is there any method to estimate genetic correlation among two traits considering the samples of two GWASs were overlap

Yes, there are methods to estimate genetic correlation among two traits, considering the samples of two GWASs were overlapping. One such method is LD-score regression, which estimates the genetic correlation between two traits by comparing the genome-wide correlation of linkage disequilibrium (LD) with the observed correlation of effect sizes between the two traits. This method is computationally efficient and has been widely used to estimate genetic correlations between many complex traits. However, it assumes that the genetic architecture of the two traits is similar and that the LD structure is consistent across the two GWASs. Other methods, such as bivariate GREML, can also be used to estimate genetic correlation between two traits, but they require individual-level data and are computationally more demanding than LD-score regression.