Lake Cajititlán is a subtropical and endorheic lake, which is heavily impacted by nutrient pollution. Agricultural runoff and poorly treated wastewater have entered this reservoir at alarming rates during past rainy seasons, causing the cultural eutrophication of this body of water and resulting in several massive fish kill events. In this study, shotgun metagenomic sequencing was used to examine the taxonomic and functional structure of microbial communities in Lake Cajititlán during the rainy season. Several water quality features and their interactions with microbial communities were also assessed to identify the major factors affecting the water quality and biota, specifically fish species. According to current water quality regulations, most of the physicochemical variables analyzed (dissolved oxygen, pH, Secchi disk, NH4+, NO3−, blue-green algae, total phosphorus, and chlorophyll-a) were outside of the permissible limits. Planktothrix agardhii and Microcystis aeruginosa were the most abundant phytoplankton species, and the dominant bacterial genera were Pseudomonas, Streptomyces, and Flavobacterium, with Pseudomonas fluorescens, Stenotrophomonas maltophilia, and Aeromonas veronii representing the most abundant bacterial species. All of these microorganisms have been reported to be potentially harmful to fish, and the latter three (P. fluorescens, S. maltophilia, A. veronii) also contain genes associated with pathogenicity in fish mortality (fur, luxS, aer, act, aha, exu, lip, ser). Genetic evidence from the microbial communities analyzed herein reveals that anthropogenic sources of nutrients in the lake altered genes involved in nitrogen, phosphorus, sulfur, and carbon metabolism, mainly at the beginning of the rainy season. These findings suggest that abiotic factors influence the structure of the microbial communities, along with the major biogeochemical cycles of Lake Cajititlán, resulting in temporal variations and an excess of microorganisms that can thrive in high-nutrient and low-oxygen environments. After reviewing the literature, this appears to be the first study that focuses on characterizing the water quality of a subtropical hypereutrophic lake through associations between physicochemical variables and shotgun metagenomic data. In addition, there are few studies that have coupled the metabolism of aquatic ecosystems with nutrient cycles.