Determining Ancestral Histories in Admixed Populations

Newswise — A new study by Dr. Eran Halperin of ICSI and colleagues provides a means of pinpointing the ancestry of each position on an individual's genome. This information can be used to reconstruct ancestral history, which can then be used in studies of complex genetic diseases. Results of the study are published in the February issue of the American Journal of Human Genetics.

Large-scale genotyping of SNPs (single nucleotide polymorphisms, mutations that occurred once in history and then were passed on through heredity and became prevalent in a population) has been used extensively to identify markers that are associated with diseases. There are ~10 million SNPs in the human genome that differ between individuals.

The pattern of genetic differences between individuals varies across different populations. In studies that involve more than one population, spurious associations between a SNP and a disease might be found due to the differences in the population patterns and not due to the disease. In order to overcome this, methods exist to estimate the ancestry of each individual in a study, based on his or her genome. However, accurate methods for finding the ancestry of individuals of recently admixed populations (such as African Americans, Latin Americans, and diverse urban populations such as San Francisco) do not exist. In these populations, each position in the genome could have been transmitted from a different ancestry.

The LAMP (Local Ancestry in adMixed Populations) method developed by Dr. Halperin in collaboration with Dr. Sriram Sankararaman (UC Berkeley), Dr. Srinath Sridhar (Carnegie Mellon University), and Dr. Gad Kimmel (ICSI and UC Berkeley is able to identify the ancestry of each position in the genome for individuals from admixed populations. The researchers performed extensive simulation studies to evaluate their method, and have shown that it is highly accurate and can precisely identify ancestral regions in the genome for typical cases of individuals that are a mix of two or three populations.

The main motivation for applying LAMP to genetic data is to avoid spurious results in disease association studies. However, LAMP will also be useful in improving studies of diseases that are much more prevalent in one population than another. For instance, Multiple Sclerosis (MS) is much more common in Northern European than in other populations, and Alzheimer's disease is more common in African Americans than in Caucasians. The combination of this variation in prevalence, together with methods such as LAMP, can be used to study these diseases through studies that are referred to as admixture mapping. As an example, in MS one can genotype African-American cases and search for regions in the genome that are of Northern European ancestry. Since MS is more prevalent in Northern Europeans than in Africans, such genomic regions are suspected to be related to MS. The LAMP method accurately provides the ancestry of every single SNP in a person's genotype, enabling scientists to perform such admixture mapping studies more accurately than ever before.

The International Computer Science Institute (lCSI) is a leading innovator in the field of computer science. Its international team of staff scientists and visitors develop cutting-edge technology and applications in computer networking, bioinformatics, artificial intelligence, computer architecture, and speech and language processing.