BACKGROUND: While genetic factors are important influences on maternal mental health, few studies have used symptom-level analyses to examine how genetic liability is related to the experience of specific mental health problems in mothers. A symptom-level approach can account for disorder heterogeneity and delineate key associations between genetic liabilities and mental health. METHODS: Three waves of data (30 weeks of gestation, 6 and 18 months postpartum) from the Norwegian Mother, Father and Child Cohort Study (MoBa) were used to assess item-level associations between genetic liabilities to depression, anxiety, neuroticism and positive affect, and maternal mental health phenotypes (i.e., symptoms of anxiety, depression, positive and negative affect) using a network analysis approach. Sample sizes ranged from 46,537 to 59,308 mothers. RESULTS: PGSs exhibited both phenotype-specific associations (e.g., depression PGS linked with hopelessness, anxiety PGS linked with worry) and cross-phenotype (e.g., depression PGS linked with nervousness, positive affect PGS inversely related to anxiety and depressive symptoms) relationships, with partial correlations ranging between r = -0.025 and r = 0.024. Some PGS-phenotype associations were consistent (e.g., depression PGS linked with feeling like screaming or banging on something across all waves) and others inconsistent (e.g., anxiety PGS linked with nervousness only at 6 months postpartum) across the perinatal and postpartum periods. CONCLUSIONS: Our findings highlight symptom-level associations between PGSs and maternal mental health, which may be obscured when global measures of mental health (e.g., overall scores) are used. Identifying symptom-specific PGS associations could advance current understanding of aetiological influences on maternal mental health.