A fast way to discover new linear motifs

Welcome to LMD2 - the next generation of linear motif discovery. This tool takes protein sequences or interactions as input and determines the set of over-represented short sequence motifs present, scoring them according to over-representation, ortholog conservation, presense of motif-like residues, fit onto interacting partner surface (PepSite2) and how repeated the motif is within interacting partners. These scores are combined by a Bayesian integration approach, which was previously benchmarked on the ELM database to give estimates for False Positive Rate (FPR).
Sequences (fasta), a list of proteins* (one per line), or interacting protein* pairs * uniprot accessions (e.g. P04637), identifiers (e.g. P53_HUMAN) or gene names (e.g. TP53)
Data type:
Job name (optional):

Interacting protein, if known (fasta sequence, gene name, accession or identifier; not needed for interactions)
Species

Advanced options

 Sequence filtering
Sequence homology redundancy filtering: Yes No
Uniprot region filtering: Yes No
Ordered sequence (IUpred) filtering: Yes No
Structural domains (Blast) filtering: Yes No
Motif over-representation
Min # of sequences with motif:
Min # of fixed (non wildcard) residues in motifs:
Max motif length:
Max number of motifs to keep (per hub):
Max set size (interactions only, per hub):
Amino acid groups to be used (max 6):
Group motifs in output (blank means do not group):
Surface binding (PepSite)
Min % sequence identity for structure: