VirFinder: a novel k-mer based tool for identifying viral sequences from assembled metagenomic data

Microbiome

Table 2 Summary information for 15 viral contig bins associated with cirrhosis (+) or healthy (−) patients samples

Bin	Coefficients of association with cirrhosis^a	No. of contigs in bin	Total nucleotides in bin (bp)	No. of predicted proteins in bin	No. of contigs with significant blastn hit to nt ^b	Bin contains proteins with similarity to viral proteins^c
2	−0.04	46	82431	92	3	Y
6	0.06	88	295063	357	2	Y
35	0.00	1	1214	2	1	N
41	0.23	40	259266	360	15	Y
48	0.05	3	4940	5	0	N
51	−0.19	36	84134	112	6	Y
59	−0.10	68	184455	245	3	Y
64	−0.05	29	130154	148	1	Y
66	0.12	6	8500	7	5	N
69	0.00	1	1197	1	0	N
72	−0.05	29	77421	110	6	Y
78	−0.05	21	43329	48	1	Y
93	0.03	1	1295	1	0	N
106	−0.06	2	5243	7	0	N
127	0.01	18	72694	110	0	Y

^aCoefficients determined by the logistic regression with lasso regularization method for variable selection (see Methods)
^bContig had at least one blastn hit to NCBI’s non-redundant nucleotide database (nt) with an E value of ≤1e-10 and an alignment length of ≥100 bp
^cBin contains at least one protein for which its best blastp search results against NCBI’s non-redundant protein database (nr) was a viral protein or the protein had significant similarity to a viral Pfam domain (see Methods). Similarity requirements: E value of ≤1e-5, bit score ≥ 50

ISSN: 2049-2618