Background: We recently conducted a double-blinded randomised controlled trial showing that fish-oil supplementation during pregnancy reduced the risk of persistent wheeze or asthma in the child by 30%. Here, we explore the mechanisms of the intervention. Methods: 736 pregnant women were given either placebo or n-3 long-chain polyunsaturated fatty acids (LCPUFAs) in the third trimester in a randomised controlled trial. Deep clinical follow-up of the 695 children in the trial was done at 12 visits until age 6 years, including assessment of genotype at the fatty acid desaturase (FADS) locus, plasma fatty acids, airway DNA methylation, gene expression, microbiome and metabolomics. Results: Supplementation with n-3 LCPUFA reduced the overall risk of non-atopic asthma by 73% at age 6 (relative risk (RR) 0.27 (95% CI 0.06 to 0.85), p=0.042). In contrast, there was no overall effect on asthma with atopic traits (RR 1.42 (95% CI 0.63 to 3.38), p=0.40), but this was significantly modified by maternal FADS genotype and LCPUFA blood levels (interaction p<0.05), and supplementation did reduce the risk of atopic asthma in the subgroup of mothers with FADS risk variants and/or low blood levels of n-3 LCPUFA before the intervention (RR 0.31 (95% CI 0.11 to 0.75), p=0.016). Furthermore, n-3 LCPUFA significantly reduced the number of infections (croup, gastroenteritis, tonsillitis, otitis media and pneumonia) by 16% (incidence rate ratio 0.84 (95% CI 0.74 to 0.96), p=0.009). Conclusions: n-3 LCPUFA supplementation in pregnancy showed protective effects on non-atopic asthma and infections. Protective effects on atopic asthma depended on maternal FADS genotype and n-3 LCPUFA levels. This indicates that the fatty acid pathway is involved in multiple mechanisms affecting the risk of asthma subtypes and infections. Trial registration number: NCT00798226.