Table 2.
SNP | Gene | Minor allele | Reported effect allele | Risk allele/weight loss lowering allele | MSE (%) | |
---|---|---|---|---|---|---|
GRSwBMI, weighted BMI | rs571312 | MC4R | A | A | C | 23.16 |
rs3810291 | TMEM160 | G | A | G | 19.65 | |
rs1555543 | PTBP2 | A | C | A | 18.58 | |
rs206936 | NUDT3 | G | G | G | 17.58 | |
rs987237 | TFAP2B | G | G | G | 17.49 | |
rs4836133 | ZNF608 | A | A | A | 17.12 | |
rs2241423 | MAP2K5 | A | G | G | 13.10 | |
GRSwWHR, weighted WHR | rs1443512 | HOXC13 | A | A | A | 13.15 |
rs4846567 | LYPLAL1 | A | G | G | 11.24 | |
rs1011731 | DNM3-PIGC | G | G | A | 10.32 |
The listed single nucleotide polymorphisms (SNP) induced a mean squared error (MSE) above 10 % in the random forest model. These variants were subsequently included in the respective GRS model (weighted BMI model (GRSwBMI) and weighted waist-hip ratio model (GRSwWHR)). Those alleles were defined as risk alleles that contributed least to the weight loss following Roux-en Y gastric bypass surgery. Interestingly, the A allele of the TMEM160-related SNP rs3810291 was reported to be associated with higher BMI by Speliotes et al. In our study, the allele, G, is associated with EBMIL