FINEMAP-miss

FINEMAP-miss extends the FINEMAP model to account for variant-specific missingness in fine-mapping of genome-wide association studies (GWAS), particularly in meta-analysis settings where sample size varies across variants.

Key Features:

• Variant-specific missingness modeling: Extends the FINEMAP model to explicitly account for missing genotype information that varies across variants.
• Meta-analysis calibration: Provides well-calibrated fine-mapping in GWAS meta-analyses where standard fine-mapping methods fail due to variation in sample size across variants.
• Summary statistics imputation alternative: Offers clear improvement over summary statistics imputation when causal variants have low imputation information or when sample size or meta-analysis complexity increase.
• Open-source implementation: Available as an R package named finemapmiss.

Scientific Applications:

• GWAS meta-analysis fine-mapping: Enables statistical fine-mapping to jointly evaluate the probability of causality of all variants in a genomic region of interest in combined multiple-study GWAS.
• Breast cancer GWAS: Successfully applied to breast cancer GWAS meta-analysis where standard fine-mapping and summary statistics imputation are not applicable.

Methodology:

FINEMAP-miss extends the FINEMAP model to account for variant-specific missingness; evaluation performed through meta-analysis simulations comparing calibration against standard fine-mapping and summary statistics imputation approaches.