MI-System

MI-System performs genotype imputation to infer untyped variants in study-specific genotype data, improving causal variant identification and health-outcome prediction across multi-ethnic populations including East Asian (Taiwanese-Chinese) ancestries.

Key Features:

• Imputation Pipelines: Provides three prephasing-imputation pipelines: SHAPEIT2-IMPUTE2, SHAPEIT4-IMPUTE5, and Beagle5.1.
• Reference Panels: Includes 1000 Genomes and HapMap3 reference panels and a customized Taiwan Biobank (TWB) reference panel tailored to Taiwanese-Chinese ancestry.
• Customization Options: Supports creation of custom reference panels and the combination of two reference panels for imputation.
• Utility Functions: Implements genotype quality control, per-chromosome splitting of whole-genome data, and genome-build conversion.

Scientific Applications:

• Population Genetics: Addresses underrepresentation of non-Caucasian populations by enabling imputation for East Asian ancestries, particularly Taiwanese-Chinese, in population-genetic analyses.
• Disease Association and Causal Variant Discovery: Facilitates identification of causal variants in complex-disease association studies by imputing untyped variants.
• Health-Outcome Prediction: Improves genotype coverage used in health-outcome prediction models and polygenic analyses.
• Cohort and Patient-Driven Studies: Supports imputation workflows applicable to cohort-based and patient-driven genomic research initiatives.

Methodology:

Imputation is performed via prephasing-imputation pipelines SHAPEIT2-IMPUTE2, SHAPEIT4-IMPUTE5, and Beagle5.1 using reference panels (1000 Genomes, HapMap3, TWB) and utilities for genotype QC, chromosome splitting, genome-build conversion, and optional custom or combined reference panels.