Methods were published as a protocol openly available here: https://openscience.bmj.com/content/3/1/e000017 **Briefly:** **Intervention:** “Bimuno®” Powder, which contains Galacto-oligosaccharides (B-GOS), lactose, glucose, and galactose **Control:** 50% lactose, 27% glucose, and 23% galactose (purchased from Sigma Aldrich **Administration:** Prebiotic and control treatment were administered for 28 consecutive days at a dose of 4g/kg. Administered orally via syringe-feeding. **Modelling Depression:** Flinders Sensitive Line rats bred in-house at the Translational Neuropsychaitry Unit, Aarhus University. **Model Control:** Flinders Resistant Line rats bred in-house. **Primary Outcome:** behaviour in the forced swim test. **Secondary Outcomes:** Elevated Plus Maze, open field test, body weight. Random allocation to intervention group, allocation concealment, blinded assessment of outcome, prespecified exclusion criteria. **Deviations to the procotol:** *Age of the Animals:* The protocol specified that 5-7 weeks old animals would be used. In this experiment animals between 8-14 weeks were used instead, as the breeding rate of the animals in the breeding facility was not sufficient to provide 64 animals required to be born within a 2-week age range. *Control Treatment:* There was an error in the protocol, which specified that the control would be the “Bimuno Free Powder”. The is in fact not a product. The control given was made up of similar components to the Bimuno® powder but without the active ingredient B-GOS, consisting of commercially sourced 50% lactose, 27% glucose, and 23% galactose. This control was the same control as was administered in Savignac et al., 2016. *Volume of Treatment:* The volume of the solution fed via syringe was reduced from 2ml to 1.5ml. This smaller volume was easier for the animals to consume, especially during the training of syringe feeding. We tested the weight of the powders each week to ensure that they properly dissolved in the amount of tap water, so that dosing was as accurate as possible. *Length of Outcome Assessment OFT:* We administered the OFT for 5 minutes instead of 15mins specified in the protocol. The primary aim of administering the OFT was to measure locomotor activity, which can be accurately assessed with a 5-minute testing period, so the length of the test was reduced to reduce the unnecessary stress to the animals. Additional Outcome Analysis: We could not secure additional funding to carry out neurochemical brain analyses, therefore we did not dissect the brains from the animals at euthanisation. *Microbial Outcome Assessment:* We collected faecal boli on the first day behavioural testing (day 24) instead of on the day of euthanisation (day 28) with the aim of isolating the effects of prebiotics on gut microbiota, without the stress of the behavioural testing. Due to an error in the shipping process analysis of the faecal boli samples was not carried out. *Statistical Analysis:* We did not specify in the protocol that correction for multiple testing would be applied. With 7 outcomes investigated and the conventional level of significance of alpha = 0.05, a Holm-Bonferroni correction to account for multiple testing is applied gives a threshold of 0.0073.