PFRMAT AL 
TARGET T0044 
AUTHOR 9070-5088-8627 
REMARK  
REMARK Prediction date: Monday June 22, 1998 
REMARK Group name: UCSC-compbio 
REMARK Authors: Christian Barrett, Melissa Cline, Mark Diekens, Kevin Karplus, 
REMARK 	 David Haussler and Richard Hughey 
REMARK University of California, Santa Cruz 
REMARK  
METHOD Overview 
METHOD  
METHOD Fold recognition was performed using the Target98 (SAM-T98) method 
METHOD [3] using SAM version 2.1.1 [1], a refinement of the methods developed 
METHOD by this group for CASP2 [2].  This method attempts to find and multiply  
METHOD align a set of homologs to a given sequence, then create an HMM from that  
METHOD multiple alignment. 
METHOD  
METHOD First, a set of sequence weights is determined from the alignment.  Next,  
METHOD Modelfromalign is used to build the model from the alignment and the  
METHOD sequence weights.  Finally, hmmscore performs a local, all-paths scoring  
METHOD of the sequences, using a reversed-sequence normalization feature. 
METHOD  
METHOD The weighting method, detailed in upcoming publications [3,4], 
METHOD combines the Henikoffs' scheme [5], Dirichlet mixtures [6], and an 
METHOD entropy method to set the final weights. 
METHOD  
METHOD Alignment generation 
METHOD  
METHOD The initial step uses BLASTP to search NRP twice: once to produce a set 
METHOD of very close homologs, and once to produce a set of possible homologs. 
METHOD  
METHOD The method then uses multiple iterations of a selection, training, and  
METHOD alignment procedure.  Each iteration involves an initial alignment, a set  
METHOD of search sequences, a threshold value, and a transition regularizer.  
METHOD  
METHOD The first iteration uses a single sequence (or seed alignment) as the  
METHOD initial alignment and the close homologs found by BLASTP are used as the  
METHOD search set.  The threshold is set very strictly, so that only good matches  
METHOD to the sequence are considered.  This iteration uses a transition regularizer  
METHOD that was designed to match the gap costs used by BLASTP. 
METHOD  
METHOD On subsequent iterations the input alignment is the output from the 
METHOD previous iteration, the search set is the larger set of possible 
METHOD homologs found by BLASTP, and the thresholds are gradually loosened. 
METHOD The second through second-from-last iteration use a ``long-match'' 
METHOD transition regularizer, and the final iteration uses a transition regularizer  
METHOD trained on FSSP alignments. 
METHOD  
METHOD References 
METHOD [1] R. Hughey and A. Krogh, CABIOS 12(2): 95-107, 1996. 
METHOD     http://www.cse.ucsc.edu/research/compbio/sam.html.   
METHOD [2] K. Karplus, K. Sjolander, C. Barrett, M. Cline, D. Haussler, R. 
METHOD     Hughey, L. Holm, and C. Sander, Proteins: Structure, Function, and  
METHOD     Genetics, Suppl. 1, 134-9, 1997. 
METHOD [3] K. Karplus, C. Barrett, and R. Hughey, Technical Report UCSC-CRL-98-06, 
METHOD     Department of Computer Engineering, Univ. of California, Santa Cruz, 1998. 
METHOD [4] J. Park, K. Karplus, C. Barrett, R. Hughey, D. Haussler, T. Hubbard, 
METHOD     and C. Chothia, http://cyrah.med.harvard.edu/~jong/assess_final.html, 1998. 
METHOD [5] S. Henikoff and J. C. Henikoff, JMB, vol 243, pp 574-578, Nov 1994. 
METHOD [6] K. Sjolander, K. Karplus, M. P. Brown, R. Hughey, A. Krogh, I. S. 
METHOD    Mian, and D. Haussler, CABIOS 12(4):327-345, 1996. 
METHOD  
METHOD  
METHOD The fold-prediction (1EPS) was fairly simple, since 1EPS scored very 
METHOD well in both direction.  The sum of the scores (-12.84) is in the 
METHOD range where only about 20% of the hits are false positives, so we 
METHOD would have been fairly confident even without the hint from the 
METHOD contributors that the fold was an already known one. 
METHOD  
METHOD This is an important target for us, as it is easy to find with our 
METHOD methods, but wu-blast and double-blast were unable to find it. 
METHOD Very few of the targets this year fell into the grey zone between the 
METHOD trivial and the impossible. 
METHOD  
METHOD The alignment turned out to be quite difficult for us.  Different 
METHOD variations gave very different alignments, and we had to use a lot of 
METHOD visual inspection in 3D to select between them. 
METHOD  
METHOD Essentially all the methods agreed on one core region: 
METHOD  
METHOD T0044	VLQTVLPALWFADGPSRVEV 
METHOD 1EPS	FLTALLMTAPLAPEDTVIRI 
METHOD  
METHOD and most had fairly similar alignments for the first half of T0044, 
METHOD but the alignment to the second domain was much less determined.  We 
METHOD finally picked this alignment, for two reasons---it had the most in 
METHOD common with other alignments and the gaps it had looked like 
METHOD they could reasonably be bridged with only minor movement of the 
METHOD backbone.  We did some very minor tweaking of the alignment, shifting 
METHOD or unaligning residues to match the better parts of other alignments. 
MODEL 1 
PARENT 1eps 
V 2 V 12 
K 3 D 13 
R 4 G 14 
M 5 T 15 
I 6 I 16 
A 7 N 17 
L 8 L 18 
D 9 P 19 
G 10 G 20 
A 11 S 21 
Q 12 K 22 
G 13 T 23 
E 14 V 24 
G 15 S 25 
G 16 N 26 
G 17 R 27 
Q 18 A 28 
I 19 L 29 
L 20 L 30 
R 21 L 31 
S 22 A 32 
A 23 A 33 
L 24 L 34 
S 25 A 35 
L 26 M 99 
S 27 R 100 
M 28 P 101 
I 29 L 102 
T 30 A 103 
G 31 A 104 
Q 32 A 105 
P 33 L 106 
F 34 C 107 
T 35 L 108 
I 36 G 109 
T 37 S 110 
S 38 N 111 
I 39 D 112 
R 40 I 113 
A 41 V 114 
G 42 L 115 
R 43 T 116 
A 44 G 117 
K 45 E 118 
P 46 P 119 
G 47 R 120 
L 48 M 121 
L 49 K 122 
R 50 E 123 
Q 51 R 124 
H 52 P 125 
L 53 I 126 
T 54 G 127 
A 55 H 128 
V 56 L 129 
K 57 V 130 
A 58 D 131 
A 59 A 132 
T 60 L 133 
E 61 R 134 
I 62 L 135 
C 63 G 136 
G 64 G 137 
A 65 A 138 
T 66 K 139 
V 67 I 140 
E 68 T 141 
G 69 Y 142 
A 70 L 143 
E 71 E 144 
L 72 Q 145 
G 73 E 146 
S 74 N 147 
Q 75 Y 148 
R 76 P 149 
L 77 P 150 
L 78 L 151 
F 79 R 152 
R 80 L 153 
P 81 Q 154 
G 82 G 155 
T 83 G 156 
V 84 F 157 
R 85 T 158 
G 86 G 159 
G 87 G 160 
D 88 N 161 
Y 89 V 162 
R 90 D 163 
F 91 V 164 
A 92 D 165 
G 94 G 166 
S 95 S 167 
A 96 V 168 
G 97 S 169 
S 98 S 170 
C 99 Q 171 
V 102 F 172 
L 103 L 173 
Q 104 T 174 
T 105 A 175 
V 106 L 176 
L 107 L 177 
P 108 M 178 
A 109 T 179 
L 110 A 180 
W 111 P 181 
F 112 L 182 
A 113 A 183 
D 114 P 184 
G 115 E 185 
P 116 D 186 
S 117 T 187 
R 118 V 188 
V 119 I 189 
E 120 R 190 
V 121 I 191 
S 122 K 192 
