Quickstart

Installation

R

To install the latest release from R-multiverse:

Sys.setenv(NOT_CRAN = "true");
install.packages("bedrockbio", repos = c("https://community.r-multiverse.org", "https://cloud.r-project.org"));

Python

To install the latest release from PyPI:

pip install bedrock-bio

Usage

R

Load the package (and dplyr for downstream data frame manipulation):

library(bedrockbio)
library(dplyr)

List available datasets:

list_datasets()

Lazily load a dataset:

lf <- load_dataset("ukb_ppp/pqtls")

Inspect the contents before downloading entire dataset into memory:

lf |>
    head() |>
    collect()

Filter rows, select columns, and download into a data frame:

df <- lf |>
    filter(
        ancestry == "EUR", 
        protein == "A0FGR8"
    ) |>
    select(
        chromosome, 
        position, 
        effect_allele, 
        other_allele, 
        beta, 
        neg_log_10_p_value
    ) |>
    collect()

Python

Load the package (and polars for downstream data frame manipulation):

import bedrock_bio as bb
import polars as pl

List available datasets:

bb.list_datasets()

Lazily load a dataset:

lf = bb.load_dataset('ukb_ppp/pqtls')

Inspect the contents before downloading entire dataset into memory:

print(lf.collect_schema())

Filter rows, select columns, and download into a data frame:

df = lf \
    .filter(
        pl.col('ancestry') == 'EUR', 
        pl.col('protein') == 'A0FGR8'
    ) \
    .select(
        'chromosome', 
        'position', 
        'effect_allele', 
        'other_allele', 
        'beta', 
        'neg_log_10_p_value'
    ) \
    .collect()

Convert to a pandas data frame (optional):

df_pd = df.to_pandas()