GEDI-ADMX – Genotype Error Detection and Imputation for Admixed Populations


The GEDI-ADMX package provides methods for

  • local ancestry inference
  • SNP genotype error detection and correction
  • imputation of missing genotypes at typed SNPs, and
  • imputation of genotypes at untyped SNPs

in admixed populations based on whole-genome SNP genotype data
and reference haplotype panels for ancestral populations.

Currently GEDI-ADMX handles genotype data from unrelated individuals.
Local ancestry inference uses a factorial HMMs trained on ancestral
haplotypes to impute genotypes at all typed SNP loci (temporarily marking
each SNP genotype as missing) under each possible local ancestry.
GEDI-ADMX assigns to each locus the local ancestry that yields the highest
imputation accuracy, as assessed using a weighted-voting scheme based on
multiple SNP windows centered on the locus of interest. Error detection
and imputation of missing genotypes at typed SNPs is performed using
multilocus genotype probabilities computed based on the factorial HMM
corresponding to the inferred local ancestry. Imputation of genotypes at
each untyped SNP is performed based on posterior genotype probabilities
computed using a similar factorial HMM spanning k (default k=10) typed
SNPs before and after the imputed locus.

For further details on the GEDI-ADMX statistical models and algorithms see the
references below.

GEDI-ADMX source code and executables

Contact Information

Related Publications

Related Presentations

Acknowledgment and Disclaimer

This material is based upon work supported in part by the National Science Foundation under Grants No. IIS-0546457, IIIS-0916948 and DBI-0543365. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.