TABLE 2.

Assignment of CheA and CheW protein Architectures to chemotaxis system categories

Protein	CheW-like	Chemotaxis	Evidence for Chemotaxis System Assignment
Architecture a	Domain Classb	System Categoryc	Figure 6 of 32	Figure 2	Figure 4
CheA.I	3	F7		+	+
CheA.I	3	F8			+
CheA.II	3	F1		+	+
CheA.III	3	ACF	+	+	+
CheA.III	3	F3	+d
CheA.IV	3	ACF	+	+	+
CheA.IV	3	F3	+d
CheA.V	3, 5	F5	e	+	+
CheA.VI	3, 5	F5	+	+	+
CheA.VII	4	F7.zf	e	+	+
CheA.VIII	3	F7.z	e	+	+
CheA.IX	3	Uncertain (F8 and ACF?)			+
CheA.X	4	Unassignedg	e
CheA.XI	3	Uncertain (F4 and F9?)			+
CheA.XII	3	F4	+	+	+
CheV.I	1	F6		+	+
CheV.I	1	F1			+
CheW.IA	6	F5, F6, F7.z			+
CheW.IB	1	F8		+
CheW.IB	1	F1, F7			+
CheW.IC	2	MAC1			+
CheW.II	1	ACF	e	+
CheW.II	1	F8	e		+
CheW.III	1	F9	+	+	+

^aOutlined in Figure 1. The CheW.I Architecture splits into three Classes.³⁶

^bFrom ³⁶. The two CheW-like domains in CheA.V and CheA.VI Architectures belong to different Classes.

^cOur sample does not contain enough representatives for analysis of system categories F3, F11-F17.

^dWuichet & Zhulin³² noted CheA proteins modified only by C-terminal receiver domains (CheA.III or CheA.IV) were consistently observed in F3 chemotaxis systems. However, our PhyDCA scores in Dataset S4 do not support linkage of either the CheA.III or CheA.IV Architectures to either F3 or F4 chemotaxis systems.

^eAlthough the CheA.V, CheA.VII, CheA.VIII, CheA.X, and CheW.II Architectures contain additional domains with respect to canonical CheA or CheW Architectures and are sufficiently abundant to be included in ³⁶, these Architectures were apparently not observed sufficiently consistently in specific chemotaxis system categories to be noted by Wuichet & Zhulin³², who analyzed a much smaller sample size.

^fThe subset of F7 chemotaxis systems that contain CheZ. See Figure 2.

^gDid not sort with a specific chemotaxis system category in Figure 2 and did not make sufficiently strong connections to be included in Figure 4.