Video summary

Analysis & Visualization | Bioinformatics & Molecular Docking Internship 2026 | Class 13 | VNIAS

Main summary

Key takeaways

Educational

Main ideas, concepts, and lessons

  • Purpose of the session: Learn how to analyze, visualize, and present molecular docking results in a way that becomes scientific evidence (i.e., “scientific fact”) through clear communication.
  • Why it matters: These skills are essential for:
    • writing research papers
    • preparing conference presentations
    • creating project reports
  • Roadmap of what’s covered:
    • 2D interaction diagrams (protein–ligand)
    • 3D visualization of docking complexes and interactions
    • a brief touch on MD simulation context and what it adds after docking
    • how to write and report findings scientifically (including the standard structure and workflow)
  • Flow from docking to publication-grade support:
    • Docking generates poses, but MD simulation helps validate/refine them by allowing dynamic behavior
    • Rescoring/free-energy methods (e.g., MMGBSA / MM-PBSA) further validate binding energetics and stability
    • RMSF/RMSD-based indicators help infer whether the ligand remains stable in the binding site
  • Scientific communication methodology (writing/reporting workflow):
    • target selection and justification
    • structure/model retrieval or prediction
    • pocket identification
    • ligand library collection + filtering strategy
    • protein/ligand preparation methods
    • docking workflow and results presentation
    • standard paper sections: Introduction → Methods → Results → Discussion → Conclusion
  • Practical tools taught for visualization and interaction analysis:
    • Protein Interaction Profiler (PIP) (referred to as “Protein Interaction Profiler” / “clip” in subtitles)
    • Discovery Studio Visualizer
    • LigPlot+ (also mentioned as an option)
    • RDKit and plotting tools (matplotlib, seaborn, ggplot2, etc.) for Python/R-based analysis
  • Interaction frequency analysis concept:
    • when multiple docked ligands exist, identify which residues interact most consistently
    • use average/frequency of residue interactions to inform pharmacophore modeling/features

A) Docking-to-evidence concept (validation steps)

1) MD simulation after docking

MD simulation is used to validate/refine docking poses by:

  • mimicking a dynamic biological environment
  • accounting for protein flexibility and solvation (water box), including:
    • ions
    • charges

The session states that for publication-quality support, MD durations on the order of ~100–200 ns are required (as described).

2) Rescoring / binding free-energy validation

After MD:

  • run rescoring using MMGBSA / MM-PBSA on trajectories to estimate binding energies

3) Interpret MD metrics and additional indicators

Use metrics and interaction evidence such as:

  • RMSF of active-site residues
    • higher RMSF in ligand-binding regions suggests less stable binding
  • binding confidence indicators including:
    • hydrogen bond quantity
    • ΔG / free-energy calculations from MM/PBSA-type results
    • assessing induced fit and conformational changes
    • observing structural water displacement (water role addressed as well)

B) Scientific reporting / writing workflow (paper structure)

Introduction

  • introduce the study and its motivation.

Methods (key required components)

  • Target selection
    • specify the target
    • justify why it was selected (with literature evidence or rationale)
  • Structure preparation of the target
    • if novel: predict 3D structure (e.g., AlphaFold or other tools referenced)
    • if literature-based: retrieve from sources such as:
      • example: UniProt ID
      • example: PDB ID
    • report structure quality details (e.g., resolution and relevant factors as described from the PDB page)
  • Pocket / binding-site identification
    • report co-crystallized ligand coordinates if present
    • confirm pockets using tools like CASTp
  • Ligand collection and filtering
    • describe ligand library source (e.g., ChEMBL)
    • filtering criteria may include:
      • potency filters such as IC50 / EC50 / Kd
      • ADMET / pharmacokinetic & pharmacodynamic criteria (including BBB permeation and solubility)
      • Lipinski’s Rule of Five (L/Ro5) related filtering (whether violations are allowed depends on the chosen criteria)
    • mention tools and parameters used
    • cite the original database/tool papers (example mentioned: PubChem)
  • Protein/ligand preparation
    • describe how protein and ligands were prepared (noted as handled earlier in the program)
  • Docking setup and configuration
    • explain prepared files such as PDBQT
    • describe configuration dimensions/parameters (as referenced)

Results / Discussion / Conclusion

  • present interaction evidence (2D + 3D) and interpretation
  • conclude with binding reliability and stability evidence

