Summix

Summix estimates and adjusts for substructure in genetic summary allele frequency data using a computationally efficient mixture model.

Key Features:

• Reference Group Proportion Estimation: The summix() function estimates reference group proportions using a mixture model.
• Adjusted Allele Frequency Generation: The adjAF() function produces adjusted allele frequencies for an observed group with reference group proportions matching a target individual or sample.
• Local Ancestry and Selection Analysis: The summix_local() function estimates local ancestry mixture proportions and performs selection scans in genetic summary data.
• Fine-Scale Structure Characterization: Summix characterizes finer-scale population structure beyond broad continental groupings.
• Ascertainment Bias Detection: Summix identifies ascertainment bias in genetic summary data.
• Local Substructure Deviation Identification: Summix identifies potential regions of selection due to local substructure deviation.

Scientific Applications:

• Risk Prediction: Summix improves the robust use of genetic summary data for risk prediction in diverse populations.
• Causal Inference: Summix enables more accurate causal inference by accounting for substructure that would otherwise confound analyses.
• Fine Mapping: Summix supports fine mapping applications by addressing heterogeneity in genetic summary data.
• External Control Incorporation: Summix facilitates the incorporation of external controls into genetic studies while adjusting for substructure.
• Admixed Population Genetics: Summix increases the usability of summary data for understudied or admixed populations where unaccounted substructure limits analysis.

Methodology:

Summix uses a computationally efficient mixture model to estimate and adjust for substructure in genetic summary allele frequency data.