Chronic kidney disease (CKD) is a common disease, affecting up to 10% of the general population, and causing significant morbidity and mortality. Apart from the risk conferred by traditional, cardiovascular risk factors, there is a strong genetic component to CKD risk. GWAS is a powerful method to unravel this genetic component by means of the unbiased association analysis of several million genetic variants distributed across the complete genome. Since the publication of the first GWAS in 2005, this method has lead to the discovery of novel loci for almost every human common disease and phenotype. Recently, GWAS have uncovered several novel loci associated with estimated glomerular filtration rate (eGFR), CKD and albuminuria in cross sectional analyses of general population cohorts. However, less than 2% of total variance of eGFR and albuminuria is explained by the identified variants, and relative risk for CKD is modified by at most 20% per locus. Further, while most of these loci are also associated with incident CKD in longitudinal analyses, only two loci are associated with the hardest renal end point, end stage renal disease (ESRD). The effort to uncover the missing heritability of renal phenotypes is hampered by the multifactorial nature and different pathogenic pathways involved at progressive stages of CKD, and by the challenges in elucidating the role of low frequency variants which are likely to have larger effects. While in African Americans much of the risk for nondiabetic ESRD is explained by common variants in the MYH9/APOL1 locus, genetic variants with large effect sizes have not been identified in individuals of European descent. The analysis of cohorts with larger sample sizes and with longitudinal renal phenotypes, using high resolution genotyping (1000G, sequencing) is ongoing work aimed at uncovering novel loci associated with severe renal phenotypes.