Diminished exposure to harmless micro-organisms, such as lactobacilli, has been suggested to play a role in the increased prevalence of allergic disorders in Westernized communities. The development of allergies depends on both environmental factors and genetic variations, including polymorphisms in genes encoding pattern recognition receptors. The present study examines the effects of both colonization with specific Lactobacillus species and genetic variations in DC-SIGN, a pattern recognition receptor on dendritic cells that recognizes lactobacilli, on the development of atopic dermatitis (AD) and sensitization in infancy. Within the KOALA Birth Cohort Study, faecal samples of 681 one-month-old infants were collected and quantitatively screened for five Lactobacillus species: L. casei, L. paracasei, L. rhamnosus, L. acidophilus and L. reuteri. Eleven haplotype-tagging polymorphisms in the DC-SIGN gene were genotyped in these children. Allergic outcomes were a clinical diagnosis of AD and sensitization (specific IgE) at age 2 years. L. rhamnosus (31.5 %), L. paracasei (31.3 %) and L. acidophilus (14.4 %) were frequently detected in the faecal samples of one-month-old infants, whereas L. casei (2.5 %) and L. reuteri (<1 %) were rare. Colonization with L. paracasei decreased the risk of AD significantly (odds ratio 0.57, 95 % confidence interval 0.32-0.99), whereas effects of L. acidophilus were of borderline statistical significance (0.46, 0.20-1.04). Two DC-SIGN polymorphisms, rs11465413 and rs8112555, were statistically significantly associated with atopic sensitization. The present study supports the 'old friends' hypothesis suggesting that certain health-beneficial micro-organisms protect us from developing allergies and that these protective effects are species-dependent. Firm conclusions on the potential interaction between lactobacillus colonization and genetic variations in DC-SIGN in association with the development of allergic disorders cannot be drawn, given the limited power of our study. Therefore, incorporation of consecutive faecal sampling in newly started (birth) cohort studies would be a first requisite to further increase our understanding of host-microbial interactions in health and disease.