In total, 22 of the 50 monitored honey bee colonies died throughout the study, although samples were obtained for only 4 of those samples when they were dead.

The majority of the colonies (i.e., 16 of 22) died between the August and October 2016 sampling dates. Therefore, we evaluated the differences in virus abundance and pathogen composition between colonies that lived or died by the end of the study for the August sampling date (i.e., the sample date prior to colony death by October 2016). Colonies that died previous to the August 2016 time point are not included in this table. The 'October status' column indicates whether the colony was alive or dead by the next sampling time point in October 2016. (XLSX)