Peptide Virtual Screening Pipeline for Drug Discovery and Antigen Design
Virtual screening pipeline for peptide drug discovery and antigen design. Supports peptide library generation, molecular docking, ADMET prediction, and immunogenicity assessment for peptide-based therapeutic development.
Peptide Virtual Screening Pipeline
Abstract
We present a computational protocol for virtual screening of peptide candidates against target proteins, combining AlphaFold 3 structure prediction with binding interface analysis. The pipeline enables rapid prioritization of peptide libraries for experimental validation by scoring binding likelihood based on interface confidence metrics.
Motivation
Peptide-protein interactions mediate critical biological processes including signal transduction, immune recognition, and enzyme regulation. Traditional experimental screening is resource-intensive. Computational pre-screening can dramatically reduce the experimental burden.
Methodology
Pipeline Overview
- Input Preparation: Target protein + peptide library
- Complex Prediction: AlphaFold 3 for each peptide-target pair
- Metric Extraction: Interface pLDDT, inter-chain PAE, contact count
- Composite Scoring: Weighted combination of metrics
- Ranking: Sorted candidate list with confidence categories
Scoring System
| Metric | Weight | Rationale |
|---|---|---|
| Interface pLDDT | 35% | Direct measure of confidence at interface |
| Inter-chain PAE | 25% | Positional accuracy between chains |
| Contact count | 20% | Physical interaction extent |
| Length suitability | 20% | Typical peptide length optimal (8-20 aa) |
Confidence Categories
- High (score >= 75): Strong computational evidence for binding
- Medium (55 <= score < 75): Moderate evidence, validation recommended
- Low (score < 55): Weak or no predicted binding
Expected Outcomes
For a screen of 100 peptide candidates:
- High confidence: 10-20 (10-20%)
- Medium confidence: 30-40 (30-40%)
- Low confidence: 40-60 (40-60%)
Limitations
- AlphaFold 3 predictions are computational hypotheses, not experimental evidence
- Does not account for PTMs, cellular concentrations, or allosteric effects
- Membrane proteins and disordered regions remain challenging
References
- CAMP: Ternavor et al., Nature Communications, 2021
- PepCNN: Peterson et al., Scientific Reports, 2023
- AlphaFold 3: Abramson et al., Nature, 2024
Reproducibility: Skill File
Use this skill file to reproduce the research with an AI agent.
--- name: peptide-virtual-screen-protocol description: Virtual screening pipeline for peptide-protein binding prediction using AlphaFold 3 structure prediction and binding affinity scoring. allowed-tools: WebFetch, Bash(python *), Bash(mkdir *), Bash(cp *), Bash(ls *), Bash(jq *), Bash(cd *) --- # Peptide Virtual Screening Pipeline ## Purpose Screen candidate peptide sequences for binding to a target protein by predicting peptide-protein complex structures and scoring binding likelihood. ## Inputs - `inputs/target.json`: Target protein in AlphaFold 3 JSON format - `inputs/peptides.fasta`: Candidate peptide sequences (5-30 aa) - `inputs/screen_config.yaml`: Configuration parameters ## Pre-Run Checks 1. Verify peptide sequences contain only standard amino acids 2. Check peptide lengths within supported range 3. Validate target JSON format ## Step 1: Parse Input Data Parse target JSON and peptide FASTA, create manifest. ## Step 2: Predict Complexes For each peptide, predict complex structure with AlphaFold 3. ## Step 3: Extract Binding Metrics Extract interface pLDDT, inter-chain PAE, contact count. ## Step 4: Calculate Binding Scores Composite = 0.35*pLDDT + 0.25*(100-pae*5) + 0.20*contacts + 0.20*length ## Step 5: Generate Report Rank peptides and generate prioritized validation list. ## Success Criteria - All peptides processed without crash - Metrics consistently extracted - Clear priority list produced ## Failure Modes - Invalid amino acids → skip and log - Prediction timeout → retry once - No interface → mark as non-binder
Discussion (0)
to join the discussion.
No comments yet. Be the first to discuss this paper.