[85]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[85]," +"))[-1])[1],(unlist(strsplit(output[85]," +"))[-1])[2]),unlist(strsplit(output[85]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[85]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[85]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[85]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[85]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[85]," +"))[-1][6:7])} else
end
e9 <- if(length(unlist(strsplit(output[86]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[86]," +"))[-1])[1],(unlist(strsplit(output[86]," +"))[-1])[2]),unlist(strsplit(output[86]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[86]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[86]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[86]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[86]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[86]," +"))[-1][6:7])} else
end
rownaames3<-c(e2[1],e3[1],e4[1],e5[1],e6[1],e7[1],e8[1],e9[1]) #make vector of first column names
e22<-CI_split(e2) #Split confidence interval numbers apart and reinsert to vector of data
e33<-CI_split(e3)
e44<-CI_split(e4)
e55<-CI_split(e5)
e66<-CI_split(e6)
e77<-CI_split(e7)
e88<-CI_split(e8)
e99<-CI_split(e9)
numbers3<-t(data.frame(as.numeric(e22),as.numeric(e33),as.numeric(e44),as.numeric(e55),as.numeric(e66),as.numeric(e77),as.numeric(e88),as.numeric(e99))) #make data frame of numbers
numbers3<-data.frame(numbers3,rownaames3) #Make data frame of row names and numbers
colnames(numbers3)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_phyl<-data.frame(Group=numbers3[,9],k=numbers3[,1],effsize=numbers3[,2],var=numbers3[,3],CI_low=numbers3[,4],CI_high=numbers3[,5],Z=numbers3[,6],df=numbers3[,7],P=numbers3[,8]) #Reorder columns
#summaryeffsizes_table2_phyl #See table
### Model fit table, AIC
f1<-unlist(strsplit(output[101]," +"))[-1] #Table header
f2<-unlist(strsplit(output[102]," +"))[-1] #Data row 1, fixed effects, all studies
rownaames4<-c(f1[1],f2[1]) #make vector of first column names
numbers4<-t(data.frame(as.numeric(f1[2:3]),as.numeric(f2[2:3]))) #make data frame of numbers
numbers4<-data.frame(numbers4,rownaames4) #Make data frame of row names and numbers
colnames(numbers4)<-c("AIC_fixed","AIC_random","Analysis") #Assign column names
summarymodelfits_AIC_table<-data.frame(Analysis=numbers4[,3],AIC_fixed=numbers4[,1],AIC_random=numbers4[,2]) #Reorder columns
#summarymodelfits_AIC_table #See table
### Model fit table, -2(likelihood)
g1 <- unlist(strsplit(output[113]," +"))[-1] #Traditional AIC data
g2 <- unlist(strsplit(output[114]," +"))[-1] #Phylogenetically controlled AIC data
rownaames5 <- c(g1[1],g2[1]) #make vector of first column names
numbers5 <- t(data.frame(as.numeric(g1[2:3]), as.numeric(g2[2:3]))) #make data frame of numbers
numbers5 <- data.frame(numbers5, rownaames5) #Make data frame of row names and numbers
colnames(numbers5) <- c("twoln_fixed", "twoln_random", "Analysis") #Assign column names
summarymodelfits_twoln_table <- data.frame(Analysis = numbers5[,3], twoln_fixed = numbers5[,1], twoln_random = numbers5[,2]) #Reorder columns
#summarymodelfits_twoln_table #See table
#Make list of tables
tables <- list(summaryfitstats_table1_trad, summaryeffsizes_table2_trad, summaryfitstats_table1_phyl, summaryeffsizes_table2_phyl, summarymodelfits_AIC_table, summarymodelfits_twoln_table)
tables
}
# 4 groups in the moderator variable
maketables_4group <- function(output){ #DONE IF/ELSE REFORMATTING
#Table 1, Traditional meta-analysis
a1<-unlist(strsplit(output[6]," +"))[-1] #Table header
a2<-unlist(strsplit(output[8]," +"))[-1] #Data row 1, between groups, fixed effects
a3<-unlist(strsplit(output[9]," +"))[-1] #Data row 2, within groups
a4<-unlist(strsplit(output[10]," +"))[-1] #Data row 3, within goup 0
a5<-unlist(strsplit(output[11]," +"))[-1] #Data row 4, within goup 1
a6<-unlist(strsplit(output[12]," +"))[-1] #Data row 5, within goup 2
a7<-unlist(strsplit(output[13]," +"))[-1] #Data row 6, Total
a8<-unlist(strsplit(output[14]," +"))[-1] #Data row 7, between groups, random effects
a9<-unlist(strsplit(output[17]," +"))[-1] #Data row 7, between groups, random effects
rownaames<-c(a2[1],a3[1],a4[1],a5[1],a6[1],a7[1],a8[1],a9[1]) #make vector of first column names
