GEO样本名称处理和对照夹杂影响后续分析吗，如何获取样本名称

rm(list = ls())
#打破下载时间的限制,改前60秒，改后10w秒
options(timeout = 100000) 
options(scipen = 20)#不要以科学计数法表示

#GEO的样本名称，处理和对照夹杂，如何获取样本名称？
#读入数据
data <- data.table::fread("gsm_group.txt",data.table = F,header = F)
names(data) <- c("GSM","group")
data[1,2]
#> [1] "whole_blood_control_1"
data[102:106,2]
#> [1] "whole_blood_control_34" "whole_blood_case_69"   
#> [3] "whole_blood_control_35" "whole_blood_case_70"   
#> [5] "whole_blood_case_71"

library(dplyr)
library(tidyr)
metadata <- data %>% 
  separate(group,into = c("drop","group"),sep = "_blood_") %>% 
  select(-drop) %>% 
  separate(group,into = c("group","drop"),sep = "_") %>% 
  select(-drop)
metadata[1,2]
#> [1] "control"
metadata[102:106,2]
#> [1] "control" "case"    "control" "case"    "case"

#只是去除了冗余信息，其实样本顺序没变，处理和对照依旧夹杂

library(stringr)
group <- str_match(data$group,"whole_blood_(.*)_.*")
metadata2 <- data.frame(GSM=data$GSM,group=group[,2])
metadata2[102:106,2]
#> [1] "control" "case"    "control" "case"    "case"

#同样只是去除了冗余信息，其实样本顺序没变，处理和对照依旧夹杂