Scaffold Structure	Receptor Structure	Complex Structure
Jmol._Canvas2D (Jmol) "jmolApplet0"[x]	Jmol._Canvas2D (Jmol) "jmolApplet1"[x]	Jmol._Canvas2D (Jmol) "jmolApplet2"[x]

Sequences generated during the Monte Carlo simulation are clustered. The top clusters are listed in the table below. Information in the table includes the relative cluster sizes as well as the number of the top sequences that originate from each cluster. Users are able to download every sequence in each cluster; the files include the sequences as well as the free-energy of each sequence predicted by EvoDesign.

Cluster Number	Relative Cluster Size	Download Each Sequence in Cluster
1	43%	Clus 1 Sequences
2	38.2%	Clus 2 Sequences
3	10.3%	Clus 3 Sequences
4	4.1%	Clus 4 Sequences

Design Number	Sequence Identity (%)[?]	Normalized Relative Error [?]
		Secondary Structure	Solvent Accessibility	Torsion Angle φ	Torsion Angle ψ
Design 1	36.8	-0.09	-0.05	0.14	-0.14
Design 2	39.5	-0.09	0.02	0.03	-0.24
Design 3	39.5	-0.09	0.00	0.19	0.21
Design 4	38.2	0.09	-0.02	-0.14	0.11
Design 5	38.2	-0.09	-0.06	0.03	0.18
Design 6	35.5	0.00	-0.02	0.01	0.13
Design 7	35.5	0.00	-0.08	0.25	-0.01
Design 8	38.2	0.00	0.02	0.14	-0.06
Design 9	34.2	0.00	-0.05	0.17	-0.10
Design 10	31.6	0.09	-0.05	0.34	-0.11
Data.zip	SI	SS	SA	φ	ψ

(a)	Sequence Identity: The percent sequence identity between the designed sequence and the scaffold sequence.
(b)	Normalized Relative Error (NRE): NRE=(EDS−ETS)/ETS, where EDS is the error of the neural-network predictions relative to the scaffold structure on the design sequence and ETS is the error of the predictions based on the sequence of the target scaffold. The secondary structure, solvent accessibility and torsion angles for the scaffold structure are assigned by DSSP.
(c)	Secondary Structure (SS): SS is predicted by PSSPred for the scaffold and design sequences. The Q3 errors of the design sequence (EDS) and scaffold sequence (ETS) with respect to the scaffold structure are used to calculate the NRE for SS.
(d)	Solvent Accessibility (SA): SA for scaffold and design sequences are predicted by neural-network method. The correlation on SA between design sequence and scaffold structure (EDS) and between scaffold sequence and scaffold structure (ETS) is used to calculate NRE on SA.
(e)	Torsion Angle (TA): TA is predicted by ANGLOR for scaffold and design sequences. The mean absolute difference of the design sequence (EDS) and scaffold sequence (ETS) from the scaffold structure is used to calculate NRE for TA.

Energy:         EvoDesign calculated energy
Iden.:          % sequence identity between the design and scaffold sequences.
DSSP_SS:        Secondary structure as assigned by DSSP on the scaffold structure.
Scaffold:       Scaffold sequence.
                Identical residues in the scaffold and design sequences are marked by '|'.
Design:         Design sequence.