library(dplyr)

data<-read.csv(file = "D:/heritability_collective/collective_motility.csv", header = TRUE)
data$cell_line<-as.character(data$cell_line)
data$family<-as.character(data$family)
data$generation<-as.character(data$generation)


unique(data$generation)

g<-unique(data$family)
h<-length(unique(data$generation))
test = array(data= NA, dim = c(0,(h+1)))
for ( i in g){
  print(i)
  l<-data[data$family == i,]
  m<-sapply(1:h, function(x) length(unique(l[l$generation == x,]$TID)))
  m = c(m,i)
  test = rbind(test, m)
}

test<-as.data.frame(test)

colnames(test)<-c("G1","G2","G3","G4","G5","G6","G7","Family")
test[]<-lapply(test, as.character)

useable_fam<-test[(test$G1 == 1) & (test$G2 == 2) & (test$G3 == 4) & (test$G4 >= 2),]
useable_fam1<-test[(test$G1 ==1) & (test$G2 ==2) & (test$G3 == 2) & (test$G4 >= 2),]
b<-useable_fam1$Family
c<-c(a,b)

data_2<-data %>% filter(data$family %in% c)

data_3<-data_2[data_2$generation <= 4,]

write.csv(data_3, file="D:/heritability_collective/Speed Heritability/Paper data/speed_gen_4.csv")