PFRMAT SS
TARGET T0065
AUTHOR 1751-3146-3362
METHOD Using neural net fssp-12345-IDaa1-9-6-11-9-3-8-7-ehl-seeded4-trained.net
METHOD This is a 4-layer network, with 
METHOD	window	units
METHOD	    9	    6
METHOD	   11	    9
METHOD	    3	    8
METHOD	    7	    3 (1 EHL )
METHOD The input amino acid frequencies were determined from
METHOD alignment t65.t98_6.a2m.gz
METHOD with weighted counts, using HenikoffWeight(1 bits/column, 0.5)
METHOD The weighting was determined by the posterior distribution
METHOD after regularizing with recode3.20comp.
METHOD Counts were regularized to probabilities using
METHOD /projects/compbio/lib/recode3.20comp
METHOD Total sequence weight for alignment was 5.12005
METHOD
MODEL 1
M C 0.83
K C 0.88
N C 0.86
A C 0.81
K C 0.70
Q C 0.64
E C 0.57
H C 0.52
F C 0.54
E C 0.66
L C 0.55
D H 0.79
Q H 0.97
E H 0.98
W H 0.99
V H 0.99
E H 1.00
L H 1.00
M H 0.99
V H 0.99
E H 0.97
A H 0.92
K H 0.80
E H 0.52
A C 0.82
N C 0.83
I C 0.84
S C 0.89
P H 0.82
E H 0.97
E H 0.97
I H 0.98
R H 0.98
K H 0.99
Y H 0.98
L H 0.95
L H 0.89
L H 0.82
N H 0.63
K H 0.49
K C 0.63
S C 0.74
A C 0.84
H C 0.92
P C 0.90
G C 0.89
P C 0.67
A C 0.50
A C 0.50
R C 0.47
S C 0.65
H C 0.85
T C 0.72
V C 0.72
N C 0.89
P C 0.77
F C 0.69
END