Fri Aug  6 10:03:24 PDT 2004
T0266

DUE 28 Aug 2004


Fri Aug  6 17:26:19 PDT 2004 Kevin Karplus

Comparative model with 1dbxA (d.116.1.1) as the template.

The rr constraints do not seem to useful on this target (not bad, but
we already have better info from the comparative model).

For try2, we just need to pack things more densely.
I'll run try2 from the initial alignments, but try3 will be a
polishing run using previous models as inputs.


Sun Aug  8 16:53:16 PDT 2004 Kevin Karplus

try2 looks good and scores a bit better than try1 with unconstrained.costfcn
I'll run try3 as a polishing run and we should be done.


Sun Aug  8 19:41:36 PDT 2004 Kevin Karplus

try3 also looks good, and scores slightly better than try2.

Rosetta likes try3-opt2.repack.nonPC better than
try2-opt2.repack-nonPC, but likes try1-opt2.repack-nonPC even better.

I'll submit
	try3-opt2
	try1-opt2
	try1-opt2.repack-nonPC
	T0266-1dbxA-t2k-local-str2+CB_burial_14_7-0.4+0.4-adpstyle5
	T0266-1vkiA-t2k-local-str2+CB_burial_14_7-0.4+0.4-adpstyle5
		
Thu Nov 18 23:48:10 PST 2004	Martina Koeva

Based on the smooth gdt scores:

best sam-t04	70.5322 (try2-opt2.repack-nonPC)
best submit	70.3496 (also model1)
model1		70.3496
auto		70.2028
align		70.0735
robetta best	71.3017 (robetta model3)
robetta1	70.4895