C) File conversion clarification: SDF → PDBQT (AutoDockTools)

Correct ligand conversion behavior

A clarification from a previous session:

  • Do not use the receptor -xr flag for ligand conversion (-xr is for an extendable receptor, not for ligand conversion)

Recommended usage for SDF → PDBQT

For ligand SDF → PDBQT:

  • don’t use -xr
  • use the -m flag to split multiple ligands in a single SDF into separate PDBQT files

Required options during conversion

  • add polar hydrogens using --p (speaker note: “double dash polar hydrogens”)

  • add Gasteiger charges using the appropriate option (speaker note: “add gasteizer charges”)

Output folder management (important)

  • run commands and outputs so that results go to a dedicated output folder
  • recommended approach:
    • generate ligand PDBQT files into a folder like “PDBQT ligands”
    • avoid leaving outputs mixed into the original “SDF ligands” folder

Context of the prior issue

  • a prior conversion produced rigid/incorrect PDBQT files in the wrong folder
  • that caused docking errors
  • this session’s workflow fixes the output-and-conversion process

D) Steps to generate 2D/3D interaction diagrams using Protein Interaction Profiler (PIP)

  1. Open the Protein Interaction Profiler web interface/link.
  2. Choose/upload the dock complex file
    • example referenced: “complex.pdb”
  3. Click Analyze.
  4. Wait for loading.

Outputs

  • 2D interaction diagram
  • option to save as image (PNG/SVG; PNG explicitly mentioned)
  • 3D view access
  • option to download/open a PyMOL session (e.g., .pml)

E) How to interpret PIP interaction diagram output

Color/legend reference (as described)

  • protein = blue
  • ligand = orange
  • additional legend items include:
    • charges
    • water
    • aromatic ring center
    • metal lines (described as separate items)
  • hydrophobic interactions: shown with dotted lines
  • hydrogen bonds: shown with solid/proper lines (line styles noted by the speaker)

Interaction table details

  • Hydrophobic interactions
    • residue number (example: chain A, residue 84)
    • distance values
    • ligand atom number and protein atom number
  • Hydrogen bonds
    • residue identity (speaker noted minor naming confusion from subtitles)
    • distance between donor/acceptor atoms
    • donor angle
    • donor/acceptor atom identifiers

Export

  • download results in .rst format (speaker referred to “R ST” / .rst)
  • open the PyMOL session to continue 3D exploration

F) Exploring interactions in PyMOL session (as described)

Within the PyMOL .pml session:

  • switch to interaction representation views
  • show labels for residues and interactions
  • display distances / interaction geometry
  • use visualization helpers such as:
    • deselect/select interacting residues
    • show/hide dots (contact points)
    • show cartoon, surface, mesh (mesh briefly mentioned)
    • center/zoom
    • export images as PNG
    • export molecules/structures as needed

G) Steps to visualize in Discovery Studio Visualizer

  1. Open the PDB complex in Discovery Studio Visualizer.
  2. Use display controls:
    • rotate and adjust visual style (including example background/display change)
    • show a ligand-only interaction view (receptor removed; ligand + interacting residues kept)
  3. Additional features noted:
    • show/hide receptor surface
    • explain surface properties (e.g., aromaticity, hydrogen bonds, charge, solvent accessibility)
    • show 2D diagram inside the tool (Discovery Studio “show 2D diagram” mentioned)
    • save visualizations as image files (PAG/PNG-style export mentioned)

Speakers / sources featured (identified in subtitles)

Speakers

  • Miss Aba Fatima (main instructor / presenter)

Other mentioned entities

  • International / VNIAS (Vinyas) Internship Program 2026 (program context/host)
  • Miss Deepa Fatima (instructor contact name used in Q&A; LinkedIn name referenced)
  • Tools / software referenced:
    • LigPlot+ / LigPlot
    • Protein Interaction Profiler (PIP)
    • Discovery Studio Visualizer
    • RDKit
    • PyMOL (pml session)
    • AutoDockTools (implied by SDF→PDBQT conversion steps)
    • AlphaFold, I-TASSER / other structure tools (mentioned in subtitles)
    • UniProt, PDB
    • CASTp
    • ChEMBL, PubChem
    • MMGBSA, MM-PBSA
    • MD simulation

Original video