Multi-omic data and genome-scale microbial metabolic models have allowed us to examine microbial communities, community
function, and interactions in ways that were not available to us historically. Now, one of our biggest challenges is determining
how to integrate data and maximize data potential. Our study demonstrates one way in which to test a hypothesis by combining
multi-omic data and community metabolic models. Specifically, we assess hydrogen sulfide production in colorectal cancer based on
stool, mucosa, and tissue samples collected on and off the tumor site within the same individuals. 16S rRNA microbial community
and abundance data were used to select and inform the metabolic models. We then used MICOM, an open source platform, to track the
metabolic flux of hydrogen sulfide through a defined microbial community that either represented on-tumor or off-tumor sample
communities. We also performed targeted and untargeted metabolomics, and used the former to quantitatively evaluate our model
predictions. A deeper look at the models identified several unexpected but feasible reactions, microbes, and microbial
interactions involved in hydrogen sulfide production for which our 16S and metabolomic data could not account. These results will
guide future in vitro, in vivo, and in silico tests to establish why hydrogen
sulfide production is increased in tumor tissue.