Skip to content

Cache kinship-aware null models to disk #81

@vineetver

Description

@vineetver

Kinship-aware null models (REML/PQL) are recomputed on every resume because the sparse Cholesky factor is not serializable in v1 of the null model cache (null_model_cache.rs line 10-12).

For large kinship analyses this wastes minutes on each restart. Serialize the factor or cache the derived quantities (residuals, variance components, projection matrices).

Metadata

Metadata

Assignees

No one assigned

    Labels

    performanceOptimization and profilingstaarSTAAR rare-variant association

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions