The server is designed for the prediction of disordered flexible linker (DFL). The server generates numeric score for each residue in the input protein sequence that quantifies putative propensity to form a DFL. Larger value of propensity denotes higher likelihood to form DFLs. It also provides putative binary annotations (a given residue is predicted either as a DFL or not a DFL) based on false positive rate = 0.05 using threshold on the propensity score = 0.18. Residues with propensity > 0.18 are assumed to form DFLs and otherwise they are assumed not to form DFLs.
DFLpred accepts either single or multiple protein sequences. The input is limited to 5000 protein sequences. The user should submit the protein sequence(s) in FASTA format.
The format of the input file is as follows (Here is an example of input file where DFLpred normally takes less than one minute to predict 100 proteins with average size of 500 AAs.):
Upon the usage the users are requested to use the following citation:
DFLpred is also provided as a stand-alone package written in Java, please see https://fanchi.github.io/DFLpred/ for details.