 Hello, everyone. So first of all, I'd like to thank the organizing committee for giving me this opportunity to share our work so this work was done in collaboration with Xiaoyue Wang, Megan McNerney and Kevin White, all at University of Chicago. So we also have a poster number 76 displayed in the hallway just outside the cafeteria. You're welcome to view it and let me know if you have any comments. So several people actually viewed it yesterday during my poster presentation and gave me very thoughtful comments. So okay, so I'll start with the definition of epistasis or genetic interaction. I'm going to use the two terms interchangeably. I'm also going to introduce this concept through a historical perspective. So this picture here illustrates this concept. This phenotype one is the phenotype of the wild type and then you can have point seven as the phenotype for mutant A and point five for the phenotype of mutant B. And then if the two genes are independent and we assume a multiplicative model, then the phenotype outcome for the double mutant would be point seven multiplied by point five. And if you have a double mutant with a phenotype different from this expected phenotype, then that would be an indication of the two genes in epistasis. In particular, if we look at these two cases, they're identical to one of the single mutants. And these are called masking. Masking was initially referred to as epistasis by basin. More than a century ago. A few years later, Fisher extended this concept to departure from an additive model. In our project, we define epistasis as departure from an additive or multiplicative model.