Fig. 1: A schematic diagram of our study workflow.

The entire analytical procedure can be segmented into four distinct parts. a Samples were collected, and DNA extraction was performed. Long contigs were obtained by hybrid assembly of short and long reads. b MAG abundance was calculated by read-length mapping after genome binning. c The genome-scale metabolic model of MAG was constructed by protein prediction after genome binning, and the pairwise metabolic complementarity index (MIcomplementarity) of MAGs was calculated. d An RMT-based method was used to generate a co-occurrence network and a metabolic complementary network, and their common edges were analyzed further to provide metabolic clues for potential co-occurrence of species. This figure was created with BioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International license.