Fig. 1

Schematic of the MMinte pipeline. Each green rectangle represents one widget. Widget 1 takes two files, a network of associations between operational taxonomic units (OTUs) and a FASTA file containing the 16S rDNA sequences from a microbiome study, and reduces the latter data set to include only the sequences for OTUs present in the network. Widget 2 identifies the sequences provided and assigns them a genome ID. The percent similarity between the query OTU and the 16S sequence of the genome to which it was matched is stored in a file to be used by Widget 7. Widget 3 calls the ModelSEED service [28] with the list of genome IDs produced by Widget 2, which reconstructs species metabolic models that are exported to the user’s local machine. Widget 4 then uses these species models to create metabolic models for two-species communities. Widget 5 estimates the growth rate of each species in the community under defined metabolic conditions, which can be changed by the user. Widget 6 assesses the types of interaction (mutualism, parasitism, commensalism, competition, amensalism, or neutralism) occurring between the pairs of species in a community based on the effect that each species has on the growth of another. Widget 7 takes the initial information about the topology of the network, the information about the percent similarity between OTUs and the closest genomes, and the types of interactions and plots an interaction network in which the color of the links represents the type of interaction (positive, green; negative, red; no interaction, grey)