It is an interesting study that demonstrates how the retroactive analysis of public sequencing data can contribute new knowledge and applications. The idea of designing synthetic bacterial consortia on the computer based on correlations observed in real data has great potential and could be applied to other cases where faecal transplantation has been shown to work.

The fact that faecal transplantation works as a treatment indicates that the contribution of some bacteria re-establishes a healthy balance, but in addition to these useful bacteria, many others are transferred, which poses a certain danger. The proposed approach selects the bacteria with the greatest healing potential, which are those that correlate negatively with the pathological situation.

The study goes beyond design and tests its effectiveness in a mouse model. The results also identify a potential mechanism that would explain the antagonism between these bacteria and C. difficile, the competition for the amino acid proline, and among all the antagonistic bacteria they find one that by itself functions like C. difficile. Although the results have to be validated in humans, the study points to a new strategy for generating probiotics based on the design of consortia based on computational meta-analysis of many studies.