Fig. 2From: Fast computation of the eigensystem of genomic similarity matricesCovariance matrix. First two principal components colored by population for the 1000 Genome Project dataset. Truncated SVD (left) and Algorithm 2 (right)Back to article page