Protein Domains & Motif Predictor
Predict transmembrane helices, hydropathy, signal peptides and common motifs from an amino acid sequence.
Learn more
What it does
The protein domains tool gives a quick heuristic scan of a sequence for transmembrane helices, hydropathy, signal peptides and common motifs. It is used for first-pass annotation of newly sequenced or designed proteins before deeper database searches.
How it works
The tool computes a Kyte-Doolittle hydropathy profile using a 19-residue sliding window, where strongly positive stretches indicate hydrophobic transmembrane regions. It also applies simple rules for an N-terminal signal peptide, hydrophobic-run coiled-coils, nuclear localization signals, glycosylation sites and RGD motifs. These are heuristic predictions meant for guidance, not definitive structural annotation.
Worked example
A membrane protein sequence with a long hydrophobic stretch of about 20 residues around positions 50 to 70 would show a high hydropathy peak and be flagged as a predicted transmembrane helix, while an N-terminal stretch of positively charged followed by hydrophobic residues may be reported as a signal peptide.
When to use it
Decide whether a protein is membrane-associated before doing a solubility or extraction test. Spot likely signal peptides that target secretion. Check for motifs like N-glycosylation or RGD that affect function or cell adhesion.
FAQ
- How does the tool predict transmembrane helices?
- It calculates a Kyte-Doolittle hydropathy profile over a 19-residue window and flags sustained hydrophobic stretches as candidate transmembrane helices.
- What is the Kyte-Doolittle hydropathy scale?
- It assigns each amino acid a value for hydrophobicity versus hydrophilicity; averaging over a window highlights membrane-spanning regions, which are typically hydrophobic.
- Can this tool find signal peptides?
- Yes, it applies a heuristic rule looking for an N-terminal pattern of positively charged residues followed by hydrophobic residues, a common signal-peptide signature.
- Are the domain predictions definitive?
- No, they are heuristic and for guidance only. Confirm with specialized databases such as Pfam or dedicated transmembrane prediction servers for research use.