PCSK9 Pharmacophore Mining in Ayurvedic and TCM Databases

ISEF Category: Biochemistry

Subcategory: Medicinal Biochemistry  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

You can think of PCSK9 as a cholesterol traffic controller. When it overacts, LDL receptors get removed faster, and LDL cholesterol can rise. That is why PCSK9 keeps showing up in heart-drug research. Your project can search Ayurvedic and TCM compound banks for molecules that look like better fits for this target.

What Is It?

A pharmacophore is the pattern of features a molecule needs to bind a target, like a shape with the right charge, sticky spots, and spacing. You do not start by testing every compound at random. You start by defining that feature pattern, then you search a database for compounds that match it.

IMPPAT and TCMSP collect compounds linked to Ayurvedic and traditional Chinese medicine sources. Docking then checks how each candidate fits into PCSK9, like trying keys in a lock. Molecular dynamics, or MD, follows the protein and ligand over time to see whether that fit stays steady or falls apart when everything starts moving.

Why This Is a Good Topic

This is a strong science fair topic because you can measure real differences between compounds, not just describe them. You connect your work to cholesterol control, drug discovery, and natural-product chemistry, which gives the project a clear medical angle. You can also learn how to clean chemical data, compare scoring methods, and judge whether a predicted hit is actually stable.

Research Questions

• How does pharmacophore filtering change the number of IMPPAT and TCMSP hits that survive PCSK9 docking?
• What is the effect of database source, IMPPAT versus TCMSP, on average docking score and interaction pattern?
• Does adding PAINS and reactivity filters change which scaffolds remain top ranked after docking?
• To what extent do top docking hits keep their key contacts during molecular dynamics runs?
• Which scaffold families from the two databases produce the best consensus score across docking and MD stability?
• How does using multiple protein conformations change the ranking of candidate PCSK9 binders?

Basic Materials

• Laptop with at least sixteen gigabytes of RAM.
• Stable internet connection for database searches.
• Spreadsheet software for tracking candidates and scores.
• Python or R for cleaning compound lists and plotting results.
• AutoDock Vina for docking calculations.
• UCSF ChimeraX or PyMOL for viewing binding poses.
• PubChem, RCSB PDB, IMPPAT, and TCMSP access for structures and annotations.

Advanced Materials

• Linux workstation or university compute cluster with enough memory for repeat docking.
• GPU-enabled or multi-core access for molecular dynamics runs.
• GROMACS or AMBER for trajectory generation and analysis.
• Protein-ligand preparation workflow for consistent structure cleanup.
• Optional binding assay access in a university biochemistry lab for follow-up validation.

Software & Tools

• PubChem: Pull compound structures, synonyms, and bioassay notes for candidate scaffolds.
• RCSB PDB: Download PCSK9 structures and compare bound ligands.
• AutoDock Vina: Score candidate compounds against PCSK9 binding sites.
• GROMACS: Run molecular dynamics simulations and track stability metrics.
• UCSF ChimeraX: Inspect binding poses, contacts, and protein conformational changes.

Experiment Steps

1. Define the exact PCSK9 binding site and the success metric you will rank by.
2. Choose one compound set from IMPPAT, TCMSP, or both, then plan how you will standardize structures before screening.
3. Build a pharmacophore rule set from known PCSK9 ligands or literature features, and decide which matches count as hits.
4. Set up a docking workflow that keeps protein preparation, pose selection, and rescoring consistent across every candidate.
5. Add a molecular dynamics check for your best hits, and decide which stability signals will count as real support.
6. Predefine controls, comparison groups, and statistics so you can tell whether your scaffold ranking beats random selection.

Common Pitfalls

• Mixing protonation states across database entries, which makes the same scaffold look better or worse for the wrong reason.
• Treating a single docking pose as proof of binding, which ignores false positives that only fit one snapshot.
• Skipping compound cleanup, which leaves salts, duplicates, and odd structures in the final hit list.
• Using one PCSK9 structure for every run, which hides how sensitive the ranking is to protein conformation.
• Reading MD stability from one short trajectory, which can turn a noisy pose into a fake lead.

What Makes This Competitive

A strong version of this project does more than list top docking scores. It compares two databases, tests whether pharmacophore filters improve enrichment, and checks if the same scaffolds survive across multiple protein conformations. If you add consensus scoring, strict false-positive filters, and a clear benchmark against known PCSK9 ligands, your project starts to look like real discovery work. That kind of design gives judges a reason to trust the ranking, not just the final list.

Project Variations

• Screen only alkaloid or terpene scaffolds from IMPPAT and compare their PCSK9 hit rate with the full database.
• Replace single-score docking with consensus scoring plus interaction fingerprints to see whether rankings get more stable.
• Compare native, apo, and alternate PCSK9 structures to test how target conformation changes candidate ordering.

Learn More

• PubMed: Search review articles on PCSK9, natural products, and molecular docking.
• RCSB PDB: Find PCSK9 structures, bound ligands, and structure annotations.
• PubChem: Check compound structures, synonyms, and bioassay records for candidate scaffolds.
• IMPPAT: Search the Indian Medicinal Plants, Phytochemistry and Therapeutics database for Ayurvedic compound lists.
• TCMSP: Search the Traditional Chinese Medicine Systems Pharmacology database for natural compounds and target links.