ABSTRACT GW190425 is the second of two binary neutron star (BNS) merger events to be significantly detected by the Laser Interferometer Gravitational Wave (GW) Observatory (LIGO), Virgo and the Kamioka Gravitational Wave (KAGRA) detector network. With a detection only in LIGO Livingston, the skymap containing the source was large and no plausible electromagnetic counterpart was found in real-time searching in 2019. Here, we summarize Asteroid Terrestrial-Impact Last Alert System (ATLAS) and Panoramic Survey Telescope and Rapid Response System (Pan-STARRS) wide-field optical coverage of the skymap beginning within 1 and 3 h, respectively, of the GW190425 merger time. More recently, a potential coincidence between GW190425 and a fast radio burst FRB 20190425A has been suggested, given their spatial and temporal coincidences. The smaller sky localization area of FRB 20190425A and its dispersion measure led to the identification of a likely host galaxy, UGC 10667 at a distance of 141 ± 10 Mpc. Our optical imaging covered the galaxy 6.0 h after GW190425 was detected and 3.5 h after the FRB 20190425A. No optical emission was detected and further imaging at +1.2 and +13.2 d also revealed no emission. If the FRB 20190425A and GW190425 association were real, we highlight our limits on kilonova emission from a BNS merger in UGC 10667. The model for producing FRB 20190425A from a BNS merger involves a supramassive magnetized neutron star spinning down by dipole emission on the time-scale of hours. We show that magnetar-enhanced kilonova emission is ruled out by optical upper limits. The lack of detected optical emission from a kilonova in UGC 10667 disfavours, but does not disprove, the FRB–GW link for this source.