Cotton (Gossypium spp.) is the most important natural fiber crop worldwide. The diversity of Gossypium species also provides an ideal model for investigating evolution and domestication of polyploids. However, the huge and complex cotton genome hinders genomic research. Technical advances in high-throughput sequencing and bioinformatics analysis have now largely overcome these obstacles, bringing about a new era of cotton genomics. Here, we review recent progress in Gossypium genomics based on whole genome sequencing, resequencing, and comparative genomics, which have provided insights about the genomic basis of fiber biogenesis and the landscape of cotton functional genomics. We address current challenges and present multidisciplinary genomics-enabled breeding strategies covering the breadth of high fiber yield, quality, and environmental resilience for future cotton breeding programs. Gossypium Genomics at a Glance As the world's most important fiber crop and a major source of seed oil and protein, cotton is cultivated in more than 75 countries around the globe [1]. Cotton fibers are seed trichomes, up to 65 mm long, composed of almost pure cellulose, and provide a unique single-celled model system for studying cell elongation and cell wall biogenesis [2]. The Gossypium genus is extraordinarily diverse, with eight diploid genome groups (A-G, and K) and one allopolyploid group (AD) [3,4]. Hybridization and polyploidization of two parental diploids, an A genome-like species with a D genome-like species, has resulted in at least seven allotetraploid species. Two allotetraploid species, Gossypium hirsutum and Gossypium barbadense, evolved independently; these two account for over 90% of annual commercial fiber production. Gossypium is ideal for investigating the origin, evolution, and domestication of polyploids [5-7]. Highlights Cotton is an important natural fiber crop cultivated worldwide that also provides an ideal model for investigating evolution and domestication of polyploids Combinations of the latest technologies, such as optical mapping, high-throughput chromosome conformation capture (Hi-C), and Pacific Biosciences (PacBio) long-reads, have been used to generate multiple high-quality reference genomes of diploid and allotetraploid cotton. Comparative population genomics illuminated the genetic history of cotton domestication and identified the genomic variation determining fiber yield, quality, and stress resistance.