PFRMAT SS
TARGET RECJ_ECOLI/382-451
AUTHOR 3670-4530-6947
METHOD Using neural net overrep-2500-IDaa13-5-10-7-10-5-10-11-ehl2-seeded9-stride-trained.net
METHOD This is a 4-layer network, with 
METHOD	window	units
METHOD	    5	   10
METHOD	    7	   10
METHOD	    5	   10
METHOD	   11	    3 (1 EHL2 )
METHOD The input amino acid frequencies were determined from
METHOD alignment pf02272-tuned.t2k.a2m.gz
METHOD with weighted counts, using EntropyWeight(1.4 bits/column, 10)
METHOD The weighting was determined by the posterior distribution
METHOD after regularizing with /projects/compbio/lib/recode3.20comp.
METHOD Counts were regularized to probabilities using
METHOD /projects/compbio/lib/recode3.20comp
METHOD Total sequence weight for alignment was 109.229
METHOD
MODEL 1
H C 0.978
R C 0.831
P C 0.552
V E 0.875
I E 0.980
A E 0.988
F E 0.978
A E 0.927
- E 0.778
P E 0.591
A C 0.778
G C 0.929
D C 0.957
G C 0.925
T E 0.732
L E 0.964
K E 0.990
G E 0.995
S E 0.992
G E 0.962
R E 0.783
S E 0.502
I C 0.727
Q C 0.777
G C 0.614
L H 0.471
H H 0.571
M H 0.606
R H 0.563
- C 0.876
- C 0.899
D C 0.873
A H 0.953
L H 0.985
E H 0.992
R H 0.988
L H 0.985
D H 0.980
T H 0.971
L H 0.935
Y H 0.917
P H 0.879
G H 0.818
M H 0.642
M C 0.887
L C 0.893
K C 0.903
F C 0.840
G C 0.811
G C 0.757
H C 0.667
A H 0.554
M H 0.571
A H 0.608
A H 0.567
G H 0.591
L C 0.627
S C 0.666
L C 0.837
E C 0.906
E C 0.923
- C 0.581
D C 0.727
K C 0.680
F H 0.982
K H 0.992
L H 0.995
F H 0.991
Q H 0.988
Q H 0.971
R H 0.956
F H 0.858
- H 0.544
- C 0.968
END