Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.

evo12129-sup-0001-TableS1.doc71KTable S1. The numbers of fixed, low-frequency (DAF < 3%) and high-frequency (DAF > 15%) indels analyzed and numbers of nucleotide sites in each recombination bin.

Figure S1. Neutral and selected single-nucleotide substitutions in D. melanogaster noncoding regions with different recombination rates. (A) fixed mutations, (B) low-frequency mutations (DAF < 3%). On each panel, the green line corresponds to mutations observed at positions 8–30 of short (<65 nucleotides) introns (neutral mutations), and the blue line corresponds to mutations in all noncoding regions.

Figure S2. Insertion/deletion ratio for fixed indels in short (<75 nucleotides) introns of D. melanogaster. Insertion/deletion ratio increases with recombination rate for indels of lengths 1-4 nucleotides, demonstrating a pattern similar to that observed for indels in all noncoding regions (Fig. 1B). Error bars are 95% confidence intervals based on 1000 bootstrap trials.

Figure S3. Single-nucleotide mutation pattern and GC-content as functions of the recombination rate in positions 8–30 of D. melanogaster short (<65 nucleotides) introns. (A) W[RIGHTWARDS ARROW]S/S[RIGHTWARDS ARROW]W ratio, calculated as the number of W[RIGHTWARDS ARROW]S substitutions per A(T)-site, divided by the number of S[RIGHTWARDS ARROW]W substitutions per G(C)-site. Error bars are 95% confidence intervals based on 1000 bootstrap trials. (B) GC-content. The independence of GC-content of the recombination rate suggests that gBGC has no substantial effect on D. melanogaster genome composition.

Figure S4. Distribution of intron lengths for the introns longer >70 nucleotides in regions of D. melanogaster genome with low (ρ < 0.01), intermediate (0.01 < ρ < 3.66), and high (ρ > 3.66) recombination rates.

Figure S5. Insertion/deletion ratio for fixed indels in D. melanogaster short (less than 65 nucleotides) introns. The difference between this figure and Fig. S2 is that introns of lengths 65–74 nucleotides are excluded here. Error bars are 95% confidence intervals based on 1000 bootstrap trials.

Figure S6. Length distribution of polymorphic and fixed indels in noncoding regions of D. melanogaster genome. (A) fixed indels, (B) low-frequency indels, (C) high-frequency indels.

Figure S7. Numbers of analyzed indels in each recombination bin of H. sapiens (A) and S. cerevisiae (B) genomes. Spo11 binding ratio was used as a proxy for the recombination rate in S. cerevisiae.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.