Table 3. Predicted molecular functions for the HEV PPR model.

Rank	TM-Score	RMSD	Identity	Coverage	PDB homolog	Molecular function GO terms
1	0.5528	3.35	0.04	0.95	3eqnA	0004338, 0016787, 0016798, 0008152
2	0.5385	3.62	0.15	0.95	3l3sB	0016853, 0003824, 0008152
3	0.5374	3.37	0.04	0.93	2iq7A	0008152, 0007047, 0004650, 0016787, 0016798, 0005975
4	0.5373	3.64	0.03	0.91	1wmrA	0051675, 0008152, 0016798, 0005576, 0016787, 0004553
5	0.533	3.5	0.01	0.93	1k5cA	0004650, 0005975
6	0.5316	3.41	0.13	0.91	1k5dC	0000070, 0005098, 0006913, 0005096, 0048471, 0051383, 0043547, 0031965, 0005829, 0030702, 0031291, 0032853, 0005737
7	0.5305	3.47	0.05	0.91	1nhcA	0004650, 0005576, 0016787, 0016798, 0007047, 0008152, 0005975
8	0.5289	3.59	0.05	0.93	3p85A	0003824, 0008152
9	0.5285	3.62	0.11	0.93	1vrgA	0046872, 0016874
10	0.5276	3.4	0.08	0.9	3gf7A	0016874

TM-score is a measure of global structural similarity between query and template protein. RMSD is root mean standard deviation between residues that are structurally aligned by TM-align [66]. Identity is percentage sequence identity in the structurally aligned region. Coverage represents coverage of global structural alignment and is equal to the number of structurally aligned residues divided by length of the query protein. PDB analog is the PDB-matched template or functional analog, from which associated GO terms was used to predict function of query sequence.