numbers<-t(data.frame(as.numeric(a2[2:4]),as.numeric(a3[2:4]),as.numeric(a4[2:4]),as.numeric(a5[2:4]),as.numeric(a6[2:4]),as.numeric(a7[2:4]),append(as.numeric(a8[2]),c(-9999,-9999),after=1),as.numeric(a9[2:4]))) #make data frame of numbers
numbers<-data.frame(numbers,rownaames) #Make data frame of row names and numbers
colnames(numbers)<-c("Q","df","P","Source") #Assign column names
summaryfitstats_table1_trad<-data.frame(Source=numbers$Source,Q=numbers$Q,df=numbers$df,P=numbers$P) #Reorder columns
#summaryfitstats_table1_trad #See table
#Table 2, Traditional meta-analysis
b1<-unlist(strsplit(output[26]," +"))[-1] #Table header
b2 <- if(length(unlist(strsplit(output[30]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[30]," +"))[-1])[1],(unlist(strsplit(output[30]," +"))[-1])[2]),unlist(strsplit(output[30]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[30]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[30]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[30]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[30]," +"))[-1][6:7])} else
end
b3 <- if(length(unlist(strsplit(output[31]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[31]," +"))[-1])[1],(unlist(strsplit(output[31]," +"))[-1])[2]),unlist(strsplit(output[31]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[31]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[31]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[31]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[31]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[31]," +"))[-1][6:7])} else
end
b4 <- if(length(unlist(strsplit(output[32]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[32]," +"))[-1])[1],(unlist(strsplit(output[32]," +"))[-1])[2]),unlist(strsplit(output[32]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[32]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[32]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[32]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[32]," +"))[-1][6:7])} else
end
b5 <- if(length(unlist(strsplit(output[33]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[33]," +"))[-1])[1],(unlist(strsplit(output[33]," +"))[-1])[2]),unlist(strsplit(output[33]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[33]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[33]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[33]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[33]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[33]," +"))[-1][6:7])} else
end
b6 <- if(length(unlist(strsplit(output[34]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[34]," +"))[-1])[1],(unlist(strsplit(output[34]," +"))[-1])[2]),unlist(strsplit(output[34]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[34]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[34]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[34]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[34]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[34]," +"))[-1][6:7])} else
end
b7 <- if(length(unlist(strsplit(output[38]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[38]," +"))[-1])[1],(unlist(strsplit(output[38]," +"))[-1])[2]),unlist(strsplit(output[38]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[38]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[38]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[38]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[38]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[38]," +"))[-1][6:7])} else
end
b8 <- if(length(unlist(strsplit(output[39]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[39]," +"))[-1])[1],(unlist(strsplit(output[39]," +"))[-1])[2]),unlist(strsplit(output[39]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[39]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[39]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[39]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[39]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[39]," +"))[-1][6:7])} else
end
b9 <- if(length(unlist(strsplit(output[40]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[40]," +"))[-1])[1],(unlist(strsplit(output[40]," +"))[-1])[2]),unlist(strsplit(output[40]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[40]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[40]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[40]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[40]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[40]," +"))[-1][6:7])} else
