Table 2.

VvMTPs were modeled using Phyre2

VvMTPs (query)	Template proteins using for modeled
VvMTP4	c2qfiB, c3j1zP
VvMTP5	c2qfiB, c3j1zP
VvMTP7	c2qfiB, c3j1zP
VvMTP8	c2qfiB, c3j1zP
VvMTP9	c2qfiB, c3j1zP
VvMTP1	c2qfiB, c3j1zP, d2qfia2
VvMTP6	c2qfiB, c3j1zP, d1ng0a, c5ensA, d1c8na, c1ng0A
VvMTP8.1	c2qfiB, c3j1zP, c2enkA,
VvMTP9.1	c2qfiB, c3j1zP, c5gasN
VvMTP11	c2qfiB, c3j1zP, d1d4ua1

Models were predicted using the maximum coverage heuristics (MCH) of alignment between templates and query, c2qfiB (from the structure of the zinc transporter yiip) and c3j1zP (from the cation efflux protein transmembrane domain-like family), d2qfia2 (from the inward-facing conformation of the zinc transporter yiip by 2 cryo-electron microscopy), d1ng0a (from the cation efflux protein cytoplasmic domain-like), c5ensA (from the rhodamine bound structure of bacterial efflux pump), d1c8na (from the Tombusviridae-like VP family), c1ng0A (from the three-dimensional structure of cocksfoot mottle virus at 2.7a2 resolution), c2enkA (from the structure of a putative DNA-binding domain of the 2 human solute carrier family 30 zinc transporter protein), c5gasN (from the Thermus thermophilus v/a-ATPase, conformation 2), d1d4ua1 (from DNA repair factor XPA DNA- and RPA-binding domain, C-terminal subdomain)