Figure 3 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Coancestry and identity-by-state as measures of pairwise similarity (A) Pairwise genetic similarities according to coancestry. (B) Pairwise genetic similarities according to IBS. The samples are organized according to subpopulations from the south (bottom left) to the north (top right). Darker colors indicate greater similarity, as shown by the scale bars with numeric ranges of coancestry and IBS. The poor performance of the IBS approach with genotyping errors caused by low-coverage data is clearly visible in the upper right corner: the dark stripes suggest that a set of samples are closely related to all other samples from the subpopulation.
Source publication
This protocol demonstrates how a fragmented reference genome and low-coverage sequencing data can be efficiently utilized for population genetic analyses of a species with highly reduced genetic diversity. Focusing on endangered Saimaa ringed seals, we detail the steps for data preparation and conventional analyses and then extend the latter by com...
Contexts in source publication
Context 1
... the XML-formatted tree file (saimaa_20ctg-s_linked_tree.xml). The coancestry matrix indicates how much other individuals have contributed to each target individual, while the accompanying tree represents the coancestry-based relationship/kinship and groups the samples within the matrix ( Figure 3A). It is notable that the coancestry matrix does not need to be symmetrical and, for specific analyses of pairwise distances, one has to decide how to convert the two coancestry values between each sample pair into one value. ...
Context 2
... used the mean of the values. The IBS similarity matrix calculated with PLINK is symmetrical and only needs be to ordered to match the coancestry matrix ( Figure 3B). ...
Context 3
... estimates of genetic similarity show the population structure within Lake Saimaa and form three blocks along the diagonal line. However, the IBS matrix ( Figure 3B) also shows prominent dark ll OPEN ACCESS stripes, most likely caused by the low sequencing coverage of some samples and the poor power of imputing the missing data. The coancestry estimates are based on extended segments of identity and similar stripes are not seen in the resulting similarity matrix ( Figure 3A). ...
Context 4
... the IBS matrix ( Figure 3B) also shows prominent dark ll OPEN ACCESS stripes, most likely caused by the low sequencing coverage of some samples and the poor power of imputing the missing data. The coancestry estimates are based on extended segments of identity and similar stripes are not seen in the resulting similarity matrix ( Figure 3A). Despite the significantly more demanding computation, we believe that the coancestry-based similarities can be a useful alternative for the IBS distances and provide a more robust measure of genetic distances for lowcoverage and low-variance data. ...
