Main features of GRAB package

The GRAB is an R package for Genome-wide Robust Analysis designed for Biobank data (GRAB). The main features of the package are as below.

  • Support multiple complex traits including
    • quantitative trait
    • binary trait
    • time-to-event trait
    • ordinal categorical trait
    • longitudinal trait
  • Perform single-variant and set-based association tests
  • Account for sample relatedness using Genetic Relationship Matrix (GRM)
  • Calibrate p-values using normal distribution approximation and Saddlepoint approximation (SPA)
    • are computationally efficient for large data sets (e.g. UK Biobank)
    • can handle unbalanced phenotypic distribution (e.g. case-control imbalance of binary traits)
    • are robust for both common variants and rare variants

For set-based association tests, GRAB package

  • performs Burden test, SKAT, and SKAT-O
  • allows for tests on multiple minor allele frequency cutoffs and functional annotations
  • allows for specifying weights for single variants in the set-based tests
  • performs conditional analysis to identify associations independent from nearly GWAS signals

Supported Approaches

POLMM / POLMM-GENE:

  • Support ordinal categorical trait
  • Single-variant / set-based tests
  • Can account for sample relatedness
  • Reference
    • Bi, Wenjian, Wei Zhou, Rounak Dey, Bhramar Mukherjee, Joshua N. Sampson, and Seunggeun Lee. Efficient mixed model approach for large-scale genome-wide association studies of ordinal categorical phenotypes. The American Journal of Human Genetics 108, no. 5 (2021): 825-839.
    • Bi, Wenjian, Wei Zhou, Peipei Zhang, Yaoyao Sun, Weihua Yue, and Seunggeun Lee. Scalable mixed model approaches for set-based association studies on large-scale categorical data analysis and its application to 450k exome sequencing data in UK Biobank. The American Journal of Human Genetics 110, no. 5 (2023): 762-773.

SPACox:

  • Support (but not limited to) time-to-event trait
  • Support model residuals (whose sum is zero) after fitting a null model to any type of trait
  • Single-variant tests
  • Cannot account for sample relatedness
  • Reference
    • Bi, Wenjian, Lars G. Fritsche, Bhramar Mukherjee, Sehee Kim, and Seunggeun Lee. A fast and accurate method for genome-wide time-to-event data analysis and its application to UK Biobank. The American Journal of Human Genetics 107, no. 2 (2020): 222-233.

SPAmix:

  • Support (but not limited to) time-to-event trait
  • Support model residuals (whose sum is zero) after fitting a null model to any type of trait
  • Can support admixture population or multiple populations
  • Single-variant tests
  • Cannot account for sample relatedness
  • Reference (to be submitted)

SPAGRM:

  • Support (but not limited to) time-to-event trait
  • Support model residuals (whose sum is zero) after fitting a null model to any type of trait
  • Single-variant tests
  • Can account for sample relatedness
  • Reference (to be submitted)

SAIGE / SAIGE-GENE+ (supported later, please refer to SAIGE package):

  • Support quantitative and binary trait
  • Single-variant / set-based tests
  • Can account for sample relatedness
  • Reference
    • Zhou, Wei, Jonas B. Nielsen, Lars G. Fritsche, Rounak Dey, Maiken E. Gabrielsen, Brooke N. Wolford, Jonathon LeFaive et al. Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies. Nature genetics 50, no. 9 (2018): 1335-1341.
    • Zhou, Wei, Zhangchen Zhao, Jonas B. Nielsen, Lars G. Fritsche, Jonathon LeFaive, Sarah A. Gagliano Taliun, Wenjian Bi et al. Scalable generalized linear mixed model for region-based association tests in large biobanks and cohorts. Nature genetics 52, no. 6 (2020): 634-639.
    • Zhou, Wei, Wenjian Bi, Zhangchen Zhao, Kushal K. Dey, Karthik A. Jagadeesh, Konrad J. Karczewski, Mark J. Daly, Benjamin M. Neale, and Seunggeun Lee. SAIGE-GENE+ improves the efficiency and accuracy of set-based rare variant association tests Nature genetics 54, no. 10 (2022): 1466-1469.

License

GRAB is distributed under an GPL license.

Contact

If you have any questions about GRAB package, please contact wenjianb@pku.edu.cn