Protein loops play important roles in intermolecular interactions, yet they are in general the least accurately predicted.
You can predict and build accurate loop structures in Flare with the help of FREAD, a loop prediction method based on database search.
FREAD searches a large database of entire protein structures (rather than just loop segments) to select loop fragments which have a similar amino acid sequence, and fit between geometrically comparable anchor residues on each side, to the loop you wish to model.
- Faster sampling speed and an increased number of loop predictions compared to ab initio-based loop prediction protocols
- Build loops faster within a reasonable computational time
- Choose between Automatic mode and Manual mode for building loops
- Automatic mode to fill all the gaps in selected proteins
- Manual mode to fill specific gaps
- Fine-tune results by adjusting
- The anchor geometry RMSD cutoff, i.e. the RMSD between the target and the predicted structures for two residues on either side of the loop
- The substitution score cut-off, controlling the loop sequence match span.
References and acknowledgements
Y. Choi, C. M. Deane, FREAD revisited: Accurate loop structure prediction using a database search algorithm, Proteins 2010, 78 (6), 1431-1440