
PFRMAT AL
TARGET T0121
AUTHOR 1020-4390-8741
REMARK Prediction team GERLOFF consists of the following
REMARK team members: Dietlind L. Gerloff (1),
REMARK Melanie H. McCarthy-Troke (1), 
REMARK Zeti A. M. Hussein (1), Cairan Duffy (1),
REMARK Siu-wai Leung (2), Gina M. Cannarozzi (3).
REMARK (1): ICMB Biocomputing Research Unit & 
REMARK (2): Division of Informatics, University of Edinburgh;
REMARK (3): Computational Biochemistry Research Group, ETH Zurich
REMARK Contact email is Dietlind.Gerloff@ed.ac.uk
REMARK
REMARK Prediction for T0121.2 was by: DLG, MHMT.
METHOD To be added (this is a test submission)
METHOD 
METHOD Our submissions for T0121 contain TWO DIFFERENT MANUAL THREADING 
METHOD ALIGNMENTS (models 1 and 2) for the C-terminal 135 residues of the 
METHOD E.coli maltose transporter protein MalK onto a duplicated OB-fold 
METHOD template (1b9m_A). Besides this parent structure, we would also
METHOD consider it possible to use a minimal Ig-like beta sandwich-core
METHOD (see below for detailed description) but preferred to concentrate 
METHOD on the alignment issue with 1b9m_A as the preferred parent structure.  
METHOD 
METHOD Prediction Details:
METHOD 
METHOD Fold Recognition
METHOD 
METHOD Based on predicted secondary structure, solvent accessibility and
METHOD turn positions, we agreed with the automated FR-servers that both
METHOD 5-stranded (OB-fold-like) and 6-stranded (Ig-core-like) topologies
METHOD would have to be considered for this part of T0121. Since the function
METHOD of the domain was basically unknown, we decided to speculate based
METHOD on sequence conservation and literature (including limited information
METHOD about mutagenesis experiments, e.g. from Schmees, Schneider, Traxler,
METHOD Saier, and Boos groups), and to assess the feasibility of the
METHOD hypothesis for assembling the structure by (manual) combinatorial
METHOD analysis of beta strand arrangements. Homologous sequences for our
METHOD alignment were copied from the automated server prediction by SAMT99,
METHOD except for a few too divergent sequences that were difficult to align
METHOD (or might be false positives). 
METHOD 
METHOD While cases can be made both for OB-fold a 6-stranded Greek Key
METHOD topology (strand connections A-(+3)-B-(-1)-C-(-1)-D-(+3)-E-(+1)),
METHOD the better coverage of the 135 residue part by a duplicated
METHOD OB-fold (compared with a single OB-fold, or the 6-stranded sand-
METHOD wich above) compelled us to favor this possibility. The insertion
METHOD point of domain 2 into domain 1 for the SCOP-listed structure
METHOD of His-kinase was not satisfactory. However, the recent structure
METHOD 1b9m (which was ranked first in the FFAS-server prediction for
METHOD CAFASP2) fit quite well with our criteria.
METHOD 
METHOD Threading Alignment.
METHOD 
METHOD Two different alignments were generated. For the alignment in Model 2
METHOD the sequence-sequence alignment was used to guide the sequence-beta-
METHOD strand correspondence; detailed position alignment within the strands
METHOD was based on Surface/Interior prediction and visual inspection.
METHOD In Model 1, preference was given to a slightly better appearance of
METHOD the mutation results from the literature when mapped on the structure,
METHOD and consideration of minimal lengths for connecting segements in
METHOD homologs of the target (which is subject to multiple alignment accuracy,
METHOD of course). Further, the speculative purpose of the strongly conserved
METHOD GIRPED sequence, namely epitope function, probably with the interacting
METHOD IIIglc, seemed more appropriately placed in the barrel-capping helix.
METHOD The strong conservation of the entire sequence motif might indicate
METHOD that the conformation of the epitope is dominated by local interactions,
METHOD which would make it an interesting candidate for a pharmaceutical 
METHOD peptide mimicking the protein. However, considerable size insertions 
METHOD and deletions had to be made to accommodate the arrangement in Model 1,
METHOD e.g. the edge strand of the inserted OB-domain had to be shortened.
METHOD 
METHOD It will be interesting to see, provided the selected parent structure
METHOD is correct, which of the two presented alignments is, in fact, superior.
METHOD 
REMARK
MODEL  2
PARENT 1b9m_A
P 244 R 128
M 245 N 129
N 246 Q 130
F 247 W 131
L 248 F 132
D 249 G 133
A 250 T 134
I 251 I 135
V 252 T 136
T 253 A 137
G 256 Q 145
F 257 H 146
V 258 V 147
D 259 D 148
F 260 V 149
G 261 L 150
E 262 L 151
R 264 R 157
L 265 L 158
K 266 K 159
L 267 V 160
L 268 A 161
P 269 I 162
E 284 E 177
V 285 V 178
I 286 L 179
F 287 I 180
G 288 L 181
I 289 L 182
R 290 K 183
P 291 A 184
E 292 P 185
D 293 W 186
L 294 V 187
Y 295 G 188
D 296 I 189
A 297 T 190
M 298 Q 191
G 306 A 199
E 307 D 200
N 308 N 201
L 309 Q 202
V 310 L 203
R 311 P 204
A 312 G 205
V 313 I 206
V 314 I 207
E 315 S 208
I 316 H 209
V 317 I 210
E 318 E 211
N 319 R 212
L 320 G 213
G 321 A 214
S 322 E 215
E 323 Q 216
R 324 C 217
I 325 E 218
V 326 V 219
H 327 L 220
L 328 X 221
R 329 A 222
V 330 L 223
G 331 P 224
G 332 G 226
V 333 Q 227
T 334 T 228
F 335 L 229
V 336 C 230
G 337 A 231
A 338 T 232
F 339 V 233
R 340 P 234
S 341 V 235
R 344 S 240
V 345 L 241
R 346 Q 242
E 347 Q 243
G 348 G 244
V 349 Q 245
E 350 N 246
V 351 V 247
D 352 T 248
V 353 A 249
V 354 Y 250
F 355 F 251
D 356 N 252
M 357 A 253
K 358 D 254
K 359 S 255
I 360 V 256
H 361 I 257
I 362 I 258
F 363 A 259
D 364 T 260
K 365 L 261
TER
END


