LinePlot

```{r keyprotein lineplot}

rm(list=ls())
library(ggplot2)
library(Rmisc)
library(stringr)
library(reshape2)
protein<-read.table("12_matrix_analyze/protein_overlap.txt",sep="\t",header = T)

Openswath <- read.table("06_protein_matrix/proteotypic/Openswath_part_ave_KMDR_proteotypic_matrix.txt",sep="\t",header = T,stringsAsFactors = F)
Openswath2<-data.frame(Openswath[which(Openswath$protein%in% protein$elements),])
row.names(Openswath2)<-Openswath2$protein
Openswath3<-data.frame("Openswath",t(Openswath2[,-1]))
Openswath4<-data.frame(str_split_fixed(row.names(Openswath3),"_",7)[,5],Openswath3)
names(Openswath4)<-c("group","software",names(Openswath4[,-c(1:2)]))


DIAnn <- read.table("06_protein_matrix/proteotypic/DIAnn_part_ave_KMDR_proteotypic_matrix.txt",sep="\t",header = T,stringsAsFactors = F)
DIAnn2<-data.frame(DIAnn[which(DIAnn$protein%in% protein$elements),])
row.names(DIAnn2)<-DIAnn2$protein
DIAnn3<-data.frame("DIAnn",t(DIAnn2[,-1]))
DIAnn4<-data.frame(str_split_fixed(row.names(DIAnn3),"_",7)[,5],DIAnn3)
names(DIAnn4)<-c("group","software",names(DIAnn4[,-c(1:2)]))

Spec <- read.table("06_protein_matrix/proteotypic/Spec_part_ave_KMDR_proteotypic_matrix.txt",sep="\t",header = T,stringsAsFactors = F)
Spec2<-data.frame(Spec[which(Spec$protein%in% protein$elements),])
row.names(Spec2)<-Spec2$protein
Spec3<-data.frame("Spec",t(Spec2[,-1]))
Spec4<-data.frame(str_split_fixed(row.names(Spec3),"_",7)[,5],Spec3)
names(Spec4)<-c("group","software",names(Spec4[,-c(1:2)]))

Encyo <- read.table("06_protein_matrix/proteotypic/Encyo_part_ave_KMDR_proteotypic_matrix.txt",sep="\t",header = T,stringsAsFactors = F)
Encyo2<-data.frame(Encyo[which(Encyo$protein%in% protein$elements),])
row.names(Encyo2)<-Encyo2$protein
Encyo3<-data.frame("Encyo",t(Encyo2[,-1]))
Encyo4<-data.frame(str_split_fixed(row.names(Encyo3),"_",7)[,5],Encyo3)
names(Encyo4)<-c("group","software",names(Encyo4[,-c(1:2)]))
df<-rbind(Openswath4,DIAnn4,Spec4,Encyo4)

df1<-df[-c(1,2)]

df2<-data.frame(apply(df1,2,function(X) sum(is.na(X)/104)))

names(df2)<-("NAratio")

df3<-data.frame(df1[,-(which(df2$NAratio>0.7))])

df4<-data.frame(df[,c(1,2)],df3)

dfIMA<-df[which(df$group%in%c("K","I1","I2","I3")),]
dfIMA$group<-gsub("I","",dfIMA$group)
dfIMA$group<-gsub("K","0",dfIMA$group)
dfIMA$group<-as.numeric(dfIMA$group)


myNormalize <- function (target) {
    (target - min(target,na.rm = T))/(max(target,na.rm = T) - min(target,na.rm = T))
}

dfIMA_normalize_Openswath<-data.frame(dfIMA[which(dfIMA$software=="Openswath"),1:2],apply(dfIMA[which(dfIMA$software=="Openswath"),3:ncol(dfIMA)],2,myNormalize))
dfIMA_normalize_DIAnn<-data.frame(dfIMA[which(dfIMA$software=="DIAnn"),1:2],apply(dfIMA[which(dfIMA$software=="DIAnn"),3:ncol(dfIMA)],2,myNormalize))
dfIMA_normalize_Spec<-data.frame(dfIMA[which(dfIMA$software=="Spec"),1:2],apply(dfIMA[which(dfIMA$software=="Spec"),3:ncol(dfIMA)],2,myNormalize))
dfIMA_normalize_Encyo<-data.frame(dfIMA[which(dfIMA$software=="Encyo"),1:2],apply(dfIMA[which(dfIMA$software=="Encyo"),3:ncol(dfIMA)],2,myNormalize))
dfIMA_normalize<-rbind(dfIMA_normalize_Openswath,dfIMA_normalize_DIAnn,dfIMA_normalize_Spec,dfIMA_normalize_Encyo)

for (i in 3:ncol(dfIMA_normalize)) {
tg <- dfIMA_normalize[,c(1,2,i)]
names(tg)<-c("IMA","software","protein")
tgc <- summarySE(tg, measurevar="protein", groupvars=c("software","IMA"))
anova<-data.frame("software","p")
names(anova)<-c("software","p")
anova0<-anova[-1,]


 for(a in c("DIAnn","Encyo", "Openswath","Spec")){
  
     tg1<- tg[which(tg$software== paste0(a)),]
        tg1[is.na(tg1)]<-0
      b <-(summary(aov(tg1$protein ~ tg1$IMA, tg1))[[1]])$`Pr(>F)`[1]
      c<-signif(b,2)
      anova2<-anova[-1,]
      anova2[1,1]<-paste0(a)
      anova2[1,2]<-c

      anova0<-rbind(anova0,anova2)
}


plot<-ggplot(tgc, aes(x=IMA, y=protein, colour=software))+   ggtitle(paste0(names(dfIMA_normalize)[i]))+
    geom_errorbar(aes(ymin=protein-se, ymax=protein+se), width=.1) +
    geom_line() +
    geom_point()+
  ylim(0,1)+
    scale_color_manual(values=c("#7E1B1E","#EB9C21","#6286BE","#B2D9AC"))+
    annotate("text", label = paste("p=",anova0[1,2],sep = ""), x = 0.5, y=max(tgc$protein,na.rm = T)*1, colour = "#7E1B1E")+
    annotate("text", label = paste("p=",anova0[2,2],sep = ""), x = 0.5, y=max(tgc$protein,na.rm = T)*0.9,colour = "#EB9C21")+
    annotate("text", label = paste("p=",anova0[3,2],sep = ""), x = 0.5, y=max(tgc$protein,na.rm = T)*0.8,colour = "#6286BE")+
    annotate("text", label = paste("p=",anova0[4,2],sep = ""), x = 0.5, y=max(tgc$protein,na.rm = T)*0.7,colour = "#B2D9AC")

p<-plot+ theme(legend.direction = 'horizontal',legend.position = 'top',legend.text = element_text(size = 15,color = "black"), legend.title = element_text(size=15,color="black") ,panel.grid.major =element_blank(), panel.grid.minor = element_blank(),panel.background = element_blank(),axis.line = element_line(colour = "black"), axis.ticks.length = unit(6, "pt"),axis.text = element_text(size=15,color="black")) 
ggsave(filename  = paste0("17_line_plot/IMA/IMA","_",names(dfIMA_normalize)[i],".pdf"),plot=p,device = NULL,width = 6,height = 6)
  
}

















```