In Sage-Bionetworks/AMPAD:

ROSMAP data present

synapser::synLogin()
foo <- synapser::synTableQuery("select * from syn18912660")$asDataFrame()
foo <- dplyr::filter(foo,study=='ROSMAP')
bar <- dplyr::select(foo,individualID,dataType,assay,futureData)

fxn1 <- function(x,y){
  if(x=='metabolomics'){
    return(x)
  }else{
    return(y)
  }
}


bar$dataDescriptor <- mapply(fxn1,bar$dataType,bar$assay)
bar$exists <- bar$futureData
bar$exists[bar$exists==TRUE] <- 2
bar$exists[is.na(bar$exists)] <- 1
bar <- dplyr::select(bar,individualID,dataDescriptor,exists)
uniqueId <- paste0(bar$individualID,bar$dataDescriptor)
dup <- duplicated(uniqueId)
dupInd <- uniqueId[dup]
keepIndex <- uniqueId%in%dupInd

bar$exists[keepIndex] <- 3
bar <- bar[!dup,]

#bar <- bar[!dup,]
baz <- tidyr::spread(bar,key=individualID,value=exists)
baz[is.na(baz)] <- 0
rownames(baz) <- baz$dataDescriptor
baz <- baz[,-1]
baz2 <- t(apply(baz,1,as.numeric))

ab<-pheatmap::pheatmap(baz2)
baz3 <- baz2[ab$tree_row$order,ab$tree_col$order]
png(file='ROSMAP.png',width=2400,height=1600,pointsize=14)
pheatmap::pheatmap(baz3,color = rev(viridis::viridis(4)),cluster_rows = F,cluster_cols =F,legend_breaks=0:3,legend_labels=c('Absent\n\n\n\n','Available\n\n','Planned','\n\nAvailable\n+Planned'))
dev.off()

synapser::synLogin()
foo <- synapser::synTableQuery("select * from syn18912660")$asDataFrame()
foo <- dplyr::filter(foo,study=='MayoRNAseq')
bar <- dplyr::select(foo,individualID,dataType,assay,futureData)

bar$dataDescriptor <- bar$assay
bar$exists <- bar$futureData
bar$exists[bar$exists==TRUE] <- 2
bar$exists[is.na(bar$exists)] <- 1
bar <- dplyr::select(bar,individualID,dataDescriptor,exists)
uniqueId <- paste0(bar$individualID,bar$dataDescriptor)
dup <- duplicated(uniqueId)
dupInd <- uniqueId[dup]
keepIndex <- uniqueId%in%dupInd

bar$exists[keepIndex] <- 3
bar <- bar[!dup,]

#bar <- bar[!dup,]
baz <- tidyr::spread(bar,key=individualID,value=exists)
baz[is.na(baz)] <- 0
rownames(baz) <- baz$dataDescriptor
baz <- baz[,-1]
baz2 <- t(apply(baz,1,as.numeric))

ab<-pheatmap::pheatmap(baz2)
baz3 <- baz2[ab$tree_row$order,ab$tree_col$order]
png(file='ROSMAP.png',width=2400,height=1600,pointsize=14)
pheatmap::pheatmap(baz3,color = rev(viridis::viridis(4)),cluster_rows = F,cluster_cols =F,legend_breaks=0:3,legend_labels=c('Absent\n\n\n\n','Available\n\n','Planned','\n\nAvailable\n+Planned'))
dev.off()

synapser::synLogin()
foo <- synapser::synTableQuery("select * from syn18912660")$asDataFrame()
foo <- dplyr::filter(foo,study=='MSBB')
bar <- dplyr::select(foo,individualID,dataType,assay,futureData)


fxn1 <- function(x){
  if(is.na(x)){
    return('MSBBproteomics')
  }else{
    return(x)
  }
}


bar$dataDescriptor <- sapply(bar$assay,fxn1)
#bar$dataDescriptor <- bar$assay
bar$exists <- bar$futureData
bar$exists[bar$exists==TRUE] <- 2
bar$exists[is.na(bar$exists)] <- 1
bar <- dplyr::select(bar,individualID,dataDescriptor,exists)
uniqueId <- paste0(bar$individualID,bar$dataDescriptor)
dup <- duplicated(uniqueId)
dupInd <- uniqueId[dup]
keepIndex <- uniqueId%in%dupInd

bar$exists[keepIndex] <- 3
bar <- bar[!dup,]

#bar <- bar[!dup,]
baz <- tidyr::spread(bar,key=individualID,value=exists)
baz[is.na(baz)] <- 0
rownames(baz) <- baz$dataDescriptor
baz <- baz[,-1]
baz2 <- t(apply(baz,1,as.numeric))

ab<-pheatmap::pheatmap(baz2)
baz3 <- baz2[ab$tree_row$order,ab$tree_col$order]
png(file='ROSMAP.png',width=2400,height=1600,pointsize=14)
pheatmap::pheatmap(baz3,color = rev(viridis::viridis(4)),cluster_rows = F,cluster_cols =F,legend_breaks=0:3,legend_labels=c('Absent\n\n\n\n','Available\n\n','Planned','\n\nAvailable\n+Planned'))
dev.off()

Sage-Bionetworks/AMPAD documentation built on Jan. 13, 2020, 9:18 p.m.