Fig. 2: Gene sharing network representation of all training and test viruses used in developing MArVDv2. | ISME Communications

Fig. 2: Gene sharing network representation of all training and test viruses used in developing MArVDv2.

From: MArVD2: a machine learning enhanced tool to discriminate between archaeal and bacterial viruses in viral datasets

Fig. 2

All sequences used for the development and testing of MArVD2 are included in this network, created by vConTACT2. Reference viruses here include viruses from RefSeq v85 as well as the OcAVdb. Training viruses are those curated from the ETSP and VirSorter datasets as detailed in the text. Benchmarking viruses are those curated from the IMG/VR and GOV2.0 test dataset as detailed in the text. Viruses from the benchmarking datasets are further color coded as either predicted archaeal viruses or phages, from both MArVD and MArVD2. Network modules were grouped according to the inclusion of reference archaeal viruses (archaeal virus), reference phage (phage), or no reference viruses (unknown host).

Back to article page