2018-09-12

# GEO数据下载，数据处理，基因与ID的转换

```

library(GEOquery) #没下载包的自行下载

##本地已经下载好了的数据

get = getGEO(filename = "./data/GSE13535_series_matrix.txt/GSE13535_series_matrix.txt")

GPL=getGEO(filename = './data/GPL/GPL1355.soft')

exprset = exprs(get)

```

![转换前表达矩阵.png](https://upload-images.jianshu.io/upload_images/13573980-5aab3e42e7627c36.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

```

exgpl=Table(GPL)

#exgpl2 = pData(get)

exprset = log2(exprset+1) # 处理数据

ids = exgpl[,c("ID","Gene Symbol")]

colnames(ids) = c("probe_id" ,"symbol") # 给ids列重新命名

save(ids,exprset,exgpl2,file = 'GSE13535input.Rdata')

length(unique(ids$symbol))

tail(sort(table(ids$symbol)))

table(sort(table(ids$symbol)))

#####开始转换数据了

table(rownames(exprset) %in% ids$probe_id)

dim(exprset)

expreSet=exprset[rownames(exprset) %in% ids$probe_id,]

dim(exprset)

dim(ids)

ids=ids[match(rownames(exprset),ids$probe_id),]

dim(ids)

head(ids)

expreSet[1:5,1:5]

tmp = by(exprset,ids$symbol,

function(x) rownames(x)[which.max(rowMeans(x))])

probes = as.character(tmp)

dim(exprset)

exprset=exprset[rownames(exprset) %in% probes ,] # 过滤

dim(exprset)#15249 22

rownames(exprset)=ids[match(rownames(exprset),ids$probe_id),2]

exprset[1:5,1:5]

```

![转换后表达矩阵.png](https://upload-images.jianshu.io/upload_images/13573980-74126181898008c1.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

####参考链接1（http://www.bio-info-trainee.com/1399.html）

####参考链接2（https://github.com/jmzeng1314/5years/blob/master/learn-R/tasks/3-r-20-codes.R）