end
b10 <- if(length(unlist(strsplit(output[41]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[41]," +"))[-1])[1],(unlist(strsplit(output[41]," +"))[-1])[2]),unlist(strsplit(output[41]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[41]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[41]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[41]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[41]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[41]," +"))[-1][6:7])} else
end
b11 <- if(length(unlist(strsplit(output[42]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[42]," +"))[-1])[1],(unlist(strsplit(output[42]," +"))[-1])[2]),unlist(strsplit(output[42]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[42]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[42]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[42]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[42]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[42]," +"))[-1][6:7])} else
end
rownaames1<-c(b2[1],b3[1],b4[1],b5[1],b6[1],b7[1],b8[1],b9[1],b10[1],b11[1]) #make vector of first column names
b22<-CI_split(b2) #Split confidence interval numbers apart and reinsert to vector of data
b33<-CI_split(b3)
b44<-CI_split(b4)
b55<-CI_split(b5)
b66<-CI_split(b6)
b77<-CI_split(b7)
b88<-CI_split(b8)
b99<-CI_split(b9)
b1010<-CI_split(b10)
b1111<-CI_split(b11)
numbers1<-t(data.frame(as.numeric(b22),as.numeric(b33),as.numeric(b44),as.numeric(b55),as.numeric(b66),as.numeric(b77),as.numeric(b88),as.numeric(b99),as.numeric(b1010),as.numeric(b1111))) #make data frame of numbers
numbers1<-data.frame(numbers1,rownaames1) #Make data frame of row names and numbers
colnames(numbers1)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_trad<-data.frame(Group=numbers1[,9],k=numbers1[,1],effsize=numbers1[,2],var=numbers1[,3],CI_low=numbers1[,4],CI_high=numbers1[,5],Z=numbers1[,6],df=numbers1[,7],P=numbers1[,8]) #Reorder columns
#summaryeffsizes_table2_trad #See table
###Phylogenetic meta-analysis
#Table 1, Phylogenetic meta-analysis
d1<-unlist(strsplit(output[56]," +"))[-1] #Table header
d2<-unlist(strsplit(output[58]," +"))[-1] #Data row 1, between groups, fixed effects
d3<-unlist(strsplit(output[59]," +"))[-1] #Data row 2, within groups
d4<-unlist(strsplit(output[60]," +"))[-1] #Data row 3, within goup 0
d5<-unlist(strsplit(output[61]," +"))[-1] #Data row 4, within goup 1
d6<-unlist(strsplit(output[62]," +"))[-1] #Data row 5, within goup 2
d7<-unlist(strsplit(output[63]," +"))[-1] #Data row 6, Total
d8<-unlist(strsplit(output[64]," +"))[-1] #Data row 7, between groups, random effects
d9<-unlist(strsplit(output[67]," +"))[-1] #Data row 7, between groups, random effects
rownaames2<-c(d2[1],d3[1],d4[1],d5[1],d6[1],d7[1],d8[1],d9[1]) #make vector of first column names
numbers2<-t(data.frame(append(as.numeric(d2[2:4]),c(-9999,-9999),after=3),as.numeric(d3[2:6]),as.numeric(d4[2:6]),as.numeric(d5[2:6]),as.numeric(d6[2:6]),as.numeric(d7[2:6]),append(as.numeric(d8[2]),c(-9999,-9999,-9999,-9999),after=3),append(as.numeric(d9[2:4]),c(-9999,-9999),after=3))) #make data frame of numbers
numbers2<-data.frame(numbers2,rownaames2) #Make data frame of row names and numbers
colnames(numbers2)<-c("Q","df","P","df_polytadj","P_polytadj","Source") #Assign column names
summaryfitstats_table1_phyl<-data.frame(Source=numbers2[,6],Q=numbers2[,1],df=numbers2[,2],P=numbers2[,3],df_polytadj=numbers2[,4],P_polytadj=numbers2[,5]) #Reorder columns
#summaryfitstats_table1_phyl #See table
#Table 2, Phylogenetic meta-analysis
e1 <- unlist(strsplit(output[76]," +"))[-1] #Table header
e2 <- if(length(unlist(strsplit(output[80]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[80]," +"))[-1])[1],(unlist(strsplit(output[80]," +"))[-1])[2]),unlist(strsplit(output[80]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[80]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[80]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[80]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[80]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[80]," +"))[-1][6:7])} else
end
e3 <- if(length(unlist(strsplit(output[81]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[81]," +"))[-1])[1],(unlist(strsplit(output[81]," +"))[-1])[2]),unlist(strsplit(output[81]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[81]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[81]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[81]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[81]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[81]," +"))[-1][6:7])} else
