We do not assume that TFBS load at a given site reduces an individual's biological fitness. Rather, we argue that binding sites that tolerate a higher load are less functionally constrained. This approach, although undoubtedly a crude one, makes it possible to consistently estimate TFBS constraints for different TFs and even different organisms and ask why TFBS mutations are tolerated differently in different contexts. Conceptual and statistical considerations associated with TFBS load are discussed at length in Materials and methods; here we will only outline several major points. First, since binding events limited to minor alleles are likely to be overlooked by a single-genome ChIP analysis, we compute the decrease in PWM match score relative to the major and not the highest-scoring allele as in the 'classic' genetic load metric. In addition, since we focus on the deleterious effects of variation, we have assumed that mutations yielding increased PWM match scores have a load of zero. We avoided the use of negative load values for these 'gain-of-score' mutations, as it is possible that such mutations will often be near-neutral, while in some cases they may even be deleterious.
