使用FastUKB包read_file函数报错‘无法打开链接’的解决方法

问题背景 #

使用base包read.csv可正常读取本地文件Covariates1.csv，但加载FastUKB包后调用read_file读取同一文件时，出现错误：

INFO [2025-10-22 10:42:10] read file: Covariates1.csv
Error in file(file, ifelse(append, "a", "w")): The link cannot be opened.

已确认文件存在、工作目录正确，且FastUKB包内其他函数可正常调用。

相关验证代码及结果：

setwd("/home/longqc/test")
file_path <- "/home/longqc/test/Covariates1.csv"
read.csv(file_path)
# 执行结果：
# eid gender age p21003_i1 p21003_i2 p21003_i3     BMI p21001_i1 p21001_i2 p21001_i3
# 1  2311101   Male  57        NA        NA        NA 28.0600        NA        NA        NA
# 2  2068215   Male  66        NA        NA        NA 25.9556        NA        NA        NA
# 3  4904062 Female  50        NA        61        66 26.5047        NA   27.0950   25.8645
# 4  1736210   Male  47        NA        NA        NA 24.8584        NA        NA        NA
# 5  3911988 Female  48        NA        NA        NA 30.1402        NA        NA        NA

library(FastUKB)
file.exists("Covariates1.csv")
# [1] TRUE
ls("package:FastUKB")
# [1] "all_field_id"              "analyse_cox"               "analyse_intercox"         
# [4] "analyse_lm"                "analyse_osmr"              "analyse_subgroup"         
# [7] "category_data"             "data_2_cross_sectional"    "data_process"             
# [10] "data_process_auto"         "data_process_phewas"       "doc_baseline"             
# [13] "encoding_data"             "field_upload_sample"       "language_data"            
# [16] "main_field_data"           "plot_manhattan_phewas"     "plot_rcs"                 
# [19] "prepare_prs_gwas"          "protein_de"                "rap_acc_batch"            
# [22] "rap_acc_progress"          "rap_acc_summary"           "rap_bind"                 
# [25] "rap_extract_clinical_data" "rap_extract_snps"          "rap_file_download"        
# [28] "rap_file_upload"           "rap_job_status"            "rap_jobs"                 
# [31] "rap_run_prs"               "read_file"                 "register_rap"             
# [34] "register_ukb"              "ui_open"                   "view_data"                
# [37] "write_file"

raw_data = read_file("Covariates1.csv")

解决方案 #

1. 适配FastUKB的文件处理逻辑 #

FastUKB是为UK Biobank数据设计的工具包，read_file可能默认适配UKB特定格式，而非普通CSV。可以先用base包读取数据，再传入FastUKB的处理函数：

base_data <- read.csv("Covariates1.csv")
processed_data <- data_process(base_data)

2. 使用绝对路径读取 #

虽然相对路径验证有效，但部分包的函数对路径解析有特殊处理，尝试传入完整绝对路径：

raw_data = read_file("/home/longqc/test/Covariates1.csv")

3. 检查文件权限与编码 #

• 验证文件可读权限：执行file.access("Covariates1.csv", 4)，返回0表示权限正常，非0则需修改权限（如终端执行chmod 644 Covariates1.csv）。
• 指定编码尝试读取：

raw_data = read_file("Covariates1.csv", encoding = "UTF-8")

4. 显式指定文件参数 #

查看read_file文档（?FastUKB::read_file）确认参数要求，比如显式指定分隔符：

raw_data = read_file("Covariates1.csv", sep = ",")

5. 重新安装FastUKB包 #

若为包安装不完整或版本bug，重新安装可尝试解决：

remove.packages("FastUKB")
install.packages("FastUKB") # 若为开发版，可从对应仓库安装
library(FastUKB)

内容的提问来源于stack exchange，提问作者user31731726