Up to 600 million people in the world are facing obesity, which is a prime risk factor for heart disease, type 2 diabetes and stoke. Previous studies in our laboratory defined the first evidence of polyphenism of obesity that Trim28+/D9 mice show bi-stable body weight distribution (Dalgaard et al., 2016). We are now working on the interaction of Trim28 and potential imprinting genes, DNA methyltransferases and even histone methyltransferases to trigger, amplify and reverse Trim28 haploinsufficiency-induced obesity. To do so, we will refine the strategy to get stable polyphenism of obesity in mice by in vitro fertilization. We also want to perform RNA-sequencing to predict future fate of obesity at early stages from embryos. Besides, we will further clarify the genetic circuitry of polyphenism of obesity to understand the frequency of polyphenism of obesity. Finally, we will approach the intergenerational and environmental effects on polyphenism of obesity. Overall, the studies give us brand-new concepts on phenotypic variation on epigenetic obesity.