*** Molecular models of SFV-chimpanzee Bet protein domains ***

generated by Daniel becker and Holger Gohlke, CPCLab, Heinrich Heine University Düsseldorf.

For details see Vasudevan et al., "Foamy Viruses, Bet, and APOBEC3 Restriction", Viruses 2021, in press.  

Domain boundaries within the sequence of the SFV-chimpanzee Bet protein sequence were predicted with TopDomain.
The two domains were modeled using the ROBETTA webserver. 
For each domain, ROBETTA generated five models. The best model was chosen based on the lowest Cα atom RMSD 
between the generated models and the corresponding ab initio modeled domain of PFV Bet. 
This consensus approach was also validated by TopScore in that domains with the lowest RMSD also have the 
lowest TopScore compared to the other four models. 
For the N-terminal domain, the overall TopScore is 0.47 and for the C-terminal one 0.45, indicating a moderate quality of the models. 
In the N-terminal domain, the helical part of the structure has low local TopScores, indicating good quality, 
but the N-terminal loops high local TopScores, indicating low quality. 
In the C-terminal domain, the β-strands have low local TopScores, indicating good quality, 
but helices and loops from residues 358–403 have high local TopScores, indicating low quality.