PFRMAT SS
TARGET T0639
AUTHOR SAM-T02-server 
METHOD This file is the result of combining several RDB files, specifically
METHOD T0639.t2k.dssp-ebghstl.rdb  (weight 1.53986)
METHOD T0639.t2k.stride-ebghtl.rdb  (weight 1.24869)
METHOD T0639.t2k.str.rdb  (weight 1.53983)
METHOD These files were combined by translating their predictions into EHL
METHOD predictions with tables generated by compare-real, and then combining
METHOD those predictions with weights proportional to their mutual information
METHOD with the EHL alphabet.  The comments from the individual files follow.
METHOD
METHOD Comments from T0639.t2k.dssp-ebghstl.rdb
METHOD ============================================
METHOD TARGET T0639
METHOD Using neural net t2k-5740-IDaaHr-5-15-7-15-9-15-13-ebghstl-seeded.net
METHOD This is a 4-layer network, with 
METHOD	window	units
METHOD	    5	   15
METHOD	    7	   15
METHOD	    9	   15
METHOD	   13	    7 (1 EBGHSTL )
METHOD The input amino acid frequencies were determined from
METHOD alignment T0639.t2k-thin90.a2m.gz
METHOD with weighted counts, using HenikoffWeight(-1 bits/column, 1)
METHOD Counts were regularized to probabilities using
METHOD /projects/compbio/lib/recode3.20comp
METHOD Total sequence weight for alignment was 2
METHOD
METHOD ============================================
METHOD Comments from T0639.t2k.stride-ebghtl.rdb
METHOD ============================================
METHOD TARGET T0639
METHOD Using neural net t2k-5651-IDaaHr-5-15-7-15-9-15-13-ebghtl-seeded.net
METHOD This is a 4-layer network, with 
METHOD	window	units
METHOD	    5	   15
METHOD	    7	   15
METHOD	    9	   15
METHOD	   13	    6 (1 EBGHTL )
METHOD The input amino acid frequencies were determined from
METHOD alignment T0639.t2k-thin90.a2m.gz
METHOD with weighted counts, using HenikoffWeight(-1 bits/column, 1)
METHOD Counts were regularized to probabilities using
METHOD /projects/compbio/lib/recode3.20comp
METHOD Total sequence weight for alignment was 2
METHOD
METHOD ============================================
METHOD Comments from T0639.t2k.str.rdb
METHOD ============================================
METHOD TARGET T0639
METHOD Using neural net t2k-5651-IDaaHr-5-15-7-15-9-15-13-str-seeded.net
METHOD This is a 4-layer network, with 
METHOD	window	units
METHOD	    5	   15
METHOD	    7	   15
METHOD	    9	   15
METHOD	   13	   13 (1 str )
METHOD The input amino acid frequencies were determined from
METHOD alignment T0639.t2k-thin90.a2m.gz
METHOD with weighted counts, using HenikoffWeight(-1 bits/column, 1)
METHOD Counts were regularized to probabilities using
METHOD /projects/compbio/lib/recode3.20comp
METHOD Total sequence weight for alignment was 2
METHOD
METHOD ============================================
MODEL 1
S C 0.88
N C 0.60
A H 0.55
M H 0.65
E H 0.64
T H 0.67
L H 0.86
N H 0.92
D H 0.94
I H 0.96
K H 0.97
K H 0.97
I H 0.97
L H 0.96
I H 0.95
N H 0.90
V H 0.73
G C 0.71
L C 0.63
Y C 0.55
Q C 0.66
G C 0.77
F C 0.63
D C 0.62
L C 0.64
T C 0.71
D C 0.76
P C 0.71
K C 0.64
V C 0.54
S C 0.55
E C 0.45
E C 0.47
V C 0.59
N C 0.75
H C 0.77
E C 0.77
T C 0.76
A C 0.71
N C 0.69
M H 0.42
K H 0.44
W H 0.48
I H 0.49
K H 0.46
D C 0.43
Y C 0.58
T C 0.76
S C 0.83
D C 0.89
G C 0.85
N C 0.83
W C 0.62
D C 0.52
N H 0.73
E H 0.82
F H 0.86
K H 0.88
E H 0.89
D H 0.87
L H 0.94
K H 0.95
N H 0.93
F H 0.95
L H 0.95
D H 0.92
Y H 0.90
M H 0.90
E H 0.91
V H 0.92
C H 0.90
Q H 0.85
L H 0.79
A H 0.70
L H 0.47
N C 0.69
D C 0.57
K H 0.54
N H 0.56
F H 0.73
K H 0.78
I H 0.78
A H 0.77
S H 0.78
N H 0.78
S H 0.78
L H 0.87
F H 0.90
M H 0.93
A H 0.93
M H 0.92
I H 0.90
Y H 0.87
A H 0.78
G H 0.54
N C 0.50
L H 0.91
S H 0.94
L H 0.95
I H 0.95
F H 0.95
D H 0.94
S H 0.91
I H 0.90
K H 0.89
T H 0.87
D H 0.86
I H 0.94
S H 0.95
T H 0.94
L H 0.91
L H 0.83
S H 0.72
A H 0.63
E H 0.57
Y H 0.46
K C 0.56
K C 0.67
N C 0.77
S C 0.80
F C 0.75
S C 0.77
W C 0.79
P C 0.75
S C 0.70
L C 0.78
D C 0.89
E C 0.97
END