Loading Video...
NTHRYS
Arrow

Molecular Descriptors 0D–3D & Feature Engineering Training | RDKit, Physicochemical, Topological, Fingerprints

NTHRYS >> Services >> Academic Services >> Training Programs >> Bioinformatics Training >> Cheminformatics, ADMET & Computational Toxicology >> Molecular Descriptors 0D–3D & Feature Engineering Training | RDKit, Physicochemical, Topological, Fingerprints

Molecular Descriptors 0D–3D & Feature Engineering — Hands-on

Develop a strong working knowledge of molecular descriptors from 0D counts to 3D shape indices, and how they feed into QSAR, ADMET and virtual screening workflows. You will generate, inspect and engineer descriptors and fingerprints using RDKit and KNIME, and learn how to prepare a high quality feature matrix that is ready for cheminformatics modeling and analytics.

Molecular Descriptors 0D–3D & Feature Engineering
Help Desk · WhatsApp
Session 1
Fee: Rs 8800
Descriptor Landscape & Physicochemical Properties
  • Types of molecular descriptors and use cases
  • 0D / 1D / 2D / 3D global vs local QSAR / ADMET / VS
  • Basic physicochemical descriptors
  • molecular weight logP / logS HBA / HBD / TPSA
  • Toolchain and calculation basics
  • RDKit descriptor modules KNIME nodes simple descriptor sanity checks
Session 2
Fee: Rs 11800
Structural, Topological & 3D Descriptors
  • 2D structural and fragment descriptors
  • atom and bond counts ring systems substructure flags
  • Topological and connectivity indices
  • Wiener index Balaban index kappa / chi indices
  • 3D and conformation based descriptors
  • shape descriptors 3D polar surface conformer generation basics
Session 3
Fee: Rs 14800
Fingerprints & Feature Engineering
  • Circular and path based fingerprints
  • ECFP / FCFP Morgan fingerprints MACCS keys
  • Scaling, transforms and encoding strategies
  • standardization / normalization log transforms one hot and binning
  • Feature quality and reduction
  • variance and correlation filters PCA overview basic feature selection ideas
Session 4
Fee: Rs 18800
Mini Capstone: Descriptor Matrix for Modeling
  • From curated structures to descriptor table
  • Theory + Practical
  • Apply descriptor calculation and feature processing
  • physicochemical panel fingerprint generation scaling and basic filtering
  • Deliverables: descriptor matrix and reproducible workflow
  • feature matrix CSV RDKit or KNIME workflow environment or node list


PDF