T 125 G 193 
D 126 D 194 
N 127 L 195 
P 128 V 196 
S 129 S 197 
A 130 K 198 
P 131 P 199 
P 132 Y 200 
A 133 I 201 
D 134 D 202 
F 135 I 203 
I 136 T 204 
R 137 L 205 
R 138 N 206 
V 139 L 207 
L 140 M 208 
E 141 K 209 
P 142 T 210 
L 143 F 211 
L 144 G 212 
A 145 V 213 
K 146 E 214 
I 147 I 215 
G 148 E 216 
I 149 N 217 
H 150 Q 218 
Q 151 H 219 
Q 152 Y 220 
T 153 Q 221 
T 154 Q 222 
L 155 F 223 
L 156 V 224 
R 157 V 225 
H 158 K 226 
G 159 G 227 
F 160 Y 231 
Y 161 Q 232 
P 162 S 233 
A 163 P 234 
G 164 G 235 
G 165 T 236 
G 166 Y 237 
V 167 L 238 
V 168 V 239 
A 169 E 240 
T 170 G 241 
E 171 D 242 
V 172 A 243 
S 173 S 244 
P 174 S 245 
V 175 A 246 
A 176 S 247 
S 177 Y 248 
F 178 F 249 
N 179 L 250 
T 180 A 251 
L 181 A 252 
Q 182 A 253 
L 183 A 254 
G 184 I 255 
E 185 K 256 
R 186 G 257 
G 187 G 258 
N 188 T 259 
I 189 V 260 
V 190 K 261 
Q 191 S 269 
M 192 M 270 
R 193 Q 271 
G 194 G 272 
E 195 D 273 
V 196 I 274 
L 197 R 275 
L 198 F 276 
A 199 A 277 
G 200 D 278 
V 201 V 279 
P 202 L 280 
A 206 G 299 
E 207 E 300 
R 208 L 301 
E 209 N 302 
I 210 A 303 
A 211 I 304 
T 212 D 305 
L 213 M 306 
A 214 D 307 
G 215 M 308 
S 216 N 309 
F 217 H 310 
S 218 I 311 
L 219 P 312 
H 220 D 313 
E 221 A 314 
Q 222 A 315 
N 223 M 316 
I 224 T 317 
H 225 I 318 
N 226 A 319 
L 227 T 320 
P 228 A 321 
R 229 A 322 
D 230 L 323 
Q 231 F 324 
G 232 A 325 
P 233 K 326 
G 234 G 327 
N 235 T 328 
T 236 T 329 
V 237 R 330 
S 238 L 331 
L 239 R 332 
E 240 N 333 
V 241 I 334 
E 242 Y 335 
S 243 N 336 
E 244 W 337 
N 245 R 338 
I 246 V 339 
T 247 K 340 
E 248 E 341 
R 249 F 346 
F 250 A 347 
F 251 M 348 
V 252 A 349 
V 253 T 350 
G 254 E 351 
E 255 L 352 
K 256 R 353 
R 257 K 354 
V 258 V 355 
S 259 G 356 
A 260 A 357 
E 261 E 358 
V 262 V 359 
V 263 E 360 
A 264 E 361 
A 265 G 362 
Q 266 H 363 
L 267 D 364 
V 268 Y 365 
K 269 I 366 
E 270 R 367 
V 271 I 368 
K 272 T 369 
R 273 P 370 
Y 274 P 371 
L 275 E 372 
A 276 L 374 
S 277 N 375 
T 278 F 376 
A 279 A 377 
A 280 E 378 
V 281 I 379 
G 282 A 380 
E 283 T 381 
Y 284 Y 382 
L 285 N 383 
A 286 D 384 
D 287 H 385 
Q 288 R 386 
L 289 M 387 
V 290 A 388 
L 291 M 389 
P 292 C 390 
M 293 F 391 
A 294 S 392 
L 295 L 393 
A 296 V 394 
G 297 A 395 
A 298 L 396 
G 299 T 399 
E 300 P 400 
F 301 V 401 
T 302 T 402 
V 303 I 403 
A 304 L 404 
H 305 D 405 
P 306 P 406 
S 307 K 407 
C 308 C 408 
H 309 T 409 
L 310 A 410 
L 311 K 411 
T 312 T 412 
N 313 F 413 
I 314 P 414 
A 315 D 415 
V 316 Y 416 
V 317 F 417 
E 318 E 418 
R 319 Q 419 
F 320 L 420 
L 321 A 421 
P 322 R 422 
V 323 I 423 
R 324 S 424 
F 325 Q 425 
S 326 A 426 
L 327 A 427 
TER 
END 
