Additional file 1: Figure S1. of Pan-cancer analysis of systematic batch effects on somatic sequence variations

Frequency of indels and non-indels in each MAF dataset. Distribution of mutation frequency across MAFs for indel variants (upper) and non-indel variants (lower). *, MAF datasets with no indels are indicated. Figure S2. Number of MAF data for each cancer type. Bar plot shows the number of MAF datasets available for each TCGA cancer type. Figure S3. Homopolymer runs in the flanking sequences of the batch-biased indels. Distribution of long (red) and short (blue) homopolymer runs for each altered nucleotide (A, T, G and C) within the 50 nucleotide sequences flanking each variant, shown for batch-biased indel variants (upper) and unbiased indel variants (lower). Figure S4. Consensus sequences in flanking sequences of batch-biased indels. The sequence-logo plots show consensus sequences in the 15 nucleotide flanking regions of the batch-biased (upper) and unbiased (lower) variants for each of the altered nucleotides (A, T, G and C). (DOCX 602 kb