end
e4 <- if(length(unlist(strsplit(output[82]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[82]," +"))[-1])[1],(unlist(strsplit(output[82]," +"))[-1])[2]),unlist(strsplit(output[82]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[82]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[82]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[82]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[82]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[82]," +"))[-1][6:7])} else
end
e5 <- if(length(unlist(strsplit(output[83]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[83]," +"))[-1])[1],(unlist(strsplit(output[83]," +"))[-1])[2]),unlist(strsplit(output[83]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[83]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[83]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[83]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[83]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[83]," +"))[-1][6:7])} else
end
e6 <- if(length(unlist(strsplit(output[84]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[84]," +"))[-1])[1],(unlist(strsplit(output[84]," +"))[-1])[2]),unlist(strsplit(output[84]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[84]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[84]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[84]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[84]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[84]," +"))[-1][6:7])} else
end
e7 <- if(length(unlist(strsplit(output[88]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[88]," +"))[-1])[1],(unlist(strsplit(output[88]," +"))[-1])[2]),unlist(strsplit(output[88]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[88]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[88]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[88]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[88]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[88]," +"))[-1][6:7])} else
end
e8 <- if(length(unlist(strsplit(output[89]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[89]," +"))[-1])[1],(unlist(strsplit(output[89]," +"))[-1])[2]),unlist(strsplit(output[89]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[89]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[89]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[89]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[89]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[89]," +"))[-1][6:7])} else
end
e9 <- if(length(unlist(strsplit(output[90]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[90]," +"))[-1])[1],(unlist(strsplit(output[90]," +"))[-1])[2]),unlist(strsplit(output[90]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[90]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[90]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[90]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[90]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[90]," +"))[-1][6:7])} else
end
e10 <- if(length(unlist(strsplit(output[91]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[91]," +"))[-1])[1],(unlist(strsplit(output[91]," +"))[-1])[2]),unlist(strsplit(output[91]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[91]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[91]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[91]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[91]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[91]," +"))[-1][6:7])} else
end
e11 <- if(length(unlist(strsplit(output[92]," +"))[-1]) == 9) {c(paste((unlist(strsplit(output[92]," +"))[-1])[1],(unlist(strsplit(output[92]," +"))[-1])[2]),unlist(strsplit(output[92]," +"))[-1][3:9])} else
if(str_detect(unlist(strsplit(output[92]," +"))[-1][6], ")") == "TRUE") {c(unlist(strsplit(output[92]," +"))[-1][1:4], paste(str_split(unlist(strsplit(output[92]," +"))[-1][6], ")")[[1]][1],")", sep=""), str_split(unlist(strsplit(output[92]," +"))[-1][6], ")")[[1]][2], unlist(strsplit(output[92]," +"))[-1][6:7])} else
end
rownaames3<-c(e2[1],e3[1],e4[1],e5[1],e6[1],e7[1],e8[1],e9[1],e10[1],e11[1]) #make vector of first column names
