Abstract: A method of electronically separating host and non-host sequence data (sequenced DNA, RNA, and/or proteins) utilizes electronic host filters, which can be generated on a just-in-time basis by a cloud-based software service. Host reads and non-host reads of a given sample are separated and stored in separate data repositories. Also disclosed is a cloud-based software service utilizing the method. The non-host reads resulting from the host filtration process can then be profiled more accurately for the microorganism content therein.

Abstract: A method is describe for predicting the viability and growth rate of at least one prokaryotic species of a non-cultured working sample (e.g., environmental, medical, food). The method utilizes a reference database containing time-dependent sequence data obtained from control samples of prokaryotic species at various stages of growth. The sequence coverages of the control samples are used to identify at least two regions of each genome active in replication. Importance scores and weight scores are assigned to these active regions. The sequence coverages over the active regions obtained for the working sample are used with the importance and weight scores to predict growth and viability of one or more prokaryotic species of the working sample.