Sample-wide shared ancestry
Summary of the ancestry shared between a target individual and all other individuals in the sample
Shared ancestry inference based on genomic data from the Simons Genome Diversity Project
The Coalescent Intensity Function (CIF) was computed from the aggregated Cumulative Coalescent Function (CCF) inferred between the target individual and each comparator individual.How are CCFs aggregated?
CCFs were computed using information from the Atlas of Variant Age, with allele age estimated using data from SGDP alone and under the joint clock model
Population group: Dinka Country: Sudan
Coalescent Intensity Function (CIF) How to interpret this figure?
Colours correspond to the relative intensity of coalescence scaled per epoch (columns)
Comparator individuals (rows) are sorted by the area under the curve (AUC) of CCFs
Download this figure in PNG format
Data shown in this figure can be downloaded from this page.
|CCF||The cumulative coalescent function (CCF) describes the fraction a given target genome shares with a comparator genome due to common ancestry; the curve is monotonically increasing back in time. The CCF was inferred using a dynamic programming method which takes the alleles carried by the target genome, compares those to the alleles carried by the comparator genome, and infers the CCF based on the estimated age of the shared alleles. Here, allele age estimated using the joint clock model was applied throughout.|
|CIF||The coalescent intensity function (CIF) corresponds to the rate of change of coalescence (i.e. common ancestry) during a given time interval ("epoch"). The CIF is computed from the CCF between the target and a comparator individual. Here, CIFs were scaled across the sample per epoch (columns), such that the maximum value is equal to 1 per epoch.|
|Aggregation||The CCF is inferred between two haploid genomes (i.e. phased chromosomes). To summarise the ancestry shared between diploid individuals, the CCFs between each of the two target genomes and the two comparator genomes are "aggregated" by computing the weighted mean across individual chromosomes (weighted by the number of dated variants per chromosome) and averaged per individual.|
|Relatedness ranking||An approximate measure of relatedness within the sample is provided by computing the area under the curve (AUC) of the CCFs inferred between target and comparator individuals (after aggregation). The ranking is obtained by sorting comparator individuals from highest to lowest AUC. The AUC was computed on log-scaled time, thereby giving higher importance to more recently shared ancestry.|