Abstract
Centromeric repetitive DNA is the largest missing piece of the Human Genome Project. Rather than being necessary or sufficient to specify the site of mitotic spindle attachment to the chromosome, centromeric DNA sequence is all but irrelevant. Instead, centromeres are thought to be specified by a protein component, a histone H3 variant called Centromere Protein A (CENP-A). This review includes a brief overview of the history of the centromere field, and the current status of knowledge on CENP-A assembly. New evidence for CENP-A recruitment in response to DNA damage implies a mechanism for neocentromere formation, and raises new questions about the epigenetic model of centromere maintenance.