GROMTools: scalable individual-level GReX imputation for mega-biobank-scale cohorts.

GROMTools enables nearly identical GReX predictions for mega-biobank cohorts with 100-fold faster computation and 33-fold less memory than existing tools.

• Why it matters: As biobank-scale datasets grow, current methods struggle with high computational costs, limiting routine and cost-effective individual-level GReX imputation at large scales.
• What they did: The team developed GROMTools, an open-source C++ engine with an R interface, that streams genotype data, exploits sparse weights, and produces compact outputs, tested on UK Biobank data with 50,000–450,000 individuals.
• The result: GROMTools achieves high accuracy (Pearson >0.999) while drastically reducing CPU time and memory, making large-scale GReX imputation practical and affordable on standard CPU infrastructure.