Peptide druggability workbench

Predict peptide properties from sequence

Sequence-derived physicochemical profiling, developability scoring, and early-stage risk flags for peptide candidates.

Input

Peptide sequences

Upload FASTA/TXT No file selected Peptide chain

Canonical amino acids only: A C D E F G H I K L M N P Q R S T V W Y. Sequences shorter than 3 aa are skipped. Up to 5000 sequences are accepted per run. FASTA headers from ProteinMPNN are parsed when present.

Output

Prediction results

ID	Sequence	Copies	MPNN score	Status	Score	Class	Hard filter	Length	MW (Da)	Charge	pI	Hydrophobicity	GRAVY	Instability	Boman	Aromaticity	Hydrophobic fraction	Ext. coeff.	Solubility	Aggregation	Proteolysis	Synthesis	Fail reasons
No predictions yet.

Model notes

Screening Thresholds

These thresholds are early-stage screening rules and do not represent experimental validation or clinical druggability conclusions. The references below provide high-impact review context for peptide developability, physicochemical design, stability, solubility, and amphipathic peptide features.

Scoring model

PepPredictor uses a two-tier framework: hard filters flag severe developability risks, then a weighted soft score ranks candidates while preserving all raw descriptors.

Physicochemical balance 25% Stability 25% Solubility / aggregation 25% Synthesizability 15% Safety proxy 10%

Length 5-50 aa

Typical short therapeutic-peptide range.

Net charge -1 to +6 at pH 7.4

Allows moderately cationic peptides while limiting very high charge.

Isoelectric point pI 4.0-11.0

Broad range used to avoid extreme ionization behavior.

Hydrophobicity -0.5 to +1.2

Eisenberg mean hydrophobicity range for balanced solubility.

Amphiphilicity ≥0.20

Eisenberg hydrophobic moment proxy for amphipathic character.

Instability index ≤40

ProtParam convention: values below 40 indicate predicted stability.

Aliphatic index 0-300

Reported descriptor; this range is intentionally non-restrictive.

GRAVY -2.0 to +0.8

Kyte-Doolittle hydropathy window to reduce extreme polarity or aggregation risk.

Boman index ≥0.5

Published binding-positive convention, calculated as the negative mean Boman scale value.

Shannon entropy ≥1.5

Low-complexity filter based on amino-acid composition.

Molecular weight Average mass, Da

Calculated from amino-acid average masses with peptide-bond water loss.

Residue class fractions Acidic, basic, charged, polar, hydrophobic, sulfur, proline

Composition descriptors for sequence polarity, charge distribution, and synthesis flags.

Aromaticity F + W + Y fraction

Fraction of aromatic residues, useful for hydrophobic and optical behavior.

Extinction coefficient Reduced / oxidized at 280 nm

Estimated from Trp, Tyr, and disulfide-forming Cys pairs.

Aggregation risk proxy Low / Medium / High

Sequence-only proxy combining hydrophobic load, low charge repulsion, aromaticity, and instability.

Degradation hotspots N/Q deamidation, D isomerization, oxidation, pyroglutamation

Regex scan for common chemical liabilities in peptide sequences.

Proteolysis risk proxy Longest clean stretch + cleavage-prone residues

Sequence-level estimate based on K/R and hydrophobic/aromatic cleavage-prone sites.

N-end rule class Stable / Intermediate / Destabilizing

Classifies the N-terminal residue as a stability proxy.

SPPS difficulty Favorable / Moderate / Difficult

Flags long sequences, hydrophobic stretches, beta-branched clusters, cysteine risk, and difficult motifs.

Hemolysis risk proxy Low / Medium / High

Heuristic based on high charge, amphiphilicity, positive GRAVY, and hydrophobic fraction.

Predict peptide properties from sequence

Peptide sequences

Prediction results

Screening Thresholds

Scoring model

References