I continue watching the ISME19 KBase workshop: “From Reads to Function, Day 2.” Jared Ellenbogen presented next. They spoke about “DRAM and genome-resolved inference.” Ellenbogen spoke about analyzing the fifteen genomes studied during the workshop with DRAM. The outputs of DRAM were described: a heat map with every row corresponding to a different genome or MAG. Every column is a pathway or protein complex/subunits. Ellenbogen noted that not every gene may be present for a pathway: the heat map corresponds to level of completeness for that pathway. For a multi subunit pathway or protein complex, each one is noted as present or absent. Percent completion for each pathway is reported. The group discussed scenarios for potential missing components of pathways, such as a MAG that is incomplete. The spreadsheet produced has tabs for different sets of pathways/functions. Hovering over a box in the heat map will provide more information: function ID, organism/MAG, EC number… Ellenbogen noted that if an organism has a full electron transport chain and TCA cycle, they perform respiration. If there is no external electron acceptor, energy generation may be through fermentation. Again, genome completion is an important consideration. DRAM can also help diagnose what chemical the organism is using. DRAM also provides information about carbon with a block of genes for CAZymes and one for short-chain fatty acids and alcohol conversions. Combinations of CAZymes are needed to break down complex polymers. The CAZymes required a lot of manual curation. Product.html is the interactive heat map file. DRAM also produces a series of spreadsheets and files. The team shared several features they are working on for DRAM2 including graphical outputs and ability to filter and sort results.
