search

nadavbra / ukbb_parser

A Python module for loading phenotypic and genetic data from the UK Biobank.
34 stars 6 forks source link

create dataset error #4

Closed jingali closed 2 years ago

jingali commented 2 years ago

For the command line, create_ukbb_phenotype_dataset --phenotype-specs-file=./phenotype_specs.py --output-dataset-file=./ukbb_dataset.csv --output-covariates-columns-file=./ukbb_covariate_columns.json No output files can be used. How to create the dataset by ukbb_paths.py?

nadavbra commented 2 years ago

What do you mean by "no output files can be used"? Have you opened your ukbb_paths.py file and followed the instructions within it? Do you have access and have you downloaded all the UKB files (both the main CSV with the phenotypes, and the genetic data files)?

jingali commented 2 years ago

Yes, I have opened my ukbb_path.py, I don't have access to UKB files and don't know how to get them. And I tried to create the files using my own dataset, but there are so many files except phenotypic and fam files. Thus I opened the ticket before in pwas to get the related files directly.

nadavbra commented 2 years ago

If you don't have access to the UKB then you shouldn't use ukbb_parser. You should provide whatever phenotypic data you do have directly to the pwas package without using create_ukbb_phenotype_dataset (or any other ukbb_parser command). The format accepted by PWAS is very simple: just provide it with a CSV file where each row represents an individual and each column represents a phenotype or covariate you'd like to include in the analysis. If you have further questions about how to use PWAS, please ask them in the pwas GitHub repository and not here.