Genome-Wide Estimation of Linkage Disequilibrium from Population-Level High-Throughput Sequencing Data

Supporting Information

Supporting Information

• Supporting Information - Figure S1 and Tables S1-S4 (PDF, 346 KB)
• Figure S1 - Predicted sampling standard deviation of ML estimates of LD under a fixed total budget, as a function of the mean depth of coverage μ, when library preparation is three times more expensive than sequencing. (PDF, 134 KB)
• Table S1 - Probability p_g(j) of observed nucleotide read j at site α with two-locus genotype g as a function of the error rate ε. (PDF, 125 KB)
• Table S2 - Probability p_g(j) of observed dinucleotide read j at the two sites of interest with two-locus genotype g as a function of the error rate ε. (PDF, 139 KB)
• Table S3 - Comparison of LD estimates by the proposed ML estimator to those by an imputation-based method when the major allele frequencies at two sites of interest, p and q, are intermediate. (PDF, 136 KB)
• Table S4 - Comparison of LD estimates by the proposed ML estimator to those by an imputation-based method when major allele frequencies at two sites of interest, p and q, are high. (PDF, 136 KB)