To identify more distantly related BMV-like sequences, further searches were performed against translated nucleotide sequences of the metagenomic databases using BMV2 protein sequences as queries.
Hallam et al. [ 17], used a composite genome sequence of Cenarchaeum symbiosum to study the diversity of Marine Group I archaeal genomes in metagenomic databases using BLAST score ratios [ 18] to identify conserved genes, and a similar approach was employed by Coleman to study genome variation in Prochlorococcus [ 19].
Metagenomic datasets were aligned to their corresponding reference genome databases using megablast in BLAST (version 2.2.26) with parameters (–m 7 –v 100 –b 100).
Using the checkfile generated above, metagenomic databases listed in the database file were then searched for homologs using PSI-TBLASTN, with no specified e-value cutoff with an effective database length set to 109.
Here we used sequences from metagenomic databases, Marseillevirus and three new viruses extending the Mimiviridae family to generate the phylogenetic trees of eight proteins involved in different steps of DNA processing.
The resulting peptide mass fingerprints were searched against the three metagenomic databases (OZ sludge, Phrap assembly; US sludge, Phrap assembly; US sludge, Jazz assembly) [4] using the MASCOT search tool (http://www.matrixscience.com).com
All PgVV-like MCP proteins were used as queries for TBLASTN searches against two metagenomic databases, CAMERA and marine metagenomes.
The sequence of the putative major capsid protein (MCP) of the unusual linear virophage associated with Phaeocystis globosa virus (PgVV) was used as a bait to identify potential related viruses in metagenomic databases.
PgVV MCP and its eukaryotic homologs found in the nr database were used as queries for TBLASTN searches (e-value ≤10) against two metagenomic databases, CAMERA [ 48] and the NCBI Whole Genome Shotgun (WGS) contigs database [ 49].
Using this tree as a guide, 20 diverse representatives were selected, and the metagenomic databases were screened once again, as described above.
As the full RefSeq database is 3,717,469,431 nucleotides and the combined metagenomic databases total 5,529,658,033 nucleotides, several subset databases (Proteobacteria, Alphaproteobacteria, Bacteroidetes, Additional File 2 and Global Ocean Survey Scaffolds) were used to reduce the number of false positive hits.
