Skip to content

Latest commit

 

History

History
150 lines (114 loc) · 4.1 KB

README.md

File metadata and controls

150 lines (114 loc) · 4.1 KB
Raw

UKBsearch

UKBsearch is a search tool to retreive term(or terms) from UKBiobank HTML files and tab files downloaded in the local drive.

Installation

  • from pypi
pip install ukbsearch
  • from github
pip install https://github.com/danielmsk/ukbsearch/raw/main/dist/ukbsearch-0.2.2-py3-none-any.whl

Dependency

This UKBsearch requires the following packages:

  • rich
  • pyreadr
  • prettytable
  • pandas
  • pytabix

Options

optional arguments:
  -h, --help            show this help message and exit
  
  -v, --version         show program's version number and exit
  
  -s, --searchterm      search. terms (ex: age smoking)
                        -s age
                        -s age smoking
                        -s 'smok*'
                        -s '*age' 'smok*'
                        
  -l, --logic           logical operator for multiple terms [or(default), and]
                        -s '*age' 'smok*' -l and
                        -s age 'smok*' -l or
  
  -o, --out             title of output file
                        -o searchresult_20220322
  
  -t, --outtype         output type [console(default), csv, udi]
                        -t csv
                        -t console csv
                        -t udi
                        -t console udi
  
  -p, --path            directory path for data files (.html, .Rdata) (default: /data2/UKbiobank/ukb_phenotype)
                        -p /other/path/for/ukb/html/.
  
  -u, --udilist         FileID and UDI list for saving data from tcf files
                        -u ukb39003 3536-0.0 3536-1.0 3536-2.0
  
  -d, --savedata        save data from .Rdata [csv, rdata]
                        -d csv
                        -d rdata
                        -d csv rdata
                        
  -i, --index
                        index tab file and make tcf file (ex. ukb39003.tab)

Usage

Search result

ukbsearch -s 'ag*' 'smok*' -l and

Search for single term

ukbsearch -s age
ukbsearch --searchterm age
ukbsearch -s 'ag*'
ukbsearch -s '*ge'

Search for multiple terms

  • The logical operators (and or or) are supported.
ukbsearch -s age smoking
ukbsearch -s age smoking -l or
ukbsearch -s age smoking -l and
ukbsearch -s 'ag*' 'smok*' -l and

Print only html and UDI

ukbsearch -s 'ag*' 'smok*' -l and -t udi

Save the search result as csv file

ukbsearch -s 'ag*' 'rep*' -l and -o test1 -t csv
(= ukbsearch --searchterm 'ag*' 'rep*' --logic and --out test1 --outtype csv)
ukbsearch -s 'ag*' 'rep*' -l and -o test1 -t console csv
ukbsearch -s 'ag*' 'rep*' -l and -o test1 -t console udi csv

Set a particular directory

  • The default path is /data2/UKbiobank/ukb_phenotype.
ukbsearch -s age -p /other/path/for/ukb/html/.

Index tab file

ukbsearch -i ukb26086.tab

This step generates .tab.tcf.gz, .tab.tcf.gz.tbi, and .tab.tcf.gz.idx. After generating tcf files, the tab file is no longer required to search.

Save data (.csv and .rdata) from .tcf.gz

ukbsearch -u ukb39003 3536-0.0 3536-1.0 3536-2.0 -d csv -o test3
(=ukbsearch --udilist ukb39003 3536-0.0 3536-1.0 3536-2.0 --savedata csv --out test3)
ukbsearch -u ukb39003 3536-0.0 3536-1.0 ukb26086 20161-0.0 21003-1.0 -d csv rdata -o test3

ukbsearch -s 'ag*' 'rep*' -l and -d csv -o test3
ukbsearch -s 'ag*' 'rep*' -l and -d rdata -o test3

Version History

  • 0.2.2 (2022-04-05)
    • change saving type for a single file
    • remove csvi (inversed form) option.
  • 0.2.1 (2022-03-25)
    • add csvi (inversed form) option.
    • debug unsaved values issue.
  • 0.2.0 (2022-03-24)
    • implementated tab file indexing based on tabix.
  • 0.1.1 (2022-03-23)
    • changed default path to /data2/UKbiobank/ukb_phenotype
  • 0.1.0 (2022-03-21)
    • first released.