Postpartum depression (PPD) is among the most frequent and incapacitating conditions following childbirth, with significant consequences for mothers, newborns and families. Genetic factors have been proposed to influence disease risk and symptom heterogeneity, and can potentially inform drug target discovery and treatment strategies. Here, we conducted a genetic association study to further our understanding of the genetic architecture of PPD. We identified PPD cases and controls in the UK Biobank using multiple sources of medical history and self-reported information. We performed genome-wide association studies of common and rare variants in in a harmonized set of up to 11,782 PPD cases and 167,480 controls among European-ancestry females. Genetic association results displayed a significantly associated locus at chromosome 18q12.1 led by the common rs10502503 marker (minor allele frequency: 29.9%, effect allele: C, odds ratio: 0.92, p=6.4x10-9), with in silico functional mapping suggesting Cadherin 2 (CDH2) as a candidate causal gene. This signal, if confirmed in independent replication cohorts with PPD diagnosis confirmed through psychometry, may contribute novel insights into the genetic basis of PPD. The results illustrate the use of minimal phenotyping in large-scale general population cohorts to investigate the genetic etiology and heterogeneity of PPD, and to generate therapeutic hypotheses.