Skip to main content

Table 12 LPM mispredictions in-sample, using 50% as the cutoff probability

From: Ridge regression estimated linear probability model predictions of O-glycosylation in proteins with structural and sequence data

UniProt Accession No.

UniProt Position

Protein Length

Sequence

Secondary Structure

Turn Type

Phi angle

Psi angle

ASA

LPM prediction

Proteins with LPM predictions < 50%, but that are deemed to be O-glycosylated in the data

 P02730

224

911

ILEKIPPDSEATLVLVG

Beta Turn

IV

-113

168

18.8

19.3%

 Q16566

57

473

ESELGRGATSIVYRCKQ

Beta Turn

I

-83

-35

69.9

34.6%

 P16157

794

1881

LKVVTDETSFVLVSDKH

Gamma Turn, Beta Turn

Inverse, IV

-54

11

94.7

36.3%

 P68431

11

136

RTKQTARKSTGGKAPRK

Loop

N/A

-81

145

44.4

36.3%

 P31749

308

480

IKDGATMKTFCGTPEYL

Loop

N/A

-115

-3

47.6

44.5%

 P04406

229

335

IPELNGKLTGMAFRVPT

Strand

N/A

-157

171

55.5

44.5%

Proteins with LPM predictions > 50%, but that are deemed to be N-glycosylated in the data

 P06756

488 SA

1048

SILNQDNKTCSLPGTAL

Gamma Turn, Beta Turn

Inverse, IV

-69

93

63.5

51.6%

 P04629

329

796

THVNNGNYTLLAANPFG

Beta Hairpin Strand

N/A

-93

119

30.7

62.7%

 Q92854

329

862

SAVCAYNLSTAEEVFSH

Beta Hairpin Strand

N/A

-76

124

35.5

70.1%