Skip to main content

New Research Demonstrates Algorithm for Tracing Disease Genes in Mixed Populations

NEW YORK (GenomeWeb News) – A new technique is improving scientists’ ability to detect ancestral disease genes in mixed human populations.
 
Researchers from Washington University in St. Louis and the Israeli Institute of Technology have developed an algorithm that expands the power of detecting and interpreting different DNA markers for revealing ancestry in mixed populations. The authors believe the approach, which was described in the most recent issue of the journal Genome Research, could prove useful for understanding disease in different populations and personalizing medical treatments.
 
The approach "extends previous methods by incorporating knowledge on population admixture, drawing a more precise picture of the mosaic of ancestries along an individual’s genome,” lead author Sivan Bercovici, a computer scientist and bioinformatics researcher at the Israeli Institute of Technology, which is also called Technion, said in a statement.
 
Certain diseases seem to be more prevalent in some populations. This means that an admixed population, which results from two ancestral populations mixing, may have a higher risk for certain diseases than one of its ancestral populations. Consequently, looking at human populations of mixed ancestry can provide a wealth of information about disease-susceptibility genes found in the original populations.
 
The admixture linkage disequilibrium, or MALD, approach exploits these genetic characteristics to understand ancestral populations. By genotyping either individuals in the admixed population alone or individuals in both the admixed and ancestral populations, researchers can find genetic patterns that reflect each ancestral population throughout the genome — information can be applied to find those chromosomal regions associated with differences in disease risk.
 
“MALD requires 200-500-fold fewer markers, in comparison to genome-wide association mapping, while offering the same power,” Bercovici and his colleagues explained in the paper. “Consequently, the method has an economical advantage over alternative methods.”
 
But despite the potential usefulness of this sort of admixture mapping, the approach depends on having adequate information about markers in different ancestral populations. In an effort to address this, the researchers developed an information theory-based measure, dubbed “expected mutual information,” as a means for finding the most useful sets of markers to be assessed by MALD.
 
The EMI “computes the total impact of a set of markers on the ability to infer ancestry at each chromosomal location, averaging all possible recombinations that could have occurred during the admixture process,” the authors explained.
 
They then created an algorithm that selects panels of genetic markers based on EMI scores with an eye to using data that generates the best EMI scores — an approach that maximizes the information gleaned from analyzing admixed populations.
 
Indeed, when they compared their algorithm with simulation tools applied to previous studies, they found that it was more powerful and accurate for determining disease gene loci.
 
“[W]e showed that the panels produced using EMI have a well-balanced high score in terms of informativeness of markers, yielding a significant improvement in both power and accuracy, compared to previous work,” the authors wrote. “The increase in power is particularly important in the detection of weak signals that underlie complex diseases.”
 
For example, when they constructed panels of genetic markers for admixed African-American populations using SNP allele frequencies for West African and European populations that were collected through the International HapMap Project, the researchers were able to get good EMI scores even with relatively few markers.
 
The researchers also produced simulations in which they generated 576 admixed cases using AncestryMap, a tool used for estimating ancestral origins of a certain loci based on genotype and for generating samples of admixed genotypes. They then selected locations on chromosome 1 to represent disease-risk loci and tested their ability to detect them. Again, the EMI approach could detect loci using as few as 100 markers.
 
In the future, the researchers noted, such an algorithm may provide insights into disease-risk genes associated with certain populations — information that could potentially make it possible to tailor disease diagnoses and treatments.
 
“We can look at many different hybrid human populations with this algorithm and use it on a diversity of diseases,” senior author Alan Templeton, biologist and geneticist at Washington University in St. Louis, said in a statement.
 

The Scan

Gap in COVAX Doses

BBC News reports that COVAX is experiencing a vaccine shortfall, as the Serum Institute of India has paused exports.

Sanofi, GSK Report Promising Results

The Wall Street Journal reports that the candidate SARS-CoV-2 vaccine from Sanofi and GlaxoSmithKline has had encouraging early results.

Influence of Luck

The New York Times examines how the US avoided variant-fueled increases in COVID-19 cases.

PLOS Papers on Retina GWAS, Hantaan Virus, COVID-19 Phenome-Wide Association Study

In PLOS this week: genome-wide association study of retinal morphology, analysis of hantaan virus found in a mouse, and more.