Table 4.

Examples of correctly and incorrectly predicted short segments in O20/D20 datasets

Dataset	PDB id	Protein name	Short segment	Location	ID status	SPA prediction (Avg. ID score)
D20	5EAU	5-EPI-aristolochene Synthase	MASAAVANYEEEIVRPVADF	1–20	D	D⋆(0.64)
	3GJY	Spermidine Synthase	SDTPQHPAETPEHSNTQP	300–317	D	D⋆(0.86)
	3FTO	Selenomethionine	MGSNQSSSTSTKKLKAG	1–17	D	D⋆(0.71)
	3C96	Flavin-containing Monooxygenase phzS	KTEKSAALEAITGSYRNQV	380–398	D	D⋆(0.49)
	2G5D	MltA from Neisseria gonorrhoeae Monoclinic	GSQSRSIQTFPQPDTSVING	1–20	D	D⋆(0.49)
	1ACC	Anthrax protective Antigen	HGNAEVHASFFDIGGS	304–319	D	S⋆(0.25)
	1B8X	Glutathione S-transferase	ATRYHTYLPPPYPGEFIVID	261–280	D	S⋆(0.23)
	1DYK	Laminin Alpha 2 Chain LG4–5 Domain PAIR	APLASVPTPAFPFPVPTMV	1–19	D	S⋆(0.29)
	1Q79	Poly(A) Polymerase	SHVLQKKKKHSTEGVK	499–514	D	S⋆(0.36)
	1SR8	Cobalamin Biosynthesis Protein	EDDMDSWVWDVQGTDH	283–298	D	S⋆(0.19)
O20	1HAR	HIV-1 Reverse Transcriptase N-terminal	WAKLVDFRELNKRTQDFWEV	71–90	S	S⋆(0.26)
	1B9D	HIV-1 integrase	GYSAGERIVDIIATDIQT	144–161	S	S⋆(0.41)
	1DHY	KKS102 BPHC Enzyme	WTVARHSRTAMWGHKSV	272–288	S	S⋆(0.44)
	2GA8	YFH7	EECTAVVARGGTANAIRIAA	119–138	S	S⋆(0.51)
	1HU3	Middle Domain of Human EIF4GII	NFRKLLLNRCQKEFEKDKA	89–107	S	S⋆(0.46)
	3E0C	DDB1	ALRPSASTQALSSSVS	751–766	S	D⋆(0.81)
	2QQH	C8alpha-MACPF	RKAMAVEDIISRVRGGSSG	250–268	S	D⋆(0.64)
	2IQC	Human FANCF	EDSLMKTQAELLLERLQEV	12–30	S	D⋆(0.69)
	1VZW	PRIA	SKLELLPAVDVRDGQAVR	2–19	S	D⋆(0.51)
	1PWA	Fibroblast Growth Factor 19	LPLSHFLPMLPMVPEEP	126–142	S	D⋆(0.58)

‘ID status’ is from PDB database, segments with missing electron density are identified as disordered ‘D’, while segments flanked by disordered regions are assigned as structured ‘S’; In SPA prediction, ‘D⋆’ refers to ratio of disordered residue equal to or higher than 50%; ‘S⋆’ is for ratio of predicted disordered residues less than 50%. Avg. ID score represents the mean intrinsic disorder propensity value estimated by SPA for a given fragment.