e22<-CI_split(e2) #Split confidence interval numbers apart and reinsert to vector of data
e33<-CI_split(e3)
e44<-CI_split(e4)
e55<-CI_split(e5)
e66<-CI_split(e6)
e77<-CI_split(e7)
e88<-CI_split(e8)
e99<-CI_split(e9)
e1010<-CI_split(e10)
e1111<-CI_split(e11)
numbers3<-t(data.frame(as.numeric(e22),as.numeric(e33),as.numeric(e44),as.numeric(e55),as.numeric(e66),as.numeric(e77),as.numeric(e88),as.numeric(e99),as.numeric(e1010),as.numeric(e1111))) #make data frame of numbers
numbers3<-data.frame(numbers3,rownaames3) #Make data frame of row names and numbers
colnames(numbers3)<-c("k","effsize","var","95CI_low","95CI_high","Z","df","P","Group") #Assign column names
summaryeffsizes_table2_phyl<-data.frame(Group=numbers3[,9],k=numbers3[,1],effsize=numbers3[,2],var=numbers3[,3],CI_low=numbers3[,4],CI_high=numbers3[,5],Z=numbers3[,6],df=numbers3[,7],P=numbers3[,8]) #Reorder columns
#summaryeffsizes_table2_phyl #See table
###Model fit table
f1<-unlist(strsplit(output[107]," +"))[-1] #Table header
f2<-unlist(strsplit(output[108]," +"))[-1] #Data row 1, fixed effects, all studies
rownaames4<-c(f1[1],f2[1]) #make vector of first column names
numbers4<-t(data.frame(as.numeric(f1[2:3]),as.numeric(f2[2:3]))) #make data frame of numbers
numbers4<-data.frame(numbers4,rownaames4) #Make data frame of row names and numbers
colnames(numbers4)<-c("AIC_fixed","AIC_random","Analysis") #Assign column names
summarymodelfits_AIC_table<-data.frame(Analysis=numbers4[,3],AIC_fixed=numbers4[,1],AIC_random=numbers4[,2]) #Reorder columns
#summarymodelfits_table #See table
### Model fit table, -2(likelihood)
g1 <- unlist(strsplit(output[119]," +"))[-1] #Traditional AIC data
g2 <- unlist(strsplit(output[120]," +"))[-1] #Phylogenetically controlled AIC data
rownaames5 <- c(g1[1],g2[1]) #make vector of first column names
numbers5 <- t(data.frame(as.numeric(g1[2:3]), as.numeric(g2[2:3]))) #make data frame of numbers
numbers5 <- data.frame(numbers5, rownaames5) #Make data frame of row names and numbers
colnames(numbers5) <- c("twoln_fixed", "twoln_random", "Analysis") #Assign column names
summarymodelfits_twoln_table <- data.frame(Analysis = numbers5[,3], twoln_fixed = numbers5[,1], twoln_random = numbers5[,2]) #Reorder columns
#summarymodelfits_twoln_table #See table
#Make list of tables
tables <- list(summaryfitstats_table1_trad, summaryeffsizes_table2_trad, summaryfitstats_table1_phyl, summaryeffsizes_table2_phyl, summarymodelfits_AIC_table, summarymodelfits_twoln_table)
tables
}
# Process phylometa output
phylometa.process <- function(phylometa.run,groups){
aaa <- data.frame(phylometa.run)
if(length(subset(aaa,aaa[,1]=="poly")[,1]) == 1) {phylometa.run <- phylometa.run[-as.numeric(rownames(subset(aaa,aaa[,1]=="poly")))]} else (phylometa.run <- phylometa.run)
aaaa <- data.frame(phylometa.run)
x <- as.numeric(rownames(subset(aaaa,aaaa[,1]=="RESULTS SECTION A. Traditional meta-analysis.")))
y <- as.numeric(rownames(subset(aaaa,aaaa[,1]=="For further details on these methods see:")))
bbb <- phylometa.run[x:y]
outputtables <- (
if (groups == 1) {maketables_1group(bbb)} else
if (groups == 2) {maketables_2group(bbb)} else
if (groups == 3) {maketables_3group(bbb)} else
if (groups == 4) {maketables_4group(bbb)} else
end)
outputtables
}
# Get output from phylometa.process
phylometa.output <- function(a) {
a
}
phylometa.output.table <- function(a,b) {
a[[b]]
}
# Function used below in the function 'arrange'
vp.layout <- function(x, y) viewport(layout.pos.row=x, layout.pos.col=y)
arrange_ <- function(..., nrow=NULL, ncol=NULL, as.table=FALSE) {
dots <- list(...)
n <- length(dots)
if(is.null(nrow) & is.null(ncol)) { nrow = floor(n/2) ; ncol = ceiling(n/nrow)}
if(is.null(nrow)) { nrow = ceiling(n/ncol)}
if(is.null(ncol)) { ncol = ceiling(n/nrow)}
## NOTE see n2mfrow in grDevices for possible alternative
grid.newpage()
pushViewport(viewport(layout=grid.layout(nrow,ncol) ) )
ii.p <- 1
for(ii.row in seq(1, nrow)){
ii.table.row <- ii.row
if(as.table) {ii.table.row <- nrow - ii.table.row + 1}
for(ii.col in seq(1, ncol)){
ii.table <- ii.p
if(ii.p > n) break
print(dots[[ii.table]], vp=vp.layout(ii.table.row, ii.col))
ii.p <- ii.p + 1
}
}
}
# Function to write trees to directory for use by Phylometa
WriteTrees2 <- function(xxx) {
for(i in 1:length(xxx)){
write.tree(xxx[[i]], paste("tree",i,".txt","",sep=""))
}
}
Created by Pretty R at inside-R.org
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment