🔬 AMP Candidates

646 computationally predicted antimicrobial peptide candidates

🧬 About these candidates

Each candidate was identified by mining extreme-environment metagenomes with ESM-2, a protein language model trained on 250M sequences. Candidates pass multi-stage filtering: biophysical scoring, novelty screening against 863K known AMPs (APD3 + DRAMP + AMPSphere), hemolysis risk prediction, and structural validation via AlphaFold2.

⚠️ All candidates are computationally predicted — no experimental validation has been performed.

🔒Amino acid sequences are not shown — sequences are proprietary research data. Interested in collaboration? Contact us.
🧊
#625

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8762

LengthPeptide length in amino acids26 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.07)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#626

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8754

LengthPeptide length in amino acids22 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.5
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.44)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#627

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8754

LengthPeptide length in amino acids22 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.5
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.44)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#628

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8754

LengthPeptide length in amino acids40 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.0
HydrophobicFraction of hydrophobic residues (0–1)45%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.37)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#629

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8752

LengthPeptide length in amino acids46 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.5
HydrophobicFraction of hydrophobic residues (0–1)52%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.37)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#630

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8749

LengthPeptide length in amino acids37 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)2.0
HydrophobicFraction of hydrophobic residues (0–1)49%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.54)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#631

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8741

LengthPeptide length in amino acids38 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)5.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.28)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#632

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8739

LengthPeptide length in amino acids42 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)4.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.29)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#633

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8739

LengthPeptide length in amino acids42 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)4.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.29)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#634

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8738

LengthPeptide length in amino acids50 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)4.5
HydrophobicFraction of hydrophobic residues (0–1)46%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.17)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#635

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8737

LengthPeptide length in amino acids39 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)5.5
HydrophobicFraction of hydrophobic residues (0–1)33%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.54)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#636

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8726

LengthPeptide length in amino acids24 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)3.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.53)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#637

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8722

LengthPeptide length in amino acids28 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)6.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.57)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#638

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8722

LengthPeptide length in amino acids28 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)6.0
HydrophobicFraction of hydrophobic residues (0–1)50%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.57)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#639

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8721

LengthPeptide length in amino acids45 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)8.0
HydrophobicFraction of hydrophobic residues (0–1)22%
PhyschemComposite physicochemical plausibility score3.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.36)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#640

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8710

LengthPeptide length in amino acids41 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)6.0
HydrophobicFraction of hydrophobic residues (0–1)32%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.79)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#641

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8708

LengthPeptide length in amino acids22 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)2.0
HydrophobicFraction of hydrophobic residues (0–1)55%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.31)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#642

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8708

LengthPeptide length in amino acids22 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)2.0
HydrophobicFraction of hydrophobic residues (0–1)55%
PhyschemComposite physicochemical plausibility score4.0
✅ Low hemolysisPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.31)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#643

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8705

LengthPeptide length in amino acids42 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)6.0
HydrophobicFraction of hydrophobic residues (0–1)40%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.73)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#644

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8705

LengthPeptide length in amino acids42 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)6.0
HydrophobicFraction of hydrophobic residues (0–1)40%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.73)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#645

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8701

LengthPeptide length in amino acids46 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)4.0
HydrophobicFraction of hydrophobic residues (0–1)54%
PhyschemComposite physicochemical plausibility score5.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.64)
0% max identityHighest sequence identity to any known AMP in databases
🧊
#646

🧊 Permafrost

database-novel
ESM-2 AMP ScoreProbability of antimicrobial activity (0–1). Higher = more likely AMP.

0.8701

LengthPeptide length in amino acids41 aa
Net ChargeCharge at pH 7. Most AMPs are cationic (+)2.5
HydrophobicFraction of hydrophobic residues (0–1)41%
PhyschemComposite physicochemical plausibility score4.0
⚠️ Hemolysis riskPredicted probability of red blood cell lysis. <0.5 = non-hemolytic(0.61)
0% max identityHighest sequence identity to any known AMP in databases
← PreviousShowing 625646 of 646

📚 Understanding the metrics

ESM-2 AMP Score: Probability (0–1) that the peptide has antimicrobial activity, predicted by Meta's ESM-2 protein language model.
Net Charge: Charge at physiological pH. Most effective AMPs carry a positive charge (+2 to +9).
Hemolysis Prob: SVM-predicted probability of red blood cell lysis (HemoPi3). Below 0.5 is considered non-hemolytic.
Novelty Tier: How similar to known AMPs. "database-novel" means <50% identity to any AMP in APD3 + DRAMP + AMPSphere.
Hydrophobic fraction: Proportion of hydrophobic amino acids. AMPs typically have 40–60% for membrane interaction.
Cross-biome: Found independently in multiple extreme environments — suggests evolutionary conservation of antimicrobial function.