Metabolism is one of the most central aspects of the biology of all organisms. All cells possess a bipartite metabolic network, composed of small molecule metabolites and enzymes which carry out biochemical reactions on them. This metabolic network is responsible for carrying out numerous cellular functions, including energy generation in the form of ATP, production of antioxidants to control reactive oxygen species levels, and production of biosynthetic molecules necessary for cellular growth, such as amino acids and nucleic acids. However, in Chapter 1, I will describe how one of the most important facets of metabolism, the rate at which each biochemical reaction occurs, or metabolic flux, cannot be easily experimentally measured on a genome-wide scale. This leads us to the need for a computational method that can efficiently infer such metabolic flux. I will describe the approach taken by a group of methods that go by the general name of constraints-based modeling, and how we have previously developed a new method under this framework, called FALCON, which uses metabolic gene expression to predict flux. I will then describe how I applied FALCON to infer differences in metabolic flux in two major groups of metabolic diseases. First, metagenomic sequencing has revealed that the composition of the gut microbiome is linked to several major metabolic diseases, including obesity, type 2 diabetes (T2D), and inflammatory bowel disease (IBD). I used the computational tool PICRUST to infer species-specific metagenomes for each of these diseases, and FALCON to infer fluxes from these results. I discovered that several major pathways, previously shown to be important in human host metabolism, have significantly different flux between the two groups. I also modeled metabolic cooperation and competition between pairs of species in the microbiome, used this to determine the compositional stability of the microbiome, and found that that the microbiome is generally unstable across controls as well as metabolic microbiomes. Second, I also used RNA-Seq data from The Cancer Genome Atlas (TCGA) as input to FALCON. I found a systematic difference in that cancer tissues have a considerably stronger correlation between RNA-seq expression and inferred metabolic flux, which may indicate a more streamlined and efficient metabolism. I also found several pathways that frequently have divergent flux. Among these are sphingolipid metabolism, methionine and cysteine synthesis, and bile